Compare commits

..

12 Commits

92 changed files with 6962 additions and 9497 deletions

View File

@ -1,5 +1,2 @@
Joe Neeman
Philippe Rouquier
Gabriel Craciunescu
George Danchev
Jean-Francois Wauthy
Philippe Rouquier

View File

@ -1,7 +1,7 @@
Derek Foreman <derek@signalmarketing.com> and Ben Jansens <xor@orodu.net>
Copyright (C) 2002-2006 Derek Foreman and Ben Jansens
Mario Danic <mario.danic@gmail.com>, Thomas Schmitt <scdbackup@gmx.net>
Copyright (C) 2006-2007 Mario Danic, Thomas Schmitt
Copyright (C) 2006 Mario Danic, Thomas Schmitt
This program is free software; you can redistribute it and/or modify

View File

@ -1 +0,0 @@
nothing here now

234
INSTALL
View File

@ -1,234 +0,0 @@
Installation Instructions
*************************
Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2004, 2005,
2006 Free Software Foundation, Inc.
This file is free documentation; the Free Software Foundation gives
unlimited permission to copy, distribute and modify it.
Basic Installation
==================
Briefly, the shell commands `./configure; make; make install' should
configure, build, and install this package. The following
more-detailed instructions are generic; see the `README' file for
instructions specific to this package.
The `configure' shell script attempts to guess correct values for
various system-dependent variables used during compilation. It uses
those values to create a `Makefile' in each directory of the package.
It may also create one or more `.h' files containing system-dependent
definitions. Finally, it creates a shell script `config.status' that
you can run in the future to recreate the current configuration, and a
file `config.log' containing compiler output (useful mainly for
debugging `configure').
It can also use an optional file (typically called `config.cache'
and enabled with `--cache-file=config.cache' or simply `-C') that saves
the results of its tests to speed up reconfiguring. Caching is
disabled by default to prevent problems with accidental use of stale
cache files.
If you need to do unusual things to compile the package, please try
to figure out how `configure' could check whether to do them, and mail
diffs or instructions to the address given in the `README' so they can
be considered for the next release. If you are using the cache, and at
some point `config.cache' contains results you don't want to keep, you
may remove or edit it.
The file `configure.ac' (or `configure.in') is used to create
`configure' by a program called `autoconf'. You need `configure.ac' if
you want to change it or regenerate `configure' using a newer version
of `autoconf'.
The simplest way to compile this package is:
1. `cd' to the directory containing the package's source code and type
`./configure' to configure the package for your system.
Running `configure' might take a while. While running, it prints
some messages telling which features it is checking for.
2. Type `make' to compile the package.
3. Optionally, type `make check' to run any self-tests that come with
the package.
4. Type `make install' to install the programs and any data files and
documentation.
5. You can remove the program binaries and object files from the
source code directory by typing `make clean'. To also remove the
files that `configure' created (so you can compile the package for
a different kind of computer), type `make distclean'. There is
also a `make maintainer-clean' target, but that is intended mainly
for the package's developers. If you use it, you may have to get
all sorts of other programs in order to regenerate files that came
with the distribution.
Compilers and Options
=====================
Some systems require unusual options for compilation or linking that the
`configure' script does not know about. Run `./configure --help' for
details on some of the pertinent environment variables.
You can give `configure' initial values for configuration parameters
by setting variables in the command line or in the environment. Here
is an example:
./configure CC=c99 CFLAGS=-g LIBS=-lposix
*Note Defining Variables::, for more details.
Compiling For Multiple Architectures
====================================
You can compile the package for more than one kind of computer at the
same time, by placing the object files for each architecture in their
own directory. To do this, you can use GNU `make'. `cd' to the
directory where you want the object files and executables to go and run
the `configure' script. `configure' automatically checks for the
source code in the directory that `configure' is in and in `..'.
With a non-GNU `make', it is safer to compile the package for one
architecture at a time in the source code directory. After you have
installed the package for one architecture, use `make distclean' before
reconfiguring for another architecture.
Installation Names
==================
By default, `make install' installs the package's commands under
`/usr/local/bin', include files under `/usr/local/include', etc. You
can specify an installation prefix other than `/usr/local' by giving
`configure' the option `--prefix=PREFIX'.
You can specify separate installation prefixes for
architecture-specific files and architecture-independent files. If you
pass the option `--exec-prefix=PREFIX' to `configure', the package uses
PREFIX as the prefix for installing programs and libraries.
Documentation and other data files still use the regular prefix.
In addition, if you use an unusual directory layout you can give
options like `--bindir=DIR' to specify different values for particular
kinds of files. Run `configure --help' for a list of the directories
you can set and what kinds of files go in them.
If the package supports it, you can cause programs to be installed
with an extra prefix or suffix on their names by giving `configure' the
option `--program-prefix=PREFIX' or `--program-suffix=SUFFIX'.
Optional Features
=================
Some packages pay attention to `--enable-FEATURE' options to
`configure', where FEATURE indicates an optional part of the package.
They may also pay attention to `--with-PACKAGE' options, where PACKAGE
is something like `gnu-as' or `x' (for the X Window System). The
`README' should mention any `--enable-' and `--with-' options that the
package recognizes.
For packages that use the X Window System, `configure' can usually
find the X include and library files automatically, but if it doesn't,
you can use the `configure' options `--x-includes=DIR' and
`--x-libraries=DIR' to specify their locations.
Specifying the System Type
==========================
There may be some features `configure' cannot figure out automatically,
but needs to determine by the type of machine the package will run on.
Usually, assuming the package is built to be run on the _same_
architectures, `configure' can figure that out, but if it prints a
message saying it cannot guess the machine type, give it the
`--build=TYPE' option. TYPE can either be a short name for the system
type, such as `sun4', or a canonical name which has the form:
CPU-COMPANY-SYSTEM
where SYSTEM can have one of these forms:
OS KERNEL-OS
See the file `config.sub' for the possible values of each field. If
`config.sub' isn't included in this package, then this package doesn't
need to know the machine type.
If you are _building_ compiler tools for cross-compiling, you should
use the option `--target=TYPE' to select the type of system they will
produce code for.
If you want to _use_ a cross compiler, that generates code for a
platform different from the build platform, you should specify the
"host" platform (i.e., that on which the generated programs will
eventually be run) with `--host=TYPE'.
Sharing Defaults
================
If you want to set default values for `configure' scripts to share, you
can create a site shell script called `config.site' that gives default
values for variables like `CC', `cache_file', and `prefix'.
`configure' looks for `PREFIX/share/config.site' if it exists, then
`PREFIX/etc/config.site' if it exists. Or, you can set the
`CONFIG_SITE' environment variable to the location of the site script.
A warning: not all `configure' scripts look for a site script.
Defining Variables
==================
Variables not defined in a site shell script can be set in the
environment passed to `configure'. However, some packages may run
configure again during the build, and the customized values of these
variables may be lost. In order to avoid this problem, you should set
them in the `configure' command line, using `VAR=value'. For example:
./configure CC=/usr/local2/bin/gcc
causes the specified `gcc' to be used as the C compiler (unless it is
overridden in the site shell script).
Unfortunately, this technique does not work for `CONFIG_SHELL' due to
an Autoconf bug. Until the bug is fixed you can use this workaround:
CONFIG_SHELL=/bin/bash /bin/bash ./configure CONFIG_SHELL=/bin/bash
`configure' Invocation
======================
`configure' recognizes the following options to control how it operates.
`--help'
`-h'
Print a summary of the options to `configure', and exit.
`--version'
`-V'
Print the version of Autoconf used to generate the `configure'
script, and exit.
`--cache-file=FILE'
Enable the cache: use and save the results of the tests in FILE,
traditionally `config.cache'. FILE defaults to `/dev/null' to
disable caching.
`--config-cache'
`-C'
Alias for `--cache-file=config.cache'.
`--quiet'
`--silent'
`-q'
Do not print messages saying which checks are being made. To
suppress all normal output, redirect it to `/dev/null' (any error
messages will still be shown).
`--srcdir=DIR'
Look for the package's source code in directory DIR. Usually
`configure' can determine that directory automatically.
`configure' also accepts some other, not widely useful, options. Run
`configure --help' for more details.

View File

@ -1,7 +1,7 @@
pkgconfigdir=$(libdir)/pkgconfig
libincludedir=$(includedir)/libburn
lib_LTLIBRARIES = libburn/libburn.la
lib_LTLIBRARIES = libburn/libburn.la libisofs/libisofs.la
## ========================================================================= ##
@ -39,7 +39,6 @@ libburn_libburn_la_SOURCES = \
libburn/null.h \
libburn/options.c \
libburn/options.h \
libburn/os.h \
libburn/read.c \
libburn/read.h \
libburn/sbc.c \
@ -65,18 +64,44 @@ libburn_libburn_la_SOURCES = \
## libburn/sg-@ARCH@.c \
libisofs_libisofs_la_LDFLAGS = \
-version-info $(LT_CURRENT):$(LT_REVISION):$(LT_AGE)
libisofs_libisofs_la_SOURCES = \
libisofs/tree.h \
libisofs/tree.c \
libisofs/volume.h \
libisofs/volume.c \
libisofs/util.h \
libisofs/util.c \
libisofs/ecma119.c \
libisofs/ecma119.h \
libisofs/ecma119_tree.c \
libisofs/ecma119_tree.h \
libisofs/susp.h \
libisofs/susp.c \
libisofs/rockridge.h \
libisofs/rockridge.c \
libisofs/joliet.c \
libisofs/joliet.h \
libisofs/exclude.c \
libisofs/exclude.h \
libisofs/hash.h \
libisofs/hash.c
libinclude_HEADERS = \
libburn/libburn.h
libburn/libburn.h \
libisofs/libisofs.h
## ========================================================================= ##
## Build test applications
noinst_PROGRAMS = \
test/libburner \
test/telltoc \
test/dewav \
test/fake_au \
test/iso \
test/poll \
test/toc \
test/structest
bin_PROGRAMS = \
@ -85,9 +110,6 @@ bin_PROGRAMS = \
test_libburner_CPPFLAGS = -Ilibburn
test_libburner_LDADD = $(libburn_libburn_la_OBJECTS) $(THREAD_LIBS)
test_libburner_SOURCES = test/libburner.c
test_telltoc_CPPFLAGS = -Ilibburn
test_telltoc_LDADD = $(libburn_libburn_la_OBJECTS) $(THREAD_LIBS)
test_telltoc_SOURCES = test/telltoc.c
test_dewav_CPPFLAGS = -Ilibburn
test_dewav_LDADD = $(libburn_libburn_la_OBJECTS) $(THREAD_LIBS)
test_dewav_SOURCES = test/dewav.c
@ -97,15 +119,22 @@ test_fake_au_SOURCES = test/fake_au.c
test_poll_CPPFLAGS = -Ilibburn
test_poll_LDADD = $(libburn_libburn_la_OBJECTS) $(THREAD_LIBS)
test_poll_SOURCES = test/poll.c
test_toc_CPPFLAGS = -Ilibburn
test_toc_LDADD = $(libburn_libburn_la_OBJECTS) $(THREAD_LIBS)
test_toc_SOURCES = test/toc.c
test_structest_CPPFLAGS = -Ilibburn
test_structest_LDADD = $(libburn_libburn_la_OBJECTS) $(THREAD_LIBS)
test_structest_SOURCES = test/structest.c
test_iso_CPPFLAGS = -Ilibisofs
test_iso_LDADD = $(libisofs_libisofs_la_OBJECTS) $(THREAD_LIBS)
test_iso_SOURCES = test/iso.c
## cdrskin construction site - ts A60816
cdrskin_cdrskin_CPPFLAGS = -Ilibburn
cdrskin_cdrskin_CFLAGS = -DCdrskin_libburn_0_3_0
cdrskin_cdrskin_CFLAGS = -DCdrskin_libburn_0_2_3
cdrskin_cdrskin_LDADD = $(libburn_libburn_la_OBJECTS) $(THREAD_LIBS)
cdrskin_cdrskin_SOURCES = cdrskin/cdrskin.c cdrskin/cdrfifo.c cdrskin/cdrfifo.h cdrskin/cdrskin_timestamp.h
## cdrskin_cdrskin_SOURCES = cdrskin/cdrskin.c cdrskin/cdrfifo.c cdrskin/cdrfifo.h cdrskin/cleanup.c cdrskin/cleanup.h cdrskin/cdrskin_timestamp.h
##
## Open questions: how to compute $timestamp and express -DX="$timestamp"
##
@ -145,9 +174,13 @@ uninstall-local:
# Indent source files
indent_files = \
$(libisofs_libisofs_la_SOURCES) \
$(libburn_libburn_la_SOURCES) \
$(test_libburner_SOURCES) \
$(test_poll_SOURCES) \
$(test_structest_SOURCES)
$(test_toc_SOURCES) \
$(test_structest_SOURCES) \
$(test_iso_SOURCES)
indent: $(indent_files)
@ -163,13 +196,12 @@ indent: $(indent_files)
# Extra things
nodist_pkgconfig_DATA = \
libburn-1.pc
# http://www.nada.kth.se/cgi-bin/info?(automake.info)Man%20pages
man_MANS = cdrskin/cdrskin.1
libburn-1.pc \
libisofs-1.pc
EXTRA_DIST = \
libburn-1.pc.in \
libisofs-1.pc.in \
version.h.in \
doc/comments \
doc/doxygen.conf.in \
@ -180,19 +212,12 @@ EXTRA_DIST = \
cdrskin/README \
cdrskin/cdrecord_spy.sh \
cdrskin/compile_cdrskin.sh \
cdrskin/convert_man_to_html.sh \
cdrskin/changelog.txt \
cdrskin/cdrskin_eng.html \
cdrskin/wiki_plain.txt \
cdrskin/cleanup.h \
cdrskin/cleanup.c \
libburn/os-freebsd.h \
libburn/os-linux.h \
libburn/sg-freebsd.c \
libburn/sg-linux.c \
COPYING \
NEWS \
ChangeLog \
INSTALL \
$(man_MANS)
COPYING

1
NEWS
View File

@ -1 +0,0 @@
nothing here now

75
README
View File

@ -1,25 +1,25 @@
------------------------------------------------------------------------------
libburnia.pykix.org
libburn.pykix.org
------------------------------------------------------------------------------
This all is under GPL.
(See GPL reference, our clarification and commitment at the end of this text)
------------------------------------------------------------------------------
libburnia.pykix.org
libburn.pykix.org
By Mario Danic <mario.danic@gmail.com> and Thomas Schmitt <scdbackup@gmx.net>
Copyright (C) 2006-2007 Mario Danic, Thomas Schmitt
Copyright (C) 2006 Mario Danic, Thomas Schmitt
Still containing parts of
Libburn. By Derek Foreman <derek@signalmarketing.com> and
Ben Jansens <xor@orodu.net>
Copyright (C) 2002-2006 Derek Foreman and Ben Jansens
These parts are to be replaced by own code of above libburnia.pykix.org
copyright holders and then libburnia.pykix.org is to be their sole copyright.
These parts are to be replaced by own code of above libburn.pykix.org-copyright
holders and then libburn.pykix.org is to be their sole copyright.
This is done to achieve the right to issue the clarification and the
commitment as written at the end of this text.
The rights and merits of the Libburn-copyright holders Derek Foreman and
Ben Jansens will be duely respected.
This libburnia.pykix.org toplevel README (C) 2006-2007 Thomas Schmitt
This libburn.pykix.org toplevel README (C) 2006 Thomas Schmitt
------------------------------------------------------------------------------
Build and Installation
@ -27,14 +27,14 @@ This libburnia.pykix.org toplevel README (C) 2006-2007 Thomas Schmitt
Our build system is based on autotools. For preparing the build of a SVN
snapshot you will need autotools of at least version 1.7.
Check out from SVN by
svn co http://libburnia-svn.pykix.org/libburn/trunk libburn_pykix
go into directory libburn_pykix and apply autotools by
svn co http://libburn-svn.pykix.org/trunk libburn_pykix
and apply autotools by
./bootstrap
Alternatively you may unpack a release tarball for which you do not need
autotools installed.
To build a libburnia.pykix.org subproject it should be sufficient to go
To build libburn.pykix.org and its subprojects it should be sufficient to go
into its toplevel directory (here: "libburn_pykix") and execute
./configure
make
@ -42,36 +42,28 @@ into its toplevel directory (here: "libburn_pykix") and execute
To make the libraries accessible for running resp. developing applications
make install
The other half of the project, libisofs, is hosted in the libburnia SVN, too:
svn co http://libburnia-svn.pykix.org/libisofs/trunk libisofs_pykix
See README file there.
------------------------------------------------------------------------------
Overview of libburn.pykix.org
Overview of libburnia.pykix.org
libburnia.pykix.org is an open-source software project for reading, mastering
and writing optical discs. For now this means only CD-R, CD-RW, DVD-RAM,
DVD+RW, DVD-RW.
libburn.pykix.org is an open-source library for reading, mastering and writing
optical discs. For now this means only CD-R and CD-RW.
The project comprises of several more or less interdependent parts which
together strive to be a usable foundation for application development.
These are libraries, language bindings, and middleware binaries which emulate
classical (and valuable) Linux tools.
Our scope is currently Linux 2.4 and 2.6 only. For ports to other systems
we would need : login on a development machine resp. a live OS on CD or DVD,
advise from a system person about the equivalent of Linux sg or FreeBSD CAM,
volunteers for testing of realistic use cases.
Our scope is currently Linux 2.4 and 2.6 and we will have a hard time to widen
this for now, because of our history. The project could need advise from or
membership of skilled kernel people and people who know how to talk CD/DVD
drives into doing things.
We have a workable code base for burning CDs and overwriteable DVDs, though.
The burn API is quite comprehensively documented and can be used to build a
presentable application.
We have a functional binary which emulates parts of cdrecord in order to
prove that usability, and in order to allow you to explore libburnia's scope
We do have a workable code base for burning data CDs, though. The burn API is
quite comprehensively documented and can be used to build a presentable
application.
We do have a functional binary which emulates parts of cdrecord in order to
prove that usability, and in order to allow you to explore libburn's scope
by help of existing cdrecord frontends.
The project components (list subject to growth, hopefully):
@ -166,25 +158,6 @@ Project history as far as known to me:
This version of cdrskin is much more cdrecord compatible in repect
to drive addressing and audio features.
- 30th October 2006 release of cdrskin-0.2.4 .
- 13th November 2006 splitting releases of libburn+cdrskin from libisofs.
- 24th November 2006 release of libburn-0.2.6 and cdrskin-0.2.6 . cdrskin has
become suitable for unaware frontends as long as they perform only the core
of cdrecord use cases (including open-ended input streams, audio, and
multi-session).
- 28th November 2006 the umbrella project which encloses both, libisofs and
libburn, is now called libburnia. For the origin of this name, see
http://en.wikipedia.org/wiki/Liburnians .
- 16th January 2007 release of libburn-0.3.0 and cdrskin-0.3.0 . Now the scope
is widened to a first class of DVD media: overwriteable single layer types
DVD-RAM, DVD+RW, DVD-RW. This is not a cdrecord emulation but rather inspired
by dvd+rw-tools' "poor man" writing facility for this class of media.
Taking a bow towards Andy Polyakov.
------------------------------------------------------------------------------
@ -204,9 +177,9 @@ Project history as far as known to me:
------------------------------------------------------------------------------
Clarification in my name and in the name of Mario Danic, upcoming copyright
holders on toplevel of libburnia. To be fully in effect after the remaining
other copyrighted code has been replaced by ours and by copyright-free
contributions of our friends:
holders on toplevel of libburn. To be fully in effect after the remaining other
copyrighted code has been replaced by ours and by copyright-free contributions
of our friends:
------------------------------------------------------------------------------
We, the copyright holders, agree on the interpretation that

View File

@ -1,5 +1,5 @@
------------------------------------------------------------------------------
libburnia.pykix.org scdbackup.sourceforge.net/cdrskin_eng.html
libburn.pykix.org scdbackup.sourceforge.net/cdrskin_eng.html
------------------------------------------------------------------------------
Installation instructions at about line 60. First the legal stuff:
------------------------------------------------------------------------------
@ -7,31 +7,31 @@ This all is under GPL.
(See GPL reference, our clarification and commitment at the end of this text)
------------------------------------------------------------------------------
Based on and sub project of:
libburnia.pykix.org
libburn.pykix.org
By Mario Danic <mario.danic@gmail.com> and Thomas Schmitt <scdbackup@gmx.net>
Copyright (C) 2006-2007 Mario Danic, Thomas Schmitt
Copyright (C) 2006 Mario Danic, Thomas Schmitt
libburnia.pykix.org is inspired by and in other components still containing
libburn.pykix.org is inspired by and in other components still containing
parts of
Libburn. By Derek Foreman <derek@signalmarketing.com> and
Ben Jansens <xor@orodu.net>
Copyright (C) 2002-2006 Derek Foreman and Ben Jansens
See toplevel README for an overview of the current copyright situation in
libburnia.pykix.org.
libburn.pykix.org.
------------------------------------------------------------------------------
My thanks to the above authors (except myself, of course) for making the
following possible.
cdrskin. By Thomas Schmitt <scdbackup@gmx.net>
Integrated sub project of libburnia.pykix.org but also published via:
Integrated sub project of libburn.pykix.org but also published via:
http://scdbackup.sourceforge.net/cdrskin_eng.html
http://scdbackup.sourceforge.net/cdrskin-0.3.0.pl01.tar.gz
Copyright (C) 2006-2007 Thomas Schmitt
http://scdbackup.sourceforge.net/cdrskin-0.2.4.pl01.tar.gz
Copyright (C) 2006 Thomas Schmitt
------------------------------------------------------------------------------
On top of libburn there is implemented cdrskin 0.3.0, a limited cdrecord
On top of libburn there is implemented cdrskin 0.2.4, a limited cdrecord
compatibility wrapper which allows to use some libburn features from
the command line.
Interested users of cdrecord are invited to participate in the development
@ -59,16 +59,16 @@ systems, including 64 bit systems. (Further reports are welcome.)
Compilation, First Glimpse, Installation
Obtain cdrskin-0.3.0.pl01.tar.gz, take it to a directory of your choice and do:
Obtain cdrskin-0.2.4.pl01.tar.gz ,take it to a directory of your choice and do:
tar xzf cdrskin-0.3.0.pl01.tar.gz
cd cdrskin-0.3.0
tar xzf cdrskin-0.2.4.pl01.tar.gz
cd cdrskin-0.2.4
Or obtain a libburnia.pykix.org SVN snapshot,
Or obtain a libburn.pykix.org SVN snapshot,
go into the toplevel directory of the snapshot (e.g. cd libburn_pykix ),
and execute the autotools script ./bootstrap . Use autools version >= 1.7 .
Within that toplevel directory of either cdrskin-0.3.0 or libburn then execute:
Within that toplevel directory of either cdrskin-0.2.4 or libburn then execute:
./configure
make
@ -86,11 +86,10 @@ In order to surely get a standalone binary, execute
cdrskin/compile_cdrskin.sh
Version identification and help texts available afterwards:
Version identification an help texts available afterwards:
cdrskin/cdrskin -version
cdrskin/cdrskin --help
cdrskin/cdrskin -help
man cdrskin/cdrskin.1
Install (eventually as superuser) cdrskin to a directory where it can be found:
If cdrskin was already installed by a previous version, or by "make install"
@ -109,10 +108,6 @@ It will not collide with an installed version of libburn either.
But libpthread must be installed on the system and glibc has to match. (See
below for a way to create a statically linked binary.)
To install the man page, you may do: echo $MANPATH and choose one of the
listed directories to copy the man-page under its ./man1 directory. Like:
cp cdrskin/cdrskin.1 /usr/share/man/man1/cdrskin.1
Usage
@ -146,34 +141,36 @@ It is not checked for the necessary degree of hacker safety.
Get an overview of cdrecord style addresses of available devices
cdrskin -scanbus
cdrskin dev=ATA -scanbus
cdrskin --devices
Adresses reported with dev=ATA need prefix "ATA:". Address examples:
dev=0,1,0 dev=ATA:1,0,0 dev=/dev/sg1 dev=/dev/hdc
See also "Drive Addressing" below.
Note: Adresses reported with dev=ATA are to be used with prefix "ATA:". You may
well use device file addresses as reported with --devices. Examples:
dev=0,1,0 dev=/dev/sg1 dev=ATA:1,0,0 dev=/dev/hdc
See also "Drive Addressing".
Note: Address numbers have changed since cdrskin-0.2.2 in order to become
compatible with cdrecord numbers. To get the old number scheme, use
option --old_pseudo_scsi_adr . See also "Pseudo-SCSI Adresses".
Sorry for any inconvenience.
Obtain some info about the drive
cdrskin dev=0,1,0 -checkdrive
Obtain some info about the drive and the inserted media
cdrskin dev=0,1,0 -atip -v
cdrskin dev=0,1,0 -atip
Thoroughly blank a CD-RW
cdrskin -v dev=0,1,0 blank=all -eject
Format DVD-RW before first use with cdrskin
cdrskin -v dev=0,1,0 blank=format_overwrite
Blank CD-RW sufficiently for making it ready for overwrite
cdrskin -v dev=0,1,0 blank=fast -eject
Burn image file my_image.iso to CD-R, CD-RW, DVD+RW, DVD-RAM, DVD-RW
cdrskin -v dev=0,1,0 speed=12 fs=8m driveropts=burnfree padsize=300k \
Burn image file my_image.iso to CD
cdrskin -v dev=0,1,0 speed=12 fs=8m -sao driveropts=burnfree padsize=300k \
-eject my_image.iso
Burn a compressed afio archive to any of the above media types on-the-fly
find . | afio -oZ - | cdrskin -v dev=0,1,0 fs=32m speed=8 -tao \
driveropts=burnfree padsize=300k -
Burn a compressed afio archive to CD on-the-fly
find . | afio -oZ - | cdrskin -v dev=0,1,0 fs=32m speed=8 -sao \
driveropts=burnfree padsize=300k tsize=650m -
Burn 6 audio tracks from files with different formats to CD.
Anything except .wav or .au files has to be converted into raw format first.
@ -184,7 +181,7 @@ See below "Audio CD" for specifications.
madplay -o raw:track04.cd /path/to/track4.mp3
mppdec --raw-le /path/to/track5.mpc track05.cd
cdrskin -v dev=0,1,0 blank=fast -eject speed=48 -sao \
cdrskin dev=0,1,0 blank=fast -eject speed=48 -sao \
-audio -swab track0[1-5].cd /path/to/track6.wav
@ -201,7 +198,7 @@ or a device file address as listed by --devices with an accessible drive :
Set usage of cdrskin with appropriate options rather than cdrecord :
export SCDBACKUP_CDRECORD="cdrskin -v -v"
export SCDBACKUP_CDRECORD="cdrskin -v -v tao_to_sao_tsize=650m"
Run a backup :
@ -210,14 +207,12 @@ Run a backup :
Restrictions
The major restrictions are lifted now: audio, TAO, multi-session do work.
Many cdrecord options are still unsupported, though.
If you have use cases for them, please report your wishes and expectations.
DVD support is restricted to single layer overwriteable DVD (-RAM, +RW, -RW)
for now.
The convenient burn mode TAO is not available with libburn yet.
Therefore it has to be defaulted to mode SAO which needs to know the track
size in advance. non-cdrecord option tao_to_sao_tsize=650m causes each CD
to get burned up to 650 MB regardless of the payload size.
No multi session yet ... Please report your wishes.
Inspiration and Standard
@ -236,10 +231,6 @@ Actually i, Thomas Schmitt, am a devoted user of cdrecord via my project
scdbackup which still runs a bit better with cdrecord than with cdrskin. TAO.
I have the hope that Joerg feels more flattered than annoyed by cdrskin.
Many thanks to Andy Polyakov for his dvd+rw-tools
http://fy.chalmers.se/~appro/linux/DVD+RW/tools
which provide me with examples and pointers into MMC specs for DVD writing.
Drive Addressing
@ -333,9 +324,8 @@ will be the right one to replace "1,0,0" in above example.
If not --no_rc is the first argument then cdrskin attempts on startup to read
arguments from the following three files:
/etc/default/cdrskin
/etc/defaults/cdrskin
/etc/opt/cdrskin/rc
/etc/cdrskin/cdrskin.conf
$HOME/.cdrskinrc
The files are read in the sequence given above.
Each readable line is treated as one single argument. No extra blanks.
@ -348,7 +338,7 @@ dev=0,1,0
dev_translation=+1,0,0+0,1,0
# Some more options
fifo_start_at=0
--fifo_start_empty
fs=16m
@ -378,66 +368,6 @@ I myself am not into audio. So libburn-hackers@pykix.org might be the
best address for suggestions, requests and bug reports.
DVD+RW and DVD-RAM
DVD+RW and DVD-RAM media get treated as blank media regardless wether they
hold data or not. Options -audio and -multi are not allowed. Only one track
is allowed. -toc does not return information about the media content.
Speed is counted in DVD units (i.e. 1x = 1,385,000 bytes/second). Currently
there is no difference between -sao and -tao. If ever, then -tao will be the
mode which preserves the current behavior.
For these media, -msinfo alone would not be enough to perform appending of an
ISO filesystem. The filesystem driver will need a hint to find the start of the
most recent session. For example put an ISO filesystem at address 1 GB:
mkisofs -C 0,524288 ... | \
cdrskin dev=/dev/sr0 -v fs=32m -eject speed=4 write_start_address=524288s -
The superuser may then do:
mount -t iso9660 -o ro,sbsector=524288 /dev/sr0 /mnt
Note: On my linux-2.4.21-215 mount works only with sbsector <= 337920 (660 MB).
To extend a filesystem already existing at address 0
mkisofs -C 0,524288 -M /dev/sr0 ... | cdrskin dev=/dev/sr0 ...
Record the number 524288 for usage as first number with -C at the next
extension:
mkisofs -C 524288,1000000 ... | cdrskin write_start_address=1000000s ...
Program growisofs can append to an ISO filesystem on DVD+RW by additionally
manipulating the first session. cdrskin does not want to get involved so deep
into the format of the burned data. Be advised to use growisofs for the
task of maintaining extendable ISO-Filesystems on DVD+RW.
DVD-RW
DVD-RW are usable if formatted to state "Restricted Overwrite". They then
behave much like DVD+RW. See above.
DVD-RW in state "Sequential" have first to be formatted by
cdrskin dev=... -v blank=format_overwrite
"Sequential" is the state of unused media and of media previously blanked
or written by cdrecord. dvd+rw-format -blank can also achieve this state.
(Command dvd+rw-format -force can achieve "Restricted Overwrite".)
Formatting or first use of freshly formatted DVD-RW can produce unusual noises
from the drive and last several minutes. Depending on mutual compatibility of
drive and media, formatting can yield unusable media. It seems that those die
too on blanking by cdrecord or dvd+rw-format. Perils of DVD-RW, i fear.
There are three formatting variants with cdrskin currently:
blank=format_overwrite uses "DVD-RW Quick" formatting (MMC-type 15h)
and writes a first session of 128 MB. This leads to media which are expandable
and random addressable by cdrskin.
blank=format_overwrite_quickest uses "DVD-RW Quick" formatting (type 15h) too,
but leaves the media in "intermediate" state. In the first session of writing
one may only write sequentially to such a DVD. After that, it gets random
addressable by cdrskin.
blank=format_overwrite_full uses preferrably "Full Format" (type 00h).
This formatting lasts as long as writing a full DVD. It includes writing of
lead-out which is said to be good for DVD ROM compatibility.
Special compilation variations
You may get a (super fat) statically linked binary by :

View File

@ -22,7 +22,7 @@ set -x
#
# libburn version used: http://libburn.pykix.org
# Downloaded by:
# $ svn co http://libburn-svn.pykix.org/libburn/trunk libburn_pykix
# $ svn co http://libburn-svn.pykix.org/trunk libburn_pykix
# packed up in a tarball just to save it from inadverted changes by
# $ tar czf libburn_svn.tgz libburn_pykix
original="./libburn_svn.tgz"
@ -33,10 +33,10 @@ original="./libburn_svn.tgz"
# The top level directory in that snapshot is named
intermediate="./libburn_pykix"
# My changes are in libburn-develop , mainly in ./cdrskin
# My changes are in libburn-0.2.3.ts.develop , mainly in ./cdrskin
changes="./libburn-develop"
skin_rev="0.3.0"
changes="./libburn-0.2.3.ts.develop"
skin_rev="0.2.4"
# The result directory and the name of the result tarballs
target="./cdrskin-${skin_rev}"
@ -51,9 +51,6 @@ compile_cmd="./cdrskin/compile_cdrskin.sh"
compile_static_opts="-static"
compile_result="cdrskin/cdrskin"
man_to_html_cmd="./cdrskin/convert_man_to_html.sh"
man_page_html="cdrskin/man_1_cdrskin.html"
bintarget_dynamic="cdrskin_${skin_rev}-x86-suse9_0"
bintarget_static="$bintarget_dynamic"-static
@ -122,8 +119,6 @@ do
rm "$cdrskin_target"/cdrskin_"$i"
fi
done
# Remove eventual SVN stuff from cdrskin directory
for i in .deps .dirstamp .libs
do
if test -e "$cdrskin_target"/"$i"
@ -132,23 +127,6 @@ do
fi
done
# Remove GIFs of cdrskin_eng.html
rm "$cdrskin_target"/doener_*.gif
# Remove automatically generated HTML man page
rm "$cdrskin_target"/man_1_cdrskin.html
# Remove all add_ts_changes_to_libburn besides this one
for i in "$cdrskin_target"/add_ts_changes_to_libburn*
do
if test $(basename "$0") = $(basename "$i")
then
dummy=dummy
else
rm $i
fi
done
# Remove unwanted SVN stuff (TODO: avoid downloading it)
for i in "$target"/.svn "$target"/*/.svn
do
@ -163,6 +141,12 @@ do
fi
done
## No more : Add own libburn-README in toplevel
# cp -a "$changes"/README "$target"
## No more : Add modified Makefile.am
# cp -a "$changes"/Makefile.am "$target"
# Make SVN state tarball for the libburn team
tar czf "$cdrskin_tarball_svn" "$target"
@ -192,20 +176,18 @@ done
# Pack it up to the new libburn+cdrskin-tarball
tar czf "$cdrskin_tarball" "$target"
# Produce a static and a dynamic binary, and a HTML man page
# Produce a static and a dynamic binary
(
cd "$compile_dir" || exit 1
./configure
make
"$compile_cmd" -do_strip
$compile_cmd -do_strip
cp "$compile_result" "../$bintarget_dynamic"
if test -n "$compile_static_opts"
then
"$compile_cmd" $compile_static_opts -do_strip
$compile_cmd $compile_static_opts -do_strip
cp "$compile_result" "../$bintarget_static"
fi
"$man_to_html_cmd"
mv "$man_page_html" ..
)
# Remove the build area
@ -218,5 +200,4 @@ rm -rf "$target"
ls -l "$cdrskin_tarball"
ls -l "$bintarget_dynamic"
ls -l "$bintarget_static"
ls -l $(basename "$man_page_html")

View File

@ -0,0 +1,193 @@
#!/bin/sh
set -x
# This script was used for producing cdrskin-0.2.4.pl01 from
# the svn snapshot made for cdrskin-0.2.4.pl00 and some changes
# in a patch directory.
#
# Script results are a source tarball and two binaries
# one dynamic and one static in respect to system libs.
# Both binaries are static in respect to libburn.
#
# The script is to be run in the directory above the toplevel
# directory of libburn resp. cdrskin development.
#
# libburn version used: http://libburn.pykix.org
# Downloaded by:
# $ svn co http://libburn-svn.pykix.org/trunk libburn_pykix
# packed up in a tarball just to save it from inadverted changes by
# $ tar czf libburn_svn.tgz libburn_pykix
# Then it went through cdrskin/add_ts_changes_to_libburn_0_2_4 where
# the following svn snapshot emerged.
original="./cdrskin-0.2.4.svn.tar.gz"
# The top level directory in that snapshot is named
intermediate="./cdrskin-0.2.4"
# My changes are in
changes="./cdrskin-0.2.4.patch01"
# Version parameters
skin_release="0.2.4"
patch_level=".pl01"
skin_rev="$skin_release""$patch_level"
# The result directory and the name of the result tarballs
target="./cdrskin-${skin_release}"
cdrskin_tarball="./cdrskin-${skin_rev}.tar.gz"
cdrskin_tarball_svn="./cdrskin-${skin_rev}.svn.tar.gz"
# Where to work
compile_dir="$target"
compile_cmd="./cdrskin/compile_cdrskin.sh"
compile_static_opts="-static"
compile_result="cdrskin/cdrskin"
bintarget_dynamic="cdrskin_${skin_rev}-x86-suse9_0"
bintarget_static="$bintarget_dynamic"-static
if test -d "$changes"
then
dummy=dummy
else
echo "$0 : FATAL : no directory $changes" >&2
exit 1
fi
for i in "$target" "$intermediate"
do
if test -e "$i"
then
echo "$0 : FATAL : already existing $i" >&2
exit 2
fi
done
if test -f "$original"
then
dummy=dummy
else
echo "$0 : FATAL : no file $original" >&2
exit 3
fi
# Unpack SVN snapshot.
tar xzf "$original"
# Rename the directory to the cdrskin name
if test "$intermediate" = "$target"
then
dummy=dummy
else
mv "$intermediate" "$target"
fi
# Copy the changes from the patch tree
#
cdrskin_dir="$changes"/cdrskin
libburn_dir="$changes"/libburn
test_dir="$changes"/test
cdrskin_target="$target"/cdrskin
libburn_target="$target"/libburn
test_target="$target"/test
# Create version timestamp
timestamp="$(date -u '+%Y.%m.%d.%H%M%S')"
echo "$timestamp"
echo '#define Cdrskin_timestamP "'"$timestamp"' (pl01)"' \
>"$cdrskin_dir"/cdrskin_timestamp.h
#
# Do the repair of patch01
#
# configure.ac used macro AM_CONFIG_HEADER() and thus disabled all
# other compile time configure macros. Among them were those which
# enabled 64 bit off_t in libburn. They were always set in cdrskin
# so that there was a 64<>32 bit mess-up.
# Negative consequences are to be expected at least with big-endian
# processors (i.e non-Intel compatible, MSB-first) on systems with
# 32 bit off_t default.
#
# bootstrap called a program named autoheader which is obsolete
# without the AM_CONFIG_HEADER macro.
#
cp -a "$changes"/bootstrap "$target"
cp -a "$changes"/configure.ac "$target"
# Unlike cdrskin, libburner is not depending on 64 bit off_t
# so the according code part has been removed. A similar remedy in
# cdrskin took place in version 0.2.5. Its consequences are wider
# than would be appropriate for this patch.
cp -a "$test_dir"/libburner.c "$test_target"
# Mention changed tarball name
cp -a "$cdrskin_dir"/README "$cdrskin_target"
# Fetch this script
cp -a "$cdrskin_dir"/add_ts_changes_to_libburn_0_2_4_patch01 "$cdrskin_target"
# Mark the patch
cp -a "$cdrskin_dir"/cdrskin_timestamp.h "$cdrskin_target"
#
# end of repair for patch01
#
# Make SVN state tarball for the libburn team
tar czf "$cdrskin_tarball_svn" "$target"
# Get over dependecy on autotools. Rely only on cc, make et. al.
# This is not the same as "make dist" but i can do it without
# having to evaluate the quality of said "make dist"
#
( cd "$target" ; ./bootstrap )
# Remove unwanted stuff after bootstrap
for i in "$target"/autom4te.cache
do
if echo "$i" | grep '\*' >/dev/null
then
dummy=dummy
else
if test -e "$i"
then
rm -rf "$i"
fi
fi
done
# Pack it up to the new libburn+cdrskin-tarball
tar czf "$cdrskin_tarball" "$target"
# Produce a static and a dynamic binary
(
cd "$compile_dir" || exit 1
./configure
make
$compile_cmd -do_strip
cp "$compile_result" "../$bintarget_dynamic"
if test -n "$compile_static_opts"
then
$compile_cmd $compile_static_opts -do_strip
cp "$compile_result" "../$bintarget_static"
fi
)
# Remove the build area
# Disable this for debugging the merge process
rm -rf "$target"
# Show the result
./"$bintarget_dynamic" -version
./"$bintarget_static" -version
ls -l "$cdrskin_tarball"
ls -l "$bintarget_dynamic"
ls -l "$bintarget_static"

View File

@ -48,9 +48,6 @@ struct CdrfifO {
int source_fd;
double in_counter;
double fd_in_counter;
double fd_in_limit;
char *buffer;
int buffer_size;
int buffer_is_full;
@ -75,27 +72,17 @@ struct CdrfifO {
double empty_counter;
double full_counter;
/* (sequential) fd chaining */
/* fds: 0=source, 1=dest */
int follow_up_fds[Cdrfifo_ffd_maX][2];
/* index of first byte in buffer which does not belong to predecessor fd */
int follow_up_eop[Cdrfifo_ffd_maX];
/* index of first byte in buffer which belongs to [this] fd pair */
int follow_up_sod[Cdrfifo_ffd_maX];
/* values for fd_in_limit */
double follow_up_in_limits[Cdrfifo_ffd_maX];
/* number of defined follow-ups */
int follow_up_fd_counter;
/* index of currently active (i.e. reading) follow-up */
int follow_up_fd_idx;
/* (simultaneous) peer chaining */
struct CdrfifO *next;
struct CdrfifO *prev;
@ -130,8 +117,6 @@ int Cdrfifo_new(struct CdrfifO **ff, int source_fd, int dest_fd,
buffer_size+= chunk_size-(buffer_size%chunk_size);
o->source_fd= source_fd;
o->in_counter= 0.0;
o->fd_in_counter= 0;
o->fd_in_limit= -1.0;
o->buffer= NULL;
o->buffer_is_full= 0;
o->buffer_size= buffer_size;
@ -155,7 +140,6 @@ int Cdrfifo_new(struct CdrfifO **ff, int source_fd, int dest_fd,
for(i= 0; i<Cdrfifo_ffd_maX; i++) {
o->follow_up_fds[i][0]= o->follow_up_fds[i][1]= -1;
o->follow_up_eop[i]= o->follow_up_sod[i]= -1;
o->follow_up_in_limits[i]= -1.0;
}
o->follow_up_fd_counter= 0;
o->follow_up_fd_idx= -1;
@ -240,26 +224,6 @@ int Cdrfifo_set_speed_limit(struct CdrfifO *o, double bytes_per_second,
}
/** Set a fixed size for input in order to cut off any unwanted tail
@param o The fifo object
@param idx index for fds attached via Cdrfifo_attach_follow_up_fds(),
first attached is 0, <0 directs limit to active fd limit
(i.e. first track is -1, second track is 0, third is 1, ...)
*/
int Cdrfifo_set_fd_in_limit(struct CdrfifO *o, double fd_in_limit, int idx,
int flag)
{
if(idx<0) {
o->fd_in_limit= fd_in_limit;
return(1);
}
if(idx >= o->follow_up_fd_counter)
return(0);
o->follow_up_in_limits[idx]= fd_in_limit;
return(1);
}
int Cdrfifo_set_fds(struct CdrfifO *o, int source_fd, int dest_fd, int flag)
{
o->source_fd= source_fd;
@ -280,7 +244,6 @@ int Cdrfifo_get_fds(struct CdrfifO *o, int *source_fd, int *dest_fd, int flag)
fifo buffer when its predecessors are exhausted. Reading will start as
soon as reading of the predecessor encounters EOF. Writing will start
as soon as all pending predecessor data are written.
@return index number of new item + 1, <=0 indicates error
*/
int Cdrfifo_attach_follow_up_fds(struct CdrfifO *o, int source_fd, int dest_fd,
int flag)
@ -290,7 +253,7 @@ int Cdrfifo_attach_follow_up_fds(struct CdrfifO *o, int source_fd, int dest_fd,
o->follow_up_fds[o->follow_up_fd_counter][0]= source_fd;
o->follow_up_fds[o->follow_up_fd_counter][1]= dest_fd;
o->follow_up_fd_counter++;
return(o->follow_up_fd_counter);
return(1);
}
@ -604,12 +567,7 @@ after_write:;
can_read= o->chunk_size;
if(o->write_idx<o->read_idx && o->write_idx+can_read > o->read_idx)
can_read= o->read_idx - o->write_idx;
if(o->fd_in_limit>=0.0)
if(can_read > o->fd_in_limit - o->fd_in_counter)
can_read= o->fd_in_limit - o->fd_in_counter;
ret= 0;
if(can_read>0)
ret= read(o->source_fd,o->buffer+o->write_idx,can_read);
ret= read(o->source_fd,o->buffer+o->write_idx,can_read);
if(ret==-1) {
/* >>> handle input error */;
@ -641,8 +599,6 @@ after_write:;
sod= 0;
o->follow_up_sod[idx]= sod;
o->write_idx= sod;
o->fd_in_counter= 0;
o->fd_in_limit= o->follow_up_in_limits[idx];
if(Cdrfifo_debuG || (flag&1))
fprintf(stderr,"\ncdrfio: new fifo source fd : %d\n",o->source_fd);
} else {
@ -652,7 +608,6 @@ after_write:;
did_work= 1;
o->put_counter++;
o->in_counter+= ret;
o->fd_in_counter+= ret;
o->write_idx+= ret;
if(o->write_idx>=o->buffer_size)
o->write_idx= 0;
@ -796,7 +751,7 @@ ex:;
/** Fill the fifo as far as possible without writing to destination fd */
int Cdrfifo_fill(struct CdrfifO *o, int size, int flag)
int Cdrfifo_fill(struct CdrfifO *o, int flag)
{
int ret,fill= 0,space,state;
@ -810,8 +765,6 @@ int Cdrfifo_fill(struct CdrfifO *o, int size, int flag)
} else if(state!=1)
break;
if(space<=0)
break;
if(size>=0 && fill>=size)
break;
ret= Cdrfifo_try_to_work(o,100000,NULL,NULL,2);
if(ret<0) {

View File

@ -15,7 +15,7 @@
/** The fifo buffer which will smoothen the data stream from data provider
to data consumer. Although this is not a mandatory lifesaver for modern
to data consumer. Although this is not a mandatory lifesavier for modern
burners any more, a fifo can speed up burning of data which is delivered
with varying bandwidths (e.g. compressed archives created on the fly
or mkisofs running at its speed limit.).
@ -64,16 +64,6 @@ int Cdrfifo_get_sizes(struct CdrfifO *o, int *chunk_size, int *buffer_size,
int Cdrfifo_set_speed_limit(struct CdrfifO *o, double bytes_per_second,
int flag);
/** Set a fixed size for input in order to cut off any unwanted tail
@param o The fifo object
@param idx index for fds attached via Cdrfifo_attach_follow_up_fds(),
first attached is 0, <0 directs limit to active fd limit
(i.e. first track is -1, second track is 0, third is 1, ...)
*/
int Cdrfifo_set_fd_in_limit(struct CdrfifO *o, double fd_in_limit, int idx,
int flag);
int Cdrfifo_set_fds(struct CdrfifO *o, int source_fd, int dest_fd, int flag);
int Cdrfifo_get_fds(struct CdrfifO *o, int *source_fd, int *dest_fd, int flag);
@ -82,7 +72,6 @@ int Cdrfifo_get_fds(struct CdrfifO *o, int *source_fd, int *dest_fd, int flag);
fifo buffer when its predecessors are exhausted. Reading will start as
soon as reading of the predecessor encounters EOF. Writing will start
as soon as all pending predecessor data are written.
@return index number of new item + 1, <=0 indicates error
*/
int Cdrfifo_attach_follow_up_fds(struct CdrfifO *o, int source_fd, int dest_fd,
int flag);
@ -145,11 +134,10 @@ int Cdrfifo_get_cdr_counters(struct CdrfifO *o,
int Cdrfifo_try_to_work(struct CdrfifO *o, int wait_usec,
char *reply_buffer, int *reply_count, int flag);
/** Fill the fifo as far as possible without writing to destination fd.
@param size if >=0 : end filling after the given number of bytes
/** Fill the fifo as far as possible without writing to destination fd
@return 1 on success, <=0 on failure
*/
int Cdrfifo_fill(struct CdrfifO *o, int size, int flag);
int Cdrfifo_fill(struct CdrfifO *o, int flag);
#endif /* Cdrfifo_headerfile_includeD */

View File

@ -1,703 +0,0 @@
.\" Hey, EMACS: -*- nroff -*-
.\" First parameter, NAME, should be all caps
.\" Second parameter, SECTION, should be 1-8, maybe w/ subsection
.\" other parameters are allowed: see man(7), man(1)
.TH CDRSKIN 1 "January 12, 2007"
.\" Please adjust this date whenever revising the manpage.
.\"
.\" Some roff macros, for reference:
.\" .nh disable hyphenation
.\" .hy enable hyphenation
.\" .ad l left justify
.\" .ad b justify to both left and right margins
.\" .nf disable filling
.\" .fi enable filling
.\" .br insert line break
.\" .sp <n> insert n+1 empty lines
.\" for manpage-specific macros, see man(7)
.SH NAME
cdrskin \- burns preformatted data to CD-R, CD-RW, DVD+/-RW, DVD-RAM
via libburn.
.SH SYNOPSIS
.B cdrskin
.RI [ options | track_source_addresses ]
.br
.SH DESCRIPTION
.PP
.\" TeX users may be more comfortable with the \fB<whatever>\fP and
.\" \fI<whatever>\fP escape sequences to invode bold face and italics,
.\" respectively.
.PP
\fBcdrskin\fP is a program that provides some of cdrecord's options
in a compatible way for CD media. With DVD it has its own ways for now.
You do not need to be superuser for its daily usage.
.PP
.B Overview of features:
.br
Blanking of CD-RW.
.br
Burning of data or audio tracks to CD,
.br
either in versatile Track at Once mode (TAO)
.br
or in Session at Once mode for seamless tracks.
.br
Multi session on CD (follow-up sessions in TAO only).
.br
Bus scan, burnfree, speed options, retrieving media info, padding, fifo.
.br
Burning of a single data track to DVD+RW, DVD-RW or DVD-RAM.
.br
See section EXAMPLES at the end of this text.
.PP
.B Track recording model:
.br
The input-output entities which get processed are called tracks.
A \fBtrack\fP stores a stream of bytes.
.br
Each track is initiated by one track source address argument, which may either
be "-" for standard input or the address of a readable file. If no write mode
is given explicitely then one will be chosen which matches the peculiarities
of track sources and the state of the output media.
.PP
More than one track can be burned by a single run of cdrskin.
In the terms of the MMC standard all tracks written by the same run constitute
a \fBsession\fP.
.br
CDs can be kept appendable so that further tracks can
be written to them in subsequent runs of cdrskin (see option -multi).
Info about the addresses of burned tracks is kept in a table of
content (TOC) on media and can be retrieved via cdrskin option -toc.
These informations are also used by the operating systems' CD-ROM read drivers.
.PP
In general there are two types of tracks: data and audio. They differ in
sector size, throughput and readability via the systems' CD-ROM drivers
resp. by music CD players.
.br
If not explicitely option -audio is given, then any track is burned as type
data, unless the track source is a file with suffix ".wav" or ".au" and has a
header part which identifies it as MS-WAVE resp. SUN Audio with suitable
parameters. Such files are burned as audio tracks by default.
.PP
While audio tracks just contain a given time span of acoustic vibrations,
data tracks may have an arbitray meaning. Nevertheless, ISO-9660 filesystems
are established as a format which can represent a tree of directories and
files on all major operating systems. Such filesystem images can be
produced by programs mkisofs or genisoimage. They can also be extended by
follow-up tracks if prepared properly. See the man pages of said programs.
cdrskin is able to fulfill the needs about their option -C.
.br
Another type of data track content are archive formats which originally
have been developed for magnetic tapes. Only formats which mark a detectable
end-of-archive in their data are suitable, though. Well tested are
the archivers afio and star. Not suitable seems GNU tar.
.PP
.B Recordable CD Media:
.br
CD-R can be initially written only once and eventually extended until they
get closed (or are spoiled because they are overly full). After that they are
read-only.
.br
CD-RW media can be blanked to make them re-usable for another
round of overwriting. Usually
.B blank=fast
is the appropriate option.
Blanking damages the previous content but does not
make it completely unreadable. It is no effective privacy precaution.
Multiple cycles of blanking and overwriting with random numbers might be.
.PP
.B Recordable DVD Media:
.br
Currently only types DVD+RW, DVD-RW and DVD-RAM can be burned via cdrskin.
.br
DVD+RW and DVD-RAM media get treated as blank media regardless wether they
hold data or not. They need no special initial formatting.
Options -audio and -multi are not allowed. Only one track is allowed.
-toc does not return information about the media content.
Speed is counted in DVD units (i.e. 1x = 1,385,000 bytes/second). Currently
there is no difference between -sao and -tao. If ever, then -tao will be the
mode which preserves the current behavior.
.br
DVD-RW need to be formatted to state "Restricted Overwrite". Then they behave
much like DVD+RW. This formatting can be done by option
.B blank=format_overwrite
It is necessary for unused media, for media written or blanked by cdrecord,
for media which have been written unformatted by growisofs or blanked by
dvd+rw-format -blank. If in doubt, just give it a try.
.PP
.B Drive preparation and addressing:
.br
The drives, either CD burners or DVD burners, are accessed via addresses which
are specific to libburn and the operating system. Those addresses get listed
by a run of \fBcdrskin --devices\fP.
.br
On Linux, they are device files which traditionally do not offer
w-permissions for normal users. Because libburn needs rw-permission,
it might be only the superuser who is able to get this list without further
precautions.
.br
It is consensus that \fBchmod a+rw /dev/sg0\fP or \fBchmod a+rw /dev/hdc\fP
is less security sensitive than chmod u+s /usr/bin/cdrskin. The risk for the
drive is somewhat higher but the overall system is much less at stake.
.br
.PP
If you only got one CD capable drive then you may leave out cdrskin option
\fBdev=\fP. Else you should use this option to address the drive you want.
.br
cdrskin option dev= not only accepts the listed addresses but also
traditional cdrecord SCSI addresses which on Linux consist of three
numbers: Bus,Target,Lun. There is also a related address family "ATA" which
accesses IDE drives not under control of Linux SCSI drivers:
ATA:Bus,Target,Lun.
.br
See option -scanbus for getting a list of cdrecord style addresses.
.br
Further are accepted on Linux: links to libburn-suitable device files,
device files which have the same major and minor device number,
and device files which have the same SCSI address parameters (e.g. /dev/sr0).
.br
.SH OPTIONS
.TP
.BI \-\-help
Show non-cdrecord compatible options.
.TP
.BI \-help
Show cdrecord compatible options.
.TP
.BI \-version
Print cdrskin id line, compatibility lure line, libburn version, cdrskin
version, version timestamp, build timestamp (if available), and then exit.
.PP
Alphabetical list of options which are intended to be compatible with
original cdrecord by Joerg Schilling:
.TP
.BI \-atip
Retrieve some info about media state. With CD-RW print "Is erasable".
With DVD media print "book type:" and a media type text.
.TP
.BI \-audio
Announces that the subsequent tracks are to be burned as audio.
The source is supposed to be uncompressed headerless PCM, 44100 Hz, 16 bit,
stereo. For little-endian byte order (which is usual on PCs) use option
-swab. Unless marked explicitely by option -data, input files with suffix
.wav are examined wether they have a header in MS-WAVE format confirming
those parameters and eventually raw audio data get extracted and burned as
audio track. Same is done for suffix .au and SUN Audio.
.TP
.BI blank= type
Blank a CD-RW or format a DVD+/-RW.
This is combinable with burning in the same run of cdrskin.
The type given with blank= selects the particular behavior:
.RS
.TP
help
Print this list of blanking types.
.TP
all
Blank an entire CD.
.TP
fast
Minimally blank an entire CD.
.TP
format_overwrite
Format a DVD-RW to "Restricted Overwrite". The user should bring some patience.
(Note: format_overwrite* are not original cdrecord options.)
.TP
format_overwrite_quickest
Like format_overwrite without creating a 128 MB trailblazer session.
Leads to "intermediate" state which only allows sequential write
beginning from address 0.
The "intermediate" state ends after the first session of writing data.
.TP
format_overwrite_full
For DVD-RW this is like format_overwrite but claims full media size
rather than just 128 MB.
Most traditional formatting is attempted. No data get written.
Much patience is required.
.br
For DVD+RW this is the only supported explicit formatting type. It provides
complete "de-icing" so no reader slips on unwritten data areas.
.RE
.TP
.BI \-checkdrive
Retrieve some info about the addressed drive.
Exits with non-zero value if the drive cannot be found and opened.
.TP
.BI \-dao
Alias for option -sao. Write disk in Session at Once mode.
.TP
.BI \-data
Subsequent tracks are data tracks. This option is default and only needed
to mark the end of the range of an eventual option -audio.
.TP
.BI dev= target
Set the address of the drive to use. Valid are at least the
addresses listed with option --devices,
X,Y,Z addresses listed with option -scanbus,
ATA:X,Y,Z addresses listed with options dev=ATA -scanbus,
and volatile libburn drive numbers (numbering starts at "0").
Other device file addresses which lead to the same drive might work too.
.br
If no dev= is given, volatile address "dev=0" is assumed. That is the first
drive found being available. Better avoid this ambiguity on systems with more
than one drive.
.br
The special target "help" lists hints about available addressing formats.
Be aware that deprecated option --old_pseudo_scsi_adr may change the meaning
of Bus,Target,Lun addresses.
.TP
.BI driveropts= opt
Set "driveropts=burnfree" to enable the drive's eventual protection mechanism
against temporary lack of source data (i.e. buffer underrun).
It is not an error to do this with a drive that has no such capabilities.
.TP
.BI \-dummy
Try to perform the drive operations without actually affecting the inserted
media. There is no guarantee that this will work with a particular drive
in a particular write mode. Blanking is prevented reliably, though.
.TP
.BI \-eject
Eject the disk after work is done.
.TP
.BI \-force
Assume that the user knows better in situations when cdrskin or libburn are
insecure about drive or media state. This includes the attempt to blank
media which are classified as unknown or unsuitable, and the attempt to use
write modes which libburn believes they are not supported by the drive.
.br
Another application is with blank=format_* to enforce re-formatting of media
which appear to be sufficiently formatted already.
.br
Use this only when in urgent need.
.TP
.BI fs= size
Set the fifo size to the given value. The value may have appended letters which
multiply the preceding number:
.br
"k" or "K" = 1024 , "m" or "M" = 1024k , "g" or "G" = 1024m , "s" or "S" = 2048
.br
Set size to 0 in order to disable the fifo (default is "4m").
.br
The fifo buffers an eventual temporary surplus of track source data in order to
provide the drive with a steady stream during times of temporary lack of track
source supply.
The larger the fifo, the longer periods of poor source supply can be
compensated.
But a large fifo needs substantial time to fill up if not curbed via
option fifo_start_at=size.
.TP
.BI gracetime= seconds
Set the grace time before starting to write. (Default is 0)
.TP
.BI msifile= path
Run option -msinfo and copy the result line into the file given by path.
Unlike -msinfo this option does not redirect all normal output away from
standard output. But it may be combined with -msinfo to achieve this.
.br
Note: msifile=path is actually an option of wodim and not of cdrecord.
.TP
.BI \-msinfo
Retrieve multi-session info for preparing a follow-up session by option -C
of programs mkisofs or genisoimage. Print result to standard output.
This option redirects to stderr all
message output besides its own result string, which consists of two numbers.
The result string shall be used as argument of option -C with said programs.
It gives the start address of the most recent session and the predicted
start address of the next session to be appended. The string is empty if
the most recent session was not written with option -multi.
.TP
.BI \-multi
This option keeps the CD appendable after the current session has been written.
Without it the disk gets closed and may not be written any more - unless it
is a CD-RW and gets blanked which causes loss of its content.
.br
The following sessions can only be written in -tao mode.
.br
In order to have all filesystem content accessible, the eventual ISO-9660
filesystem of a follow-up
session needs to be prepared in a special way by the filesystem formatter
program. mkisofs and genisoimage expect particular info about the situation
which can be retrieved by cdrskin option -msinfo.
.br
To retrieve an archive file which was written as follow-up session,
you may use option -toc to learn about the "lba" of the desired track number.
.TP
.BI \-nopad
Do not add trailing zeros to the data stream. Nevertheless, since there seems
to be no use for audio tracks with incomplete last sector, this option applies
only to data tracks. There it is default.
.TP
.BI \-pad
Add 30 kB of trailing zeros to each data track. (This is not sufficient to
avoid problems with various CD-ROM read drivers.)
.TP
.BI padsize= size
Add the given amount of trailing zeros to the next data track. This option
gets reset to padsize=0 after that next track is written. It may be set
again before the next track argument. About size specifiers, see option fs=.
.TP
.BI \-raw96r
Write disk in RAW/RAW96R mode. This mode allows to put more payload bytes
into a CD sector but obviously at the cost of error correction. It can only
be used for tracks of fixely predicted size. Some drives allow this mode but
then behave strange or even go bad for the next few attempts to burn a CD.
One should use it only if inavoidable.
.TP
.BI \-sao
Write disk in Session At Once mode. This mode is able to put several audio
tracks on CD without producing audible gaps between them. It can only
be used for tracks of fixely predicted size. This implies that track arguments
which depict stdin or named pipes need to be preceeded by option tsize= or
by option tao_to_sao_tsize=.
.TP
.BI \-scanbus
Scan the system for drives. On Linux the drives at /dev/s* and at /dev/hd*
are to be scanned by two separate runs. One without dev= for /dev/s* and
one with dev=ATA for /dev/hd* devices. (Option --drives lists all available
drives in a single run.)
.br
Drives which are busy or which offer no rw-permission to the user of cdrskin
are not listed. Busy drives get reported in form of warning messages.
.br
The useful fields in a result line are:
.br
Bus,Target,Lun Number) 'Vendor' 'Mode' 'Revision'
.TP
.BI speed= number
Set speed of drive. With data CD, 1x speed corresponds to a throughput of
150 kB/s. It is not an error to set a speed higher than is suitable for drive
and media. One should stay within a realistic speed range, though.
.TP
.BI \-swab
Announce that the raw audio data source of subsequent tracks is byte swapped
versus the expectations of cdrecord. This option is suitable for audio where
the least significant byte of a 16 bit word is first (little-endian, Intel).
Most raw audio data on PC systems are available in this byte order.
Less guesswork is needed if track sources are in format MS-WAVE in a file with
suffix ".wav".
.TP
.BI \-tao
Write disk in Track At Once (TAO) mode. This mode can be used with track
sources of unpredictable size, like standard input or named pipes. It is
also the only mode that can be used for writing to appendable CD which
already hold data.
.TP
.BI \-toc
Print the table of content (TOC) which describes the tracks recorded on CD.
The output contains all info from option -atip plus lines which begin with
"track:", the track number, the word "lba:" and a number which gives the
start address of the track. Addresses are counted in CD sectors which with
SAO or TAO data tracks hold 2048 bytes each.
.RS
.TP
Example. Retrieve an afio archive from track number 2:
.br
tracknumber=2
.br
lba=$(cdrskin dev=/dev/cdrom -toc 2>&1 | \\
.br
grep '^track:[ ]*[ 0-9][0-9]' | \\
.br
tail +"$tracknumber" | head -1 | \\
.br
awk '{ print $4}' )
.br
dd if=/dev/cdrom bs=2048 skip="$lba" | \\
.br
afio -t - | less
.RE
.TP
.BI tsize= size
Announces the exact size of the next track source. This is necessary with any
write mode other than -tao if the track source is not a regular disk file, but
e.g. "-" (standard input) or a named pipe.
About size specifiers, see option fs=.
.br
If the track source does not deliver the predicted amount of bytes, the
remainder of the track is padded with zeros. This is not considered an error.
If on the other hand the track source delivers more than the announced bytes
then the track on CD gets truncated to the predicted size and cdrskin exits
with non-zero value.
.TP
.BI \-v
Increment verbose level by one. Startlevel is 0 with only few messages.
Level 1 prints progress report with long running operations and also causes
some extra lines to be put out with info retrieval options.
Level 2 additionally reports about option settings derived from arguments or
startup files. Level 3 is for debugging and useful mainly in conjunction with
somebody who had a look into the program sourcecode.
.PP
Alphabetical list of options which are genuine to cdrskin and intended for
normal use:
.TP
.BI \--allow_setuid
Disable the loud warning about insecure discrepance between login user and
effective user which indicates application of chmod u+s to the program binary.
One should not do this chmod u+s , but it is an old cdrecord tradition.
.TP
.BI \--any_track
Allow source_addresses to begin with "-" (plus further characters) or to
contain a "=" character.
By default such arguments are seen as misspelled options. It is nevertheless
not possible to use one of the options listed with --list_ignored_options.
.TP
.BI \--demand_a_drive
Exit with a nonzero value if no drive can be found during a bus scan.
.TP
.BI \--devices
List the device file addresses of all accessible CD drives. In order to get
listed, a drive has to offer rw-permission for the cdrskin user and it may
not be busy. The superuser should be able to see all idle drives listed and
busy drives reported as "SORRY" messages.
.br
Each available drive gets listed by a line containing the following fields:
.br
Number dev='Devicefile' rw-Permissions : 'Vendor' 'Model'
.br
Number and Devicefile can both be used with option dev=, but number is
volatile (numbering changes if drives become busy).
.TP
.BI fifo_start_at= size
Do not wait for full fifo but start burning as soon as the given number
of bytes is read. This option may be helpful to bring the average throughput
near to the maximum throughput of a drive. A large fs= and a small
fifo_start_at= combine a quick burn start and a large savings buffer to
compensate for temporary lack of source data. At the beginning of burning,
the software protection against buffer underun is as weak as the size of
fifo_start_at= . So it is best if the drive offers hardware protection which
has to be enabled by driveropts=burnfree.
.TP
.BI \--list_ignored_options
List all ignored cdrecord options. The --options cannot be used as addresses
of track sources. No track source address may begin with a text equal to an
option which ends by "=". The list is ended by an empty line.
.TP
.BI \--no_rc
Only if used as first command line argument this option prevents reading and
interpretation of eventual startup files. See section FILES below.
.TP
.BI \--single_track
Accept only the last argument of the command line as track source address.
.TP
.BI write_start_address= byte_offset
Set the address on media where to start writing the track. With DVD+RW or
DVD-RAM byte_offset must be aligned to 2 KB blocks, but better is 32 kB.
With DVD-RW 32 kB alignment is mandatory.
.br
Other media are not suitable for this option yet.
.PP
Alphabetical list of options which are only intended for very special
situations and not for normal use:
.TP
.BI \--abort_handler
Establish default signal handling not to leave a drive in busy state
but rather to shut it down and to wait until it has ended the final operations.
This option is only needed for revoking eventual --ignore_signals or
--no_abort_handler.
.TP
.BI dev_translation= <sep><from><sep><to>
Set drive address alias. This was necessary before cdrskin-0.2.4 to manually
translate cdrecord addresses into cdrskin addresses.
.br
<sep> is a single character which may not occur in the address string
<from>. <from> is an address as expected to be given by the user via option
dev=. <to> is the address to be used instead whenever <from> is given.
More than one translation instruction can be given in one cdrskin run.
.br
E.g.: dev_translation=+ATA:1,0,0+/dev/sg1 dev_translation=+ATA:1,1,0+/dev/sg2
.TP
.BI \--drive_abort_on_busy
Linux specific: Abort process if a busy drive is encountered.
.TP
.BI \--drive_blocking
Linux specific: Try to wait for a busy drive to become free.
This is not guaranteed to work with all drivers. Some need nonblocking i/o.
.TP
.BI \--drive_not_exclusive
Linux specific: Do not ask the operating system to prevent opening busy drives.
Wether this leads to senseful behavior depends on operating system and kernel.
.TP
.BI \--drive_scsi_exclusive
Linux specific:
Try to exclusively reserve device files /dev/srN, /dev/scdM, /dev/stK of drive.
this would be helpful to protect against collisions with program growisofs.
Regrettably on Linux kernel 2.4 with ide-scsi emulation this seems not to
work. Wether it becomes helpful with new Linux systems has to be evaluated.
.TP
.BI \--fifo_disable
Disable fifo despite any fs=.
.TP
.BI \--fifo_per_track
Use a separate fifo for each track.
.TP
.BI grab_drive_and_wait= seconds
Open the addressed drive, wait the given number of seconds, release the drive,
and do normal work as indicated by the other options used. This option helps
to explore the program behavior when faced with busy drives. Just start a
second cdrskin with option --devices while grab_drive_and_wait= is still
active.
.TP
.BI \--ignore_signals
Try to ignore any signals rather than to abort the program. This is not a
very good idea. You might end up waiting a very long time for cdrskin
to finish.
.TP
.BI \--no_abort_handler
On signals exit even if the drive is in busy state. This is not a very good
idea. You might end up with a stuck drive that refuses to hand out the media.
.TP
.BI \--no_blank_appendable
Refuse to blank appendable CD-RW. This is a feature that was once builtin with
libburn. No information available for what use case it was needed.
.TP
.BI \--no_convert_fs_adr
Do only literal translations of dev=. This prevents cdrskin from test-opening
device files in order to find one that matches the given dev= specifier.
.br
Partly Linux specific:
Such opening is needed for Bus,Target,Lun addresses unless option
--old_pseudo_scsi_adr is given. It is also needed to resolve device file
addresses which are not listed with cdrskin --devices but nevertheless point
to a usable drive. (Like /dev/sr0 using the same SCSI address as /dev/sg0.)
.TP
.BI \--old_pseudo_scsi_adr
Linux specific:
Use and report literal Bus,Target,Lun addresses rather than real SCSI and
pseudo ATA addresses. This method is outdated and was never compatible with
original cdrecord.
.TP
.BI tao_to_sao_tsize= size
Set an exact fixed size for the next track to be in effect only if the track
source cannot deliver a size prediction and no tsize= was specified.
This is the fallback from bad old times when cdrskin was unable to burn
in mode -tao.
.br
.SH EXAMPLES
.SS
.B Get an overview of drives:
.br
cdrskin -scanbus
.br
cdrskin dev=ATA -scanbus
.br
cdrskin --devices
.SS
.B Get info about a particular drive or loaded media:
.br
cdrskin dev=0,1,0 -checkdrive
.br
cdrskin dev=ATA:1,0,0 -atip
.br
cdrskin dev=/dev/hdc -toc
.SS
.B Make used CD-RW writable again:
.br
cdrskin -v dev=/dev/sg1 blank=all -eject
.br
cdrskin -v dev=/dev/dvd blank=fast -eject
.SS
.B Format DVD-RW before first use with cdrskin:
.br
cdrskin -v dev=/dev/sr0 blank=format_overwrite
.SS
.B Write ISO-9660 filesystem image:
.br
cdrskin -v dev=/dev/hdc speed=12 fs=8m \\
.br
driveropts=burnfree -sao -eject \\
.br
padsize=300k my_image.iso
.SS
.B Write compressed afio archive on-the-fly:
.br
find . | afio -oZ - | \\
.br
cdrskin -v dev=0,1,0 fs=32m speed=8 driveropts=burnfree \\
.br
padsize=300k -tao -
.SS
.B Write several sessions to the same CD:
.br
cdrskin dev=/dev/hdc padsize=300k -multi 1.iso
.br
cdrskin dev=/dev/hdc padsize=300k -multi -tao 2.afio
.br
cdrskin dev=/dev/hdc padsize=300k -multi -tao 3.afio
.br
cdrskin dev=/dev/hdc padsize=300k -tao 4.afio
.SS
.B Get CD multi-session info for option -C of program mkisofs:
.br
c_values=$(cdrskin dev=/dev/sr0 -msinfo 2>/dev/null)
.br
mkisofs ... -C "$c_values" ...
.SS
.B Write audio tracks to CD:
.br
cdrskin -v dev=ATA:1,0,0 speed=48 \\
.br
driveropts=burnfree -sao \\
.br
track1.wav track2.au -audio -swab track3.raw
.br
.SH FILES
If not --no_rc is given as the first argument then cdrskin attempts on
startup to read the arguments from the following files:
.PP
.br
.B /etc/default/cdrskin
.br
.B /etc/opt/cdrskin/rc
.br
.B /etc/cdrskin/cdrskin.conf
.br
.B $HOME/.cdrskinrc
.br
.PP
The files are read in the sequence given above, but none of them is
required for cdrskin to function properly. Each readable line is treated
as one single argument. No extra blanks.
A first character '#' marks a comment, empty lines are ignored.
.SS
.B Example content of a startup file:
.br
# This is the default device
.br
dev=0,1,0
.br
# To accomodate to remnant cdrskin-0.2.2 addresses
.br
dev_translation=+1,0,0+0,1,0
.br
# Some more options
.br
fifo_start_at=0
.br
fs=16m
.br
.SH SEE ALSO
.TP
Formatting track sources for cdrskin:
.br
.BR mkisofs (8),
.BR genisoimage (8),
.BR afio (1),
.BR star (1)
.br
.TP
Other CD/DVD burn programs:
.br
.BR cdrecord (1),
.BR wodim (1)
.br
.TP
For DVD burning:
.br
.BR growisofs (1)
.br
.SH AUTHOR
cdrskin was written by Thomas Schmitt <scdbackup@gmx.net>.
.PP
This manual page was written by George Danchev <danchev@spnet.net> and
Thomas Schmitt, for the Debian project and for all others.

File diff suppressed because it is too large Load Diff

View File

@ -2,7 +2,7 @@
<HEAD>
<META NAME="description" CONTENT="cdrskin, a limited cdrecord compatibility wrapper for libburn">
<META NAME="keywords" CONTENT="cdrskin, libburn, libburnia, burn, CD, linux, CDR, CD-R, CDRW, CD-RW, cdrecord, compatible, scdbackup, burning">
<META NAME="keywords" CONTENT="cdrskin, libburn, burn, CD, linux, CDR, CD-R, CDRW, CD-RW, cdrecord, compatible, scdbackup, burning">
<META NAME="robots" CONTENT="follow">
<TITLE>cdrskin homepage english</TITLE>
</HEAD>
@ -11,12 +11,7 @@
<FONT SIZE=+1>
<CENTER>
<A HREF="http://en.wikipedia.org/wiki/D%C3%B6ner_kebab">
<IMG SRC="doener_150x200_tr_octx.gif" BORDER=0
ALT="cdrskin logo: Doener mit Scharf">
</A>
<P><H2> Homepage of </H2>
<H1> cdrskin </H1>
<P><H2>Homepage of</H2><H1><BR>cdrskin</H1><BR>
<!-- <FONT SIZE=+0><A HREF="cdrskin_ger.html">deutsch (german)</A></FONT> -->
<H2>Limited cdrecord compatibility wrapper for libburn</H2>
@ -25,7 +20,7 @@
<P>
<H2>Purpose:</H2>
<UL>
<LI>Burns preformatted data to CD-R, CD-RW, DVD-RAM, DVD+RW, DVD-RW</LI>
<LI>Burns preformatted data to CD-R or CD-RW</LI>
</UL>
</P>
<P>
@ -35,8 +30,8 @@
<P>
<H2>Hardware requirements:</H2>
A CD recorder suitable for
<A HREF="http://libburnia.pykix.org">libburnia.pykix.org</A>
(SCSI or IDE/ATAPI writers compliant to mmc-3 standard).
<A HREF="http://libburn.pykix.org">libburn.pykix.org</A>
(SCSI or IDE/ATAPI writers compliant to mmc standard).
<BR>
</P>
@ -56,8 +51,8 @@ A CD recorder suitable for
GPL software included:<BR>
</H2>
<DL>
<DT>libburn-0.3.0</DT>
<DD>(by Derek Foreman, Ben Jansens, and team of libburnia.pykix.org)</DD>
<DT>libburn-0.2.3 stabilized SVN snapshot</DT>
<DD>(by Derek Foreman, Ben Jansens, and team of libburn.pykix.org)</DD>
<DD>transfers data to CD</DD>
</DL>
</P>
@ -88,8 +83,6 @@ and for data CD projects of <A HREF="http://www.k3b.org">K3b</A>
(see <A HREF="#examples">examples</A>).
Suitability for audio CD frontends has been improved much and is now being
evaluated.<BR>
DVD are written in a pseudo -tao mode which is very different from the
write mode used by cdrecord(-ProDVD).<BR>
Further enhancements depend on people who can describe and discuss their
wishes as well as on the development of libburn.</DT>
<BR><BR>
@ -104,26 +97,21 @@ wishes as well as on the development of libburn.</DT>
<DT>Make used CD-RW writable again:</DT>
<DD>$ cdrskin -v dev=/dev/sg1 blank=all -eject</DD>
<DD>$ cdrskin -v dev=/dev/dvd blank=fast -eject</DD>
<DT>Format DVD-RW before first use with cdrskin</DT>
<DD>$ cdrskin -v dev=0,1,0 blank=format_overwrite -eject<DD>
<DT>Write ISO-9660 filesystem image:</DT>
<DD>$ cdrskin -v dev=/dev/hdc speed=12 fs=8m driveropts=burnfree -eject padsize=300k my_image.iso</DD>
<DT>Write compressed afio archive on-the-fly :</DT>
<DD>$ find . | afio -oZ - | cdrskin -v dev=0,1,0 fs=32m speed=8 driveropts=burnfree padsize=300k -tao -</DD>
<DD>$ cdrskin -v dev=/dev/hdc speed=12 fs=8m driveropts=burnfree -sao -eject padsize=300k my_image.iso</DD>
<DT>Write compressed afio archive on-the-fly:</DT>
<DD>$ find . | afio -oZ - | cdrskin -v dev=0,1,0 fs=32m speed=8 driveropts=burnfree padsize=300k -sao tsize=650m -</DD>
<DT>Write audio tracks:</DT>
<DD>$ cdrskin -v dev=ATA:1,0,0 speed=48 driveropts=burnfree -sao track1.wav track2.au -audio -swab track3.raw
<DD>
<BR>
<DT>Get overview of the cdrecord compatible options:</DT>
<DD><A HREF="cdrskin_help">$ cdrskin -help</A></DD>
<DT>Get overview of the non-cdrecord options:</DT>
<DD><A HREF="cdrskin__help">$ cdrskin --help</A></DD>
<DT>Read the detailed manual page:</DT>
<DD><A HREF="man_1_cdrskin.html">$ man cdrskin</A></DD>
<DT>Read about the standard for which cdrskin is striving:</DT>
<DD><A HREF="http://cdrecord.berlios.de/old/private/man/cdrecord-2.0.html">
$ man cdrecord</A></DD>
<DD><B>Do not bother Joerg Schilling with any cdrskin problems.</B>
<DT><A HREF="cdrskin_help">cdrskin -help</A></DT>
<DD>reports the cdrecord compatible options</DD>
<DT><A HREF="cdrskin__help">cdrskin --help</A></DT>
<DD>reports the non-cdrecord options</DD>
<DT><A HREF="http://cdrecord.berlios.de/old/private/man/cdrecord-2.0.html">man cdrecord</A></DT>
<DD>documents the standard for which cdrskin is striving.
<B>Do not bother Joerg Schilling with any cdrskin problems.</B>
(Be cursed if you install cdrskin as "cdrecord" without clearly forwarding
this "don't bother Joerg" demand.)
</DD>
@ -135,7 +123,11 @@ $ man cdrecord</A></DD>
<UL>
<DT></DT>
<LI>
Appending sessions to unclosed media is restricted to write mode TAO.
Burns only a single closed session. No -multi option yet.
</LI>
<LI>
No TAO mode and therefore no writing on-the-fly without a predefined
source size.
</LI>
<LI>
cdrskin -scanbus or --devices hangs for quite a while if there is
@ -157,8 +149,8 @@ rw-permissions and retry the bus scan as non-superuser.
<P>
<DL>
<DT>Download as source code (see README):</DT>
<DD><A HREF="cdrskin-0.3.0.pl01.tar.gz">cdrskin-0.3.0.pl01.tar.gz</A>
(540 KB).
<DD><A HREF="cdrskin-0.2.4.tar.gz">cdrskin-0.2.4.tar.gz</A>
(510 KB).
</DD>
<DD>
The "stable" cdrskin tarballs are source code identical with "stable"
@ -168,14 +160,14 @@ cdrskin is part of libburn - full libburn is provided with cdrskin releases.
</DD>
<DD>&nbsp;</DD>
<DT>Download as single x86 binaries (untar and move to /usr/bin/cdrskin):</DT>
<DD><A HREF="cdrskin_0.3.0.pl01-x86-suse9_0.tar.gz">
cdrskin_0.3.0.pl01-x86-suse9_0.tar.gz</A>, (75 KB),
<DD><A HREF="cdrskin_0.2.4-x86-suse9_0.tar.gz">
cdrskin_0.2.4-x86-suse9_0.tar.gz</A>, (60 KB),
<DL>
<DD>runs on SuSE 9.0 (2.4.21) , RIP-14.4 (2.6.14) ,
Gentoo (2.6.15 x86_64 Athlon).</DD>
</DL>
<DD><A HREF="cdrskin_0.3.0.pl01-x86-suse9_0-static.tar.gz">
cdrskin_0.3.0.pl01-x86-suse9_0-static.tar.gz</A>, (275 KB), -static compiled,
<DD><A HREF="cdrskin_0.2.4-x86-suse9_0-static.tar.gz">
cdrskin_0.2.4-x86-suse9_0-static.tar.gz</A>, (260 KB), -static compiled,
<DL>
<DD>runs on SuSE 7.2 (2.4.4), and on the systems above.</DD>
</DL>
@ -185,13 +177,10 @@ cdrskin_0.3.0.pl01-x86-suse9_0-static.tar.gz</A>, (275 KB), -static compiled,
<DD><A HREF="README_cdrskin">README</A> a short introduction</DD>
<DD><A HREF="cdrskin__help">cdrskin --help</A> non-cdrecord options</DD>
<DD><A HREF="cdrskin_help">cdrskin -help</A> cdrecord compatible options</DD>
<DD><A HREF="man_1_cdrskin.html">man cdrskin</A> the manual page</DD>
<DD>&nbsp;</DD>
</DL>
<DL><DT>Contact:</DT>
<DD>Thomas Schmitt, <A HREF="mailto:scdbackup@gmx.net">scdbackup@gmx.net</A></DD>
<DD>libburn development mailing list,
<A HREF="mailto:libburn-hackers@pykix.org">libburn-hackers@pykix.org</A></DD>
</DL>
<DL><DT>License:</DT>
<DD><A HREF="COPYING_cdrskin">GPL</A>, an <A HREF="http://www.opensource.org/">Open Source</A> approved license</DD>
@ -202,45 +191,44 @@ cdrskin_0.3.0.pl01-x86-suse9_0-static.tar.gz</A>, (275 KB), -static compiled,
<HR>
<P>
Enhancements towards previous stable version cdrskin-0.2.6:
Enhancements towards previous stable version cdrskin-0.2.2:
<UL>
<LI>Improved recognition of unsuitable media types</LI>
<LI>Ban of chmod u+s is replaced by a loud warning</LI>
<LI>Detailed man page</LI>
<LI>Burning to DVD+RW and DVD-RAM as non-multi, non-appending,
single-track session</LI>
<LI>Formatting and then burning to DVD-RW like to DVD+RW</LI>
<LI>Emulation of new wodim option msifile=path</LI>
<LI>cdrecord compatibility with drive addresses of form [ATA:]Bus,Target,Lun.
<BR>(use option --old_pseudo_scsi_adr to get back the incompatible
Bus,Traget,Lun addressing of version 0.2.2)
</LI>
<LI>Drives adressable via links and device siblings (/dev/cdrom , /dev/scd0).
</LI>
<LI>Automatic -audio extraction with .wav files and .au files.
</LI>
<LI>Bug fix about failure to eject.</LI>
<LI>Comments and empty lines allowed in startup files.</LI>
<LI>Options -scanbus and --devices print SORRY messages about busy drives.
</LI>
<LI>Drive buffer fill indicator reports realistic percentage numbers.</LI>
<LI>Option -toc is supported, drive firmware revision gets displayed.</LI>
</UL>
</P>
<P>
Bug fix towards previous patch level 0:
<UL>
<LI>Tracks >= 2 GB were only possible via a pipe to stdin but not
directly from a disk file</LI>
</UL>
</P>
<HR>
<P>
<DL>
<DT><H3>Development snapshot, version 0.3.1 :</H3></DT>
<DD>Enhancements towards stable version 0.3.0:
<UL>
<LI>-none yet-</LI>
</UL>
<DT>Development snapshot, version 0.2.5 :</DT>
<DD>&nbsp;</DD>
<DD>Enhancements towards stable version 0.2.4:
<UL>
<LI>-none for now-</LI>
</UL>
</DD>
<DD>&nbsp;</DD>
<DD><A HREF="README_cdrskin_devel">README 0.3.1</A>
<DD><A HREF="cdrskin__help_devel">cdrskin_0.3.1 --help</A></DD>
<DD><A HREF="cdrskin_help_devel">cdrskin_0.3.1 -help</A></DD>
<DD><A HREF="man_1_cdrskin_devel.html">man cdrskin (as of 0.3.1)</A></DD>
<DD><A HREF="README_cdrskin_devel">README 0.2.5</A>
<DD><A HREF="cdrskin__help_devel">cdrskin_0.2.5 --help</A></DD>
<DD><A HREF="cdrskin_help_devel">cdrskin_0.2.5 -help</A></DD>
<DD>&nbsp;</DD>
<DT>Maintainers of cdrskin unstable packages please use SVN of
<A HREF="http://libburnia.pykix.org"> libburnia.pykix.org</A></DT>
<DD>Download: <KBD><B>svn co http://libburnia-svn.pykix.org/libburn/trunk libburn_pykix</B>
<A HREF="http://libburn.pykix.org"> libburn.pykix.org</A></DT>
<DD>Download: <KBD><B>svn co http://libburn-svn.pykix.org/trunk libburn_pykix</B>
</KBD></DD>
<DD>Build: <KBD><B>cd libburn_pykix ; ./bootstrap ; ./configure ; make</B>
</KBD></DD>
@ -252,19 +240,19 @@ vanilla tools like make and gcc are needed.</DD>
<DD>&nbsp;</DD>
<DT>The following downloads are intended for adventurous end users or
admins with full system souvereignty.</DT>
<DD>Source (./bootstrap is already applied, build tested, for more see
<DD>Source (./bootstrap is already applied, build tested, for more see above
<A HREF="README_cdrskin_devel">upcoming README</A> ):
</DD>
<DD>
<A HREF="cdrskin-0.3.1.tar.gz">cdrskin-0.3.1.tar.gz</A>
(550 KB).
<A HREF="cdrskin-0.2.5.tar.gz">cdrskin-0.2.5.tar.gz</A>
(500 KB).
</DD>
<DD>Binary (untar and move to /usr/bin/cdrskin):</DD>
<DD><A HREF="cdrskin_0.3.1-x86-suse9_0.tar.gz">
cdrskin_0.3.1-x86-suse9_0.tar.gz</A>, (75 KB).
<DD><A HREF="cdrskin_0.2.5-x86-suse9_0.tar.gz">
cdrskin_0.2.5-x86-suse9_0.tar.gz</A>, (60 KB).
</DD>
<DD><A HREF="cdrskin_0.3.1-x86-suse9_0-static.tar.gz">
cdrskin_0.3.1-x86-suse9_0-static.tar.gz</A>, (275 KB)
<DD><A HREF="cdrskin_0.2.5-x86-suse9_0-static.tar.gz">
cdrskin_0.2.5-x86-suse9_0-static.tar.gz</A>, (260 KB)
</DD>
</DL>
</P>
@ -317,6 +305,14 @@ is a GUI frontend which uses cdrecord for CD burning.)
<DT>Example for a test session with a cdrecord based scdbackup installation:</DT>
<DD>$ <KBD><B>cdrskin -scanbus</B></KBD></DD>
<DD><KBD>...</KBD></DD>
</DL>
If your system is stricken with some ill CD device then this can stall
and you will have to press <KBD>Ctrl+C</KBD> to abort.
In this case, you may execute
<KBD>export SCDBACKUP_NO_SCANBUS=1</KBD>
and try again.
<DL>
<DT></DT>
<DD><KBD>&nbsp;&nbsp;&nbsp; 2,0,0 &nbsp;&nbsp; 0)&nbsp; 'TEAC' 'CD-ROM CD-532S' '?' Removable CD-ROM</KBD></DD>
<DD>$ <KBD><B>cdrskin -scanbus dev=ATA</B></KBD></DD>
<DD><KBD>...</KBD></DD>
@ -331,15 +327,8 @@ is a GUI frontend which uses cdrecord for CD burning.)
<DD>$ <KBD><B>export SCDBACKUP_USE_CDRSKIN=1</B></KBD></DD>
<DD>$ <KBD><B>./CONFIGURE_CD</B></KBD></DD>
<DD><KBD>...</KBD></DD>
<DD><KBD>cdrskin 0.3.0 : limited cdrecord compatibility wrapper for libburn</KBD></DD>
</DL>
If your system is stricken with some ill CD device then this can stall
and you will have to press <KBD>Ctrl+C</KBD> to abort.
In this case, you may execute
<KBD>export SCDBACKUP_NO_SCANBUS=1</KBD>
and try again.
<DL>
<DT></DT>
<DD><KBD>cdrskin 0.2.4 : limited cdrecord compatibility wrapper for libburn</KBD></DD>
<DD><KBD>...</KBD></DD>
<DD><KBD> ------------------- SCSI devices. To be used like &nbsp;&nbsp; 0,0,0</KBD></DD>
<DD><KBD>&nbsp;&nbsp;&nbsp; 2,0,0 &nbsp;&nbsp; 0)&nbsp; 'TEAC' 'CD-ROM CD-532S' '?' Removable CD-ROM</KBD></DD>
<DD><KBD> ------------------- end of SCSI device list</KBD></DD>
@ -389,11 +378,10 @@ the gestures necessary for their cdrecord applications.
Contact me. Let's see what we can achieve.
<BR>
<BR>
libburn and cdrskin are now mature enough to substitute cdrecord in its
major use cases of CD burning. It is possible to foist cdrskin on various
software packages if it gets falsely named "cdrecord".
I do not encourage this approach, but of course such a replacement
opportunity is the goal of a cdrecord compatibility wrapper.
I am aware that libburn and cdrskin still have way to go until you can simply
install cdrskin as cdrecord and may expect any application to run with it.
Currently i do not encourage this approach, but of course such a replacement
opportunity is the long term goal of a cdrecord compatibility wrapper.
<BR>
<BR>
It is very important to me that this project is not perceived as hostile
@ -407,10 +395,6 @@ I owe him much. For cdrecord, for mkisofs, for star. Chapeau.
<!-- <A NAME="bottom" HREF="main_ger.html#bottom">deutsch (german)</A>
<BR><BR>
-->
<A HREF="http://en.wikipedia.org/wiki/D%C3%B6ner_kebab">
<IMG SRC="doener_150x200_tr.gif" BORDER=0
ALT="cdrskin logo: Doener mit Scharf"></A>
<BR><BR>
<FONT SIZE=+0>Enjoying free Open Source hosting by <A HREF="http://www.webframe.org">www.webframe.org</A><BR>
<A HREF="http://www.webframe.org">
<IMG SRC="msfree.gif" ALT="100 % Microsoft free" BORDER=0></A><BR>

View File

@ -1 +1 @@
#define Cdrskin_timestamP "2007.01.25.180001"
#define Cdrskin_timestamP "2006.11.02.140329 (pl01)"

File diff suppressed because it is too large Load Diff

View File

@ -23,27 +23,31 @@ typedef void (*sighandler_t)(int);
#include "cleanup.h"
#ifndef Cleanup_has_no_libburn_os_H
#include "../libburn/os.h"
/* see os.h for name of particular os-*.h where this is defined */
static int signal_list[]= { BURN_OS_SIGNAL_MACRO_LIST , -1};
static char *signal_name_list[]= { BURN_OS_SIGNAL_NAME_LIST , "@"};
static int signal_list_count= BURN_OS_SIGNAL_COUNT;
static int non_signal_list[]= { BURN_OS_NON_SIGNAL_MACRO_LIST, -1};
static int non_signal_list_count= BURN_OS_NON_SIGNAL_COUNT;
#else /* ! Cleanup_has_no_libburn_os_H */
/* Outdated. Linux only. For backward compatibility with pre-libburn-0.2.3 */
#ifdef __FreeBSD__
/* Signals to be caught */
static int signal_list[]= {
static int signal_list[]= {
SIGHUP, SIGINT, SIGQUIT, SIGILL, SIGABRT,
SIGFPE, SIGSEGV, SIGPIPE, SIGALRM, SIGTERM,
SIGUSR1, SIGUSR2, SIGXCPU, SIGTSTP, SIGTTIN,
SIGTTOU,
SIGBUS, SIGPROF, SIGSYS, SIGTRAP,
SIGVTALRM, SIGXCPU, SIGXFSZ, -1
};
static char *signal_name_list[]= {
"SIGHUP", "SIGINT", "SIGQUIT", "SIGILL", "SIGABRT",
"SIGFPE", "SIGSEGV", "SIGPIPE", "SIGALRM", "SIGTERM",
"SIGUSR1", "SIGUSR2", "SIGXCPU", "SIGTSTP", "SIGTTIN",
"SIGTTOU",
"SIGBUS", "SIGPROF", "SIGSYS", "SIGTRAP",
"SIGVTALRM", "SIGXCPU", "SIGXFSZ", "@"
};
static int signal_list_count= 23;
#else /* __FreeBSD__ */
/* Signals to be caught */
static int signal_list[]= {
SIGHUP, SIGINT, SIGQUIT, SIGILL, SIGABRT,
SIGFPE, SIGSEGV, SIGPIPE, SIGALRM, SIGTERM,
SIGUSR1, SIGUSR2, SIGXCPU, SIGTSTP, SIGTTIN,
@ -51,7 +55,7 @@ static int signal_list[]= {
SIGBUS, SIGPOLL, SIGPROF, SIGSYS, SIGTRAP,
SIGVTALRM, SIGXCPU, SIGXFSZ, -1
};
static char *signal_name_list[]= {
static char *signal_name_list[]= {
"SIGHUP", "SIGINT", "SIGQUIT", "SIGILL", "SIGABRT",
"SIGFPE", "SIGSEGV", "SIGPIPE", "SIGALRM", "SIGTERM",
"SIGUSR1", "SIGUSR2", "SIGXCPU", "SIGTSTP", "SIGTTIN",
@ -61,21 +65,18 @@ static char *signal_name_list[]= {
};
static int signal_list_count= 24;
#endif /* ! __FreeBSD__ */
/* Signals not to be caught */
static int non_signal_list[]= {
SIGKILL, SIGCHLD, SIGSTOP, SIGURG, SIGWINCH, -1
};
static int non_signal_list_count= 5;
#endif /* Cleanup_has_no_libburn_os_H */
SIGKILL, SIGCHLD, SIGSTOP, SIGURG, -1
};
static int non_signal_list_count= 4;
/* run time dynamic part */
static char cleanup_msg[4096]= {""};
static int cleanup_exiting= 0;
static int cleanup_has_reported= -1234567890;
static void *cleanup_app_handle= NULL;
static Cleanup_app_handler_T cleanup_app_handler= NULL;
@ -86,10 +87,8 @@ static int Cleanup_handler_exit(int exit_value, int signum, int flag)
{
int ret;
if(cleanup_msg[0]!=0 && cleanup_has_reported!=signum) {
if(cleanup_msg[0]!=0)
fprintf(stderr,"\n%s\n",cleanup_msg);
cleanup_has_reported= signum;
}
if(cleanup_perform_app_handler_first)
if(cleanup_app_handler!=NULL) {
ret= (*cleanup_app_handler)(cleanup_app_handle,signum,0);

View File

@ -6,8 +6,7 @@
debug_opts=
def_opts=
largefile_opts="-D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE=1"
libvers="-DCdrskin_libburn_0_3_0"
libvers="-DCdrskin_libburn_0_2_3"
cleanup_src_or_obj="libburn/cleanup.o"
libdax_msgs_o="libburn/libdax_msgs.o"
libdax_audioxtr_o="libburn/libdax_audioxtr.o"
@ -32,16 +31,16 @@ do
libvers="-DCdrskin_libburn_cvs_A60220_tS"
libdax_audioxtr_o=
libdax_msgs_o="libburn/message.o"
cleanup_src_or_obj="-DCleanup_has_no_libburn_os_H cdrskin/cleanup.c"
elif test "$i" = "-libburn_0_3_0"
cleanup_src_or_obj="cdrskin/cleanup.c"
elif test "$i" = "-libburn_0_2_2"
then
libvers="-DCdrskin_libburn_0_3_0"
libdax_audioxtr_o="libburn/libdax_audioxtr.o"
libdax_msgs_o="libburn/libdax_msgs.o"
cleanup_src_or_obj="libburn/cleanup.o"
elif test "$i" = "-libburn_svn"
libvers="-DCdrskin_libburn_0_2_2"
libdax_audioxtr_o=
libdax_msgs_o="libburn/message.o"
cleanup_src_or_obj="cdrskin/cleanup.c"
elif test "$i" = "-libburn_0_2_3"
then
libvers="-DCdrskin_libburn_0_3_1"
libvers="-DCdrskin_libburn_0_2_3"
libdax_audioxtr_o="libburn/libdax_audioxtr.o"
libdax_msgs_o="libburn/libdax_msgs.o"
cleanup_src_or_obj="libburn/cleanup.o"
@ -51,10 +50,6 @@ do
elif test "$i" = "-oldfashioned"
then
def_opts="$def_opts -DCdrskin_oldfashioned_api_usE"
cleanup_src_or_obj="-DCleanup_has_no_libburn_os_H cdrskin/cleanup.c"
elif test "$i" = "-no_largefile"
then
largefile_opts=
elif test "$i" = "-do_not_compile_cdrskin"
then
compile_cdrskin=0
@ -79,9 +74,8 @@ do
echo " -compile_cdrfifo compile program cdrskin/cdrfifo."
echo " -compile_dewav compile program test/dewav without libburn."
echo " -cvs_A60220 set macro to match libburn-CVS of 20 Feb 2006."
echo " -libburn_0_3_0 set macro to match libburn-0.3.0."
echo " -libburn_svn set macro to match current libburn-SVN."
echo " -no_largefile do not use 64 bit off_t (must match libburn)."
echo " -libburn_0_2_2 set macro to match libburn-0.2.2."
echo " -libburn_0_2_3 set macro to match current libburn-SVN."
echo " -do_not_compile_cdrskin omit compilation of cdrskin/cdrskin."
echo " -experimental use newly introduced libburn features."
echo " -oldfashioned use pre-0.2.2 libburn features only."
@ -105,14 +99,7 @@ echo "Build timestamp : $timestamp"
if test "$compile_cdrskin"
then
echo "compiling program cdrskin/cdrskin.c $static_opts $debug_opts $libvers $def_opts $cleanup_src_or_obj"
cc -I. \
$warn_opts \
$static_opts \
$debug_opts \
$libvers \
$largefile_opts \
$def_opts \
\
cc $warn_opts -I. $static_opts $debug_opts $libvers $def_opts \
-DCdrskin_build_timestamP='"'"$timestamp"'"' \
\
-o cdrskin/cdrskin \

View File

@ -1,72 +0,0 @@
#!/bin/sh
#
# convert_man_to_html.sh - ts A61214
#
# Generates a HTML version of man page cdrskin.1
#
# To be executed within the libburn toplevel directory (like ./libburn-0.2.7)
#
# set -x
man_dir=$(pwd)"/cdrskin"
export MANPATH="$man_dir"
manpage="cdrskin"
raw_html=$(pwd)/"cdrskin/raw_man_1_cdrskin.html"
htmlpage=$(pwd)/"cdrskin/man_1_cdrskin.html"
if test -r "$manpage"
then
dummy=dummy
else
echo "Cannot find readable man page source $1" >&2
exit 1
fi
if test -e "$man_dir"/man1
then
dummy=dummy
else
ln -s . "$man_dir"/man1
fi
if test "$1" = "-work_as_filter"
then
# set -x
sed \
-e 's/<meta name="generator" content="groff -Thtml, see www.gnu.org">/<meta name="generator" content="groff -Thtml, via man -H, via cdrskin\/convert_man_to_html.sh">/' \
-e 's/<meta name="Content-Style" content="text\/css">/<meta name="Content-Style" content="text\/css"><META NAME="description" CONTENT="man page of cdrskin"><META NAME="keywords" CONTENT="man cdrskin, manual, cdrskin, CD, CD-RW, CD-R, burning, cdrecord, compatible"><META NAME="robots" CONTENT="follow">/' \
-e 's/<title>CDRSKIN<\/title>/<title>man 1 cdrskin<\/title>/' \
-e 's/<h1 align=center>CDRSKIN<\/h1>/<h1 align=center>man 1 cdrskin<\/h1>/' \
-e 's/<body>/<body BGCOLOR="#F5DEB3" TEXT=#000000 LINK=#0000A0 VLINK=#800000>/' \
-e 's/<b>Overview of features:<\/b>/\&nbsp;<BR><b>Overview of features:<\/b>/' \
-e 's/<b>Track recording model:<\/b>/\&nbsp;<BR><b>Track recording model:<\/b>/' \
-e 's/In general there are two types of tracks: data and audio./\&nbsp;<BR>In general there are two types of tracks: data and audio./' \
-e 's/While audio tracks just contain a given/\&nbsp;<BR>While audio tracks just contain a given/' \
-e 's/<b>Recordable CD Media:<\/b>/\&nbsp;<BR><b>Recordable CD Media:<\/b>/' \
-e 's/<b>Recordable DVD Media:<\/b>/\&nbsp;<BR><b>Recordable DVD Media:<\/b>/' \
-e 's/<b>Drive preparation and addressing:<\/b>/\&nbsp;<BR><b>Drive preparation and addressing:<\/b>/' \
-e 's/If you only got one CD capable drive/\&nbsp;<BR>If you only got one CD capable drive/' \
-e 's/^Alphabetical list of options/\&nbsp;<BR>Alphabetical list of options/' \
-e 's/and for all others\.<\/td><\/table>/and for all others.<\/td><\/table> <BR><HR><FONT SIZE=-1><CENTER>(HTML generated from '"$manpage"'.1 on '"$(date)"' by '$(basename "$0")' )<\/CENTER><\/FONT>/' \
-e 's/See section EXAMPLES/See section <A HREF="#EXAMPLES">EXAMPLES<\/A>/' \
<"$2" >"$htmlpage"
set +x
chmod u+rw,go+r,go-w "$htmlpage"
echo "Emerged file:"
ls -l "$htmlpage"
else
export BROWSER='cp "%s" '"$raw_html"
man -H "$manpage"
"$0" -work_as_filter "$raw_html"
rm "$raw_html"
rm "$man_dir"/man1
fi

Binary file not shown.

Before

Width:  |  Height:  |  Size: 34 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 29 KiB

View File

@ -1,9 +1,8 @@
--------------------------------------------------------------------------
cdrskin Wiki - plain text copy
--------------------------------------------------------------------------
[[Image(source:/libburn/trunk/cdrskin/doener_150x200_tr.gif)]] [http://en.wikipedia.org/wiki/D%C3%B6ner_kebab Doener]
'''cdrskin is the cdrecord compatibility middleware of libburn.'''
cdrskin-0.2.3 is the cdrecord compatibility middleware of libburn.
Its paragon, cdrecord, is a powerful GPL'ed burn program included in Joerg
Schilling's cdrtools. cdrskin strives to be a second source for the services
@ -13,52 +12,43 @@ Its future ability to burn DVD media depends on the development of libburn.
cdrskin does not contain any bytes copied from cdrecord's sources.
Many bytes have been copied from the message output of cdrecord
runs, though. The most comprehensive technical overview of cdrskin
can be found in [http://libburnia.pykix.org/browser/libburn/trunk/cdrskin/README?format=txt cdrskin/README].
About libburn API for burning CD: http://libburnia-api.pykix.org
can be found in cdrskin/README .
--------------------------------------------------------------------------
Appending sessions to an unclosed CD is restricted to write mode TAO.
(Users who have a burner which succeeds with a follow-up session via
cdrecord -sao : please contact us.)
cdrskin with CD media fails to match its paragon cdrecord on two major
fields: convenient TAO burn mode and multi session.
The development version of cdrskin is able to burn a single track to DVD+RW
or DVD-RW media.
For other DVD types and for appending sessions to ISO filesystems see the
advise to use dvd+rw-tools at the end of this text.
cdrskin does not provide DVD burning yet. See advise to use dvd+rw-tools
at the end of this text.
--------------------------------------------------------------------------
About the command line options of cdrskin:
They are described in detail in [http://scdbackup.sourceforge.net/man_1_cdrskin_devel.html#OPTIONS section OPTIONS] of
[http://scdbackup.sourceforge.net/man_1_cdrskin_devel.html man cdrskin]
There are two families of options: cdrecord-compatible ones and options
which are specific to cdrskin. The latter are mostly used to configure
cdrskin for its task to emulate cdrecord. There are some, nevertheless,
which provide rather exotic unique features of cdrskin.
The cdrecord-compatible options are listed in the output of
{{{
cdrskin -help
}}}
where the option "help" has *one* dash. Online: [http://scdbackup.sourceforge.net/cdrskin_help_devel cdrskin -help]
cdrskin -help
where the option "help" has *one* dash.
For these options you may expect program behavior that is roughly the
same as described in original man cdrecord .
same as described in original man 1 cdrecord .
Online: http://cdrecord.berlios.de/old/private/man/cdrecord-2.0.html
The cdrskin-specific options are listed by
{{{
cdrskin --help
}}}
where the option "help" has *two* dashes. Online: [http://scdbackup.sourceforge.net/cdrskin__help_devel cdrskin --help]
Some are very experimental and should only be
used in coordination with the libburnia developer team.
cdrskin --help
where the option "help" has *two* dashes.
Those have no man page yet. Some are very experimental and should only be
used in coordination with the libburn developer team.
Some are of general user interest, though:
--------------------------------------------------------------------------
@ -66,10 +56,11 @@ Some are of general user interest, though:
--devices allows the sysadmin to scan the system for possible drives
and displays their detected properties.
The drives are listed one per line, with fields:
libburn-drive-number, sysadmin-device-file, permissions, vendor, type
{{{
libburn-drive-number sysadmin-device-file permissions : vendor type
0 dev='/dev/sg0' rwrw-- : 'HL-DT-ST' 'DVDRAM GSA-4082B'
}}}
This feature is valuable since cdrskin -scanbus will not give you
the device file name and its current permissions.
cdrskin will accept of course the proposed dev= option as address
@ -83,23 +74,7 @@ has to offer both, r- and w-permission.
--------------------------------------------------------------------------
Non-cdrecord blank mode blank=format_overwrite is needed to bring a DVD-RW
disc from its initial profile "Sequential Recording" into profile state
"Restricted Overwrite". The latter is usable with cdrskin.
{{{
cdrskin dev=/dev/sr0 -v blank=format_overwrite
}}}
DVD-RW "Restricted Overwrite" and DVD+RW appear to cdrskin as blank media
which are capable of taking only a single track. This track may be positioned
on a 32KiB aligned address, though.
{{{
cdrskin ... write_start_address=2412m ...
}}}
--------------------------------------------------------------------------
fifo_start_at=<num> is a throughput enhancer for unsteady data streams
--fifo_start_empty is a throughput enhancer for unsteady data streams
like they are produced by a compressing archiver program when piping to
CD on-the-fly. It makes better use of the general property of a FIFO
buffer to transport surplus bandwidth into the future. Yep. A time machine.
@ -123,16 +98,10 @@ underruns, of course.
With a very fat fs=# buffer (128 MB for 12x CD is not unrealistic) this
can cause a big delay until burning finally starts and takes its due time.
fifo_start_at=<num> makes cdrskin start burning after the given number of bytes
is read rather than waiting for the FIFO to be completely full resp. the data
stream to end. It risks a few drive buffer underruns at the beginning of burn
- but modern drives stand this.
Usage examples:
{{{
cdrskin ... fs=128m fifo_start_at=20m ...
cdrskin ... fifo_start_at=0 ...
}}}
--fifo_start_empty makes cdrskin start burning without waiting for the
FIFO to be full resp. the data stream to end. It can make use of the
seconds spend with drive preparation and lead-in, it risks a few drive
buffer underruns at the beginning of burn - but modern drives stand this.
Note: no FIFO can give you better average throughput than the average
throughput of the data source and the throughput of the burner.
@ -148,8 +117,6 @@ default settings of cdrskin. Possible locations for such settings:
/etc/opt/cdrskin/rc
/etc/cdrskin/cdrskin.conf
$HOME/.cdrskinrc
--------------------------------------------------------------------------
@ -158,10 +125,6 @@ tao_to_sao_tsize=<num> allows the - actually unsupported - cdrecord option
-tao and defines a default track size to be used if - as custom with -tao -
no option tsize=# is given.
Since -tao is supported in cdrskin-0.2.6 the TAO-to-SAO workaround has become
quite obsolete. Nevertheless, tao_to_sao_tsize= allows to preset a default
size for SAO mode which is in effect only if no track size is available.
As in general with cdrskin tsize=# the data source does not have to provide
the full annouced amount of data. Missing data will be padded up by 0-bytes.
Surplus data is supposed to cause an error, though. The burn will then
@ -184,15 +147,15 @@ cdrskin the necessary hint.
Example: Your frontend insists in using "0,0,0" and --devices reported
dev='/dev/hdc' resp. cdrskin dev=ATA -scanbus reported "1,0,0" then this
would be the appropriate translation:
{{{
dev_translation=+0,0,0+/dev/hdc
}}}
The "+" character is a separator to be choosen by you.
Currently i am not aware of the need to choose any other than "+"
unless you get playful with custom translations like
{{{
dev_translation=-"cd+dvd"-1,0,0
}}}
See http://scdbackup.sourceforge.net/k3b_on_cdrskin.html
for an illustrated example with K3b 0.10 .
@ -210,9 +173,12 @@ They are not compatible or related to cdrecord resp. cdrecord-ProDVD
(now obsoleted by original source cdrtools cdrecord with identical
capabilities besides the license key).
libburn and thus the cdrskin project are currently aquiring own capabilities
to burn to DVD media. For now restricted to DVD+RW and DVD-RW and to single
tracks.
If there is sincere and well motivated interest, the cdrskin project could try
to employ growisofs as DVD burning engine. The cdrskin project would prefer to
wait for DVD support being included in libburn, though.
A very limited and specialized cdrecord-compatibility wrapper for growisofs
serves in my project scdbackup. It is not overly hard to make one that serves
some very few fixed use cases.
To my knowledge, Linux kernels 2.6 do write to DVD+RW via block devices as
they would write to a traditional tape device. Try old tape archiver

View File

@ -1,4 +1,4 @@
AC_INIT([libburn], [0.3.0], [http://libburnia.pykix.org])
AC_INIT([libburn], [0.2.3], [http://libburn.pykix.org])
AC_PREREQ([2.50])
dnl AC_CONFIG_HEADER([config.h])
@ -24,8 +24,8 @@ dnl
dnl if MAJOR or MINOR version changes, be sure to change AC_INIT above to match
dnl
BURN_MAJOR_VERSION=0
BURN_MINOR_VERSION=3
BURN_MICRO_VERSION=0
BURN_MINOR_VERSION=2
BURN_MICRO_VERSION=3
BURN_INTERFACE_AGE=0
BURN_BINARY_AGE=0
BURN_VERSION=$BURN_MAJOR_VERSION.$BURN_MINOR_VERSION.$BURN_MICRO_VERSION
@ -63,7 +63,6 @@ AC_C_BIGENDIAN
dnl Large file support
AC_SYS_LARGEFILE
AC_FUNC_FSEEKO
AC_CHECK_FUNC([fseeko])
if test ! $ac_cv_func_fseeko; then
AC_ERROR([Libburn requires largefile support.])
fi
@ -113,5 +112,6 @@ AC_CONFIG_FILES([
doc/doxygen.conf
version.h
libburn-1.pc
libisofs-1.pc
])
AC_OUTPUT

View File

@ -1,31 +1,29 @@
/**
@author Mario Danic, Thomas Schmitt
@mainpage Libburnia Documentation Index
@mainpage Libburn Documentation Index
@section intro Introduction
Libburnia is an open-source project for reading, mastering and writing
Libburn is an open-source library for reading, mastering and writing
optical discs. For now this means only CD-R and CD-RW.
Support for DVD+RW and DVD-RW is emerging.
The project comprises of several more or less interdependent parts which
together strive to be a usable foundation for application development.
These are libraries, language bindings, and middleware binaries which emulate
classical (and valuable) Linux tools.
Our scope is currently Linux 2.4 and 2.6 only. For ports to other systems
we would need : login on a development machine resp. a live OS on CD or DVD,
advise from a system person about the equivalent of Linux sg or FreeBSD CAM,
volunteers for testing of realistic use cases.
Our scope is currently Linux 2.4 and 2.6 and we will have a hard time to widen
this for now, because of our history. The project could need advise from or
membership of skilled kernel people and people who know how to talk CD/DVD
drives into doing things.
We have a workable code base for burning data and audio CDs. The burn API is
We do have a workable code base for burning data CDs, though. The burn API is
quite comprehensively documented and can be used to build a presentable
application.
We have a functional binary which emulates the core use cases of cdrecord in
order to prove that usability, and in order to allow you to explore libburn's
scope by help of existing cdrecord frontends.
We do have a functional binary which emulates parts of cdrecord in order to
prove that usability, and in order to allow you to explore libburn's scope
by help of existing cdrecord frontends.
@subsection components The project components (list subject to growth, hopefully):
@ -84,25 +82,25 @@ libburner is a minimal demo application for the library libburn
(see: libburn/libburn.h) as provided on http://libburn.pykix.org .
It can list the available devices, can blank a CD-RW and
can burn to CD-R or CD-RW.
New: burning to DVD+/-RW (single data track, single session only).
It's main purpose, nevertheless, is to show you how to use libburn and also
to serve the libburnia team as reference application. libburner does indeed
to serve the libburn team as reference application. libburner does indeed
define the standard way how above three gestures can be implemented and
stay upward compatible for a good while.
@subsection libburner-help Libburner --help
<pre>
Usage: test/libburner
[--drive <address>|<driveno>|"-"] [--audio]
[--blank_fast|--blank_full] [--try_to_simulate]
[--multi] [one or more imagefiles|"-"]
[--drive <address>|<driveno>|"-"]
[--blank_fast|--blank_full] [--audio]
[--try_to_simulate] [--stdin_size <bytes>]
[<one or more imagefiles>|"-"]
Examples
A bus scan (needs rw-permissions to see a drive):
test/libburner --drive -
Burn a file to drive chosen by number, leave appendable:
test/libburner --drive 0 --multi my_image_file
Burn a file to drive chosen by persistent address, close:
Burn a file to drive chosen by number:
test/libburner --drive 0 my_image_file
Burn a file to drive chosen by persistent address:
test/libburner --drive /dev/hdc my_image_file
Blank a used CD-RW (is combinable with burning in one run):
test/libburner --drive /dev/hdc --blank_fast
@ -110,22 +108,12 @@ Burn two audio tracks
lame --decode -t /path/to/track1.mp3 track1.cd
test/dewav /path/to/track2.wav -o track2.cd
test/libburner --drive /dev/hdc --audio track1.cd track2.cd
Burn a compressed afio archive on-the-fly:
Burn a compressed afio archive on-the-fly, pad up to 700 MB:
( cd my_directory ; find . -print | afio -oZ - ) | \
test/libburner --drive /dev/hdc -
test/libburner --drive /dev/hdc --stdin_size 734003200 -
To be read from *not mounted* CD via: afio -tvZ /dev/hdc
Program tar would need a clean EOF which our padded CD cannot deliver.
</pre>
libburner has two companions, telltoc and dewav, which help to perform some
peripheral tasks of burning.
telltoc prints a table of content (sessions, tracks and leadouts), it tells
about type and state of CD media, and also is able to provide the necessary
multi-session information for program mkisofs option -C.
See: test/telltoc --help.
dewav extracts raw byte-swapped audio data from files of format .wav (MS WAVE)
or .au (SUN Audio). See example in libburner --help.
@subsection libburner-source Sourceode of libburner

121
doc/comments_test_ts Normal file
View File

@ -0,0 +1,121 @@
/**
@author Mario Danic, Thomas Schmitt
@mainpage Libburn Documentation Index
@section intro Introduction
Libburn is an open-source library for reading, mastering and writing
optical discs. For now this means only CD-R and CD-RW.
The project comprises of several more or less interdependent parts which
together strive to be a usable foundation for application development.
These are libraries, language bindings, and middleware binaries which emulate
classical (and valuable) Linux tools.
Our scope is currently Linux 2.4 and 2.6 and we will have a hard time to widen
this for now, because of our history. The project could need advise from or
membership of skilled kernel people and people who know how to talk CD/DVD
drives into doing things.
We do have a workable code base for burning data CDs, though. The burn API is
quite comprehensively documented and can be used to build a presentable
application.
We do have a functional binary which emulates parts of cdrecord in order to
prove that usability, and in order to allow you to explore libburn's scope
by help of existing cdrecord frontends.
@subsection components The project components (list subject to growth, hopefully):
- libburn is the library by which preformatted data get onto optical media.
It uses either /dev/sgN (e.g. on kernel 2.4 with ide-scsi) or
/dev/hdX (e.g. on kernel 2.6).
libburn is the foundation of our cdrecord emulation.
- libisofs is the library to pack up hard disk files and directories into a
ISO 9660 disk image. This may then be brought to CD via libburn.
libisofs is to be the foundation of our upcoming mkisofs emulation.
- cdrskin is a limited cdrecord compatibility wrapper for libburn.
cdrecord is a powerful GPL'ed burn program included in Joerg
Schilling's cdrtools. cdrskin strives to be a second source for
the services traditionally provided by cdrecord.
cdrskin does not contain any bytes copied from cdrecord's sources.
Many bytes have been copied from the message output of cdrecord
runs, though.
See cdrskin/README for more.
- "test" is a collection of application gestures and examples given by the
authors of the library features. The main API example of libburn
is named test/libburner.c .
Explore these examples if you look for inspiration.
We plan to be a responsive upstream. Bear with us.
@section using Using the libraries
Our build system is based on autotools.
User experience tells us that you will need at least autotools version 1.7.
To build libburn and its subprojects it should be sufficient to go into
its toplevel directory and execute
- ./bootstrap (needed if you downloaded from SVN)
- ./configure
- make
To make the libraries accessible for running resp. developing applications
- make install
Both libraries are written in C language and get built by autotools.
Thus we expect them to be useable by a wide range of Linux-implemented
languages and development tools.
@section libburner Libburner
libburner is a minimal demo application for the library libburn
(see: libburn/libburn.h) as provided on http://libburn.pykix.org .
It can list the available devices, can blank a CD-RW and
can burn to CD-R or CD-RW.
It's main purpose, nevertheless, is to show you how to use libburn and also
to serve the libburn team as reference application. libburner does indeed
define the standard way how above three gestures can be implemented and
stay upward compatible for a good while.
@subsection libburner-help Libburner --help
<pre>
Usage: test/libburner
[--drive <address>|<driveno>|"-"]
[--verbose <level>] [--blank_fast|--blank_full]
[--burn_for_real|--try_to_simulate] [--stdin_size <bytes>]
[<imagefile>|"-"]
Examples
A bus scan (needs rw-permissions to see a drive):
test/libburner --drive -
Burn a file to drive chosen by number:
test/libburner --drive 0 --burn_for_real my_image_file
Burn a file to drive chosen by persistent address:
test/libburner --drive /dev/hdc --burn_for_real my_image_file
Blank a used CD-RW (is combinable with burning in one run):
test/libburner --drive 0 --blank_fast
Burn a compressed afio archive on-the-fly, pad up to 700 MB:
( cd my_directory ; find . -print | afio -oZ - ) | \
test/libburner --drive /dev/hdc --burn_for_real --stdin_size 734003200 -
To be read from *not mounted* CD via:
afio -tvZ /dev/hdc
Program tar would need a clean EOF which our padded CD cannot deliver.
</pre>
@subsection libburner-source Sourceode of libburner
Click on blue names of functions, structures, variables, etc in oder to
get to the according specs of libburn API or libburner sourcecode.
@include libburner.c
*/

View File

@ -56,8 +56,8 @@ WARN_LOGFILE =
#---------------------------------------------------------------------------
# configuration options related to the input files
#---------------------------------------------------------------------------
INPUT = libburn doc test
FILE_PATTERNS = libburn.h comments libburner.c
INPUT = libburn libisofs doc test
FILE_PATTERNS = libburn.h libisofs.h comments libburner.c
RECURSIVE = NO
EXCLUDE =
EXCLUDE_SYMLINKS = NO

View File

@ -7,11 +7,9 @@
#include "options.h"
#include "async.h"
#include "init.h"
#include "back_hacks.h"
#include <pthread.h>
#include <stdlib.h>
#include <stdio.h>
/*
#include <a ssert.h>
@ -37,14 +35,6 @@ struct erase_opts
int fast;
};
/* ts A61230 */
struct format_opts
{
struct burn_drive *drive;
off_t size;
int flag;
};
struct write_opts
{
struct burn_drive *drive;
@ -52,7 +42,6 @@ struct write_opts
struct burn_disc *disc;
};
struct w_list
{
struct burn_drive *drive;
@ -64,7 +53,6 @@ struct w_list
{
struct scan_opts scan;
struct erase_opts erase;
struct format_opts format;
struct write_opts write;
} u;
};
@ -210,9 +198,6 @@ void burn_disc_erase(struct burn_drive *drive, int fast)
{
struct erase_opts o;
/* A70103 : will be set to 0 by burn_disc_erase_sync() */
drive->cancel = 1;
/* ts A61006 */
/* a ssert(drive); */
/* a ssert(!SCAN_GOING()); */
@ -233,81 +218,11 @@ void burn_disc_erase(struct burn_drive *drive, int fast)
return;
}
/* ts A70103 moved up from burn_disc_erase_sync() */
/* ts A60825 : allow on parole to blank appendable CDs */
if ( ! (drive->status == BURN_DISC_FULL ||
(drive->status == BURN_DISC_APPENDABLE &&
! libburn_back_hack_42) ) ) {
libdax_msgs_submit(libdax_messenger, drive->global_index,
0x00020130,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
"Drive and media state unsuitable for blanking",
0, 0);
return;
}
o.drive = drive;
o.fast = fast;
add_worker(drive, (WorkerFunc) erase_worker_func, &o);
}
/* ts A61230 */
static void *format_worker_func(struct w_list *w)
{
burn_disc_format_sync(w->u.format.drive, w->u.format.size,
w->u.format.flag);
remove_worker(pthread_self());
return NULL;
}
/* ts A61230 */
void burn_disc_format(struct burn_drive *drive, off_t size, int flag)
{
struct format_opts o;
int ok = 0;
char msg[160];
if ((SCAN_GOING()) || find_worker(drive)) {
libdax_msgs_submit(libdax_messenger, drive->global_index,
0x00020102,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
"A drive operation is still going on (want to format)",
0, 0);
return;
}
if (flag & 128) /* application prescribed format type */
flag |= 16; /* enforce re-format */
if (drive->current_profile == 0x14)
ok = 1; /* DVD-RW sequential */
else if (drive->current_profile == 0x13 && (flag & 16))
ok = 1; /* DVD-RW Restricted Overwrite with force bit */
else if (drive->current_profile == 0x1a) {
ok = 1; /* DVD+RW */
size = 0;
flag &= ~(2|8); /* no insisting in size 0, no expansion */
flag |= 4; /* format up to maximum size */
}
if (!ok) {
sprintf(msg,"Will not format media type %4.4Xh",
drive->current_profile);
libdax_msgs_submit(libdax_messenger, drive->global_index,
0x00020129,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
msg, 0, 0);
drive->cancel = 1;
return;
}
o.drive = drive;
o.size = size;
o.flag = flag;
add_worker(drive, (WorkerFunc) format_worker_func, &o);
}
static void *write_disc_worker_func(struct w_list *w)
{
burn_disc_write_sync(w->u.write.opts, w->u.write.disc);

View File

@ -23,27 +23,31 @@ typedef void (*sighandler_t)(int);
#include "cleanup.h"
#ifndef Cleanup_has_no_libburn_os_H
#include "../libburn/os.h"
/* see os.h for name of particular os-*.h where this is defined */
static int signal_list[]= { BURN_OS_SIGNAL_MACRO_LIST , -1};
static char *signal_name_list[]= { BURN_OS_SIGNAL_NAME_LIST , "@"};
static int signal_list_count= BURN_OS_SIGNAL_COUNT;
static int non_signal_list[]= { BURN_OS_NON_SIGNAL_MACRO_LIST, -1};
static int non_signal_list_count= BURN_OS_NON_SIGNAL_COUNT;
#else /* ! Cleanup_has_no_libburn_os_H */
/* Outdated. Linux only. For backward compatibility with pre-libburn-0.2.3 */
#ifdef __FreeBSD__
/* Signals to be caught */
static int signal_list[]= {
static int signal_list[]= {
SIGHUP, SIGINT, SIGQUIT, SIGILL, SIGABRT,
SIGFPE, SIGSEGV, SIGPIPE, SIGALRM, SIGTERM,
SIGUSR1, SIGUSR2, SIGXCPU, SIGTSTP, SIGTTIN,
SIGTTOU,
SIGBUS, SIGPROF, SIGSYS, SIGTRAP,
SIGVTALRM, SIGXCPU, SIGXFSZ, -1
};
static char *signal_name_list[]= {
"SIGHUP", "SIGINT", "SIGQUIT", "SIGILL", "SIGABRT",
"SIGFPE", "SIGSEGV", "SIGPIPE", "SIGALRM", "SIGTERM",
"SIGUSR1", "SIGUSR2", "SIGXCPU", "SIGTSTP", "SIGTTIN",
"SIGTTOU",
"SIGBUS", "SIGPROF", "SIGSYS", "SIGTRAP",
"SIGVTALRM", "SIGXCPU", "SIGXFSZ", "@"
};
static int signal_list_count= 23;
#else /* __FreeBSD__ */
/* Signals to be caught */
static int signal_list[]= {
SIGHUP, SIGINT, SIGQUIT, SIGILL, SIGABRT,
SIGFPE, SIGSEGV, SIGPIPE, SIGALRM, SIGTERM,
SIGUSR1, SIGUSR2, SIGXCPU, SIGTSTP, SIGTTIN,
@ -51,7 +55,7 @@ static int signal_list[]= {
SIGBUS, SIGPOLL, SIGPROF, SIGSYS, SIGTRAP,
SIGVTALRM, SIGXCPU, SIGXFSZ, -1
};
static char *signal_name_list[]= {
static char *signal_name_list[]= {
"SIGHUP", "SIGINT", "SIGQUIT", "SIGILL", "SIGABRT",
"SIGFPE", "SIGSEGV", "SIGPIPE", "SIGALRM", "SIGTERM",
"SIGUSR1", "SIGUSR2", "SIGXCPU", "SIGTSTP", "SIGTTIN",
@ -61,21 +65,18 @@ static char *signal_name_list[]= {
};
static int signal_list_count= 24;
#endif /* ! __FreeBSD__ */
/* Signals not to be caught */
static int non_signal_list[]= {
SIGKILL, SIGCHLD, SIGSTOP, SIGURG, SIGWINCH, -1
};
static int non_signal_list_count= 5;
#endif /* Cleanup_has_no_libburn_os_H */
SIGKILL, SIGCHLD, SIGSTOP, SIGURG, -1
};
static int non_signal_list_count= 4;
/* run time dynamic part */
static char cleanup_msg[4096]= {""};
static int cleanup_exiting= 0;
static int cleanup_has_reported= -1234567890;
static void *cleanup_app_handle= NULL;
static Cleanup_app_handler_T cleanup_app_handler= NULL;
@ -86,10 +87,8 @@ static int Cleanup_handler_exit(int exit_value, int signum, int flag)
{
int ret;
if(cleanup_msg[0]!=0 && cleanup_has_reported!=signum) {
if(cleanup_msg[0]!=0)
fprintf(stderr,"\n%s\n",cleanup_msg);
cleanup_has_reported= signum;
}
if(cleanup_perform_app_handler_first)
if(cleanup_app_handler!=NULL) {
ret= (*cleanup_app_handler)(cleanup_app_handle,signum,0);

View File

@ -25,9 +25,7 @@
#include "util.h"
#include "sg.h"
#include "structure.h"
/* ts A70107 : to get BE_CANCELLED */
#include "error.h"
#include "back_hacks.h"
#include "libdax_msgs.h"
extern struct libdax_msgs *libdax_messenger;
@ -54,9 +52,8 @@ void burn_drive_free(struct burn_drive *d)
return;
/* ts A60822 : close open fds before forgetting them */
if (burn_drive_is_open(d))
d->release(d);
sg_close_drive(d);
free((void *) d->idata);
burn_mdata_free_subs(d->mdata);
free((void *) d->mdata);
if(d->toc_entry != NULL)
free((void *) d->toc_entry);
@ -138,39 +135,53 @@ unsigned int burn_drive_count(void)
return drivetop + 1;
}
/* ts A61125 : media status aspects of burn_drive_grab() */
int burn_drive_inquire_media(struct burn_drive *d)
int burn_drive_grab(struct burn_drive *d, int le)
{
int errcode;
int was_equal = 0, must_equal = 3, max_loop = 20;
/* ts A61225 : after loading the tray, mode page 2Ah can change */
d->getcaps(d);
/* ts A60907 */
int loop_count, old_speed = -1234567890, new_speed = -987654321;
int old_erasable = -1234567890, new_erasable = -987654321;
/* ts A61020 : d->status was set to BURN_DISC_BLANK as pure guess */
if (!d->released) {
burn_print(1, "can't grab - already grabbed\n");
return 0;
}
errcode = d->grab(d);
if (errcode == 0) {
burn_print(1, "low level drive grab failed\n");
return 0;
}
d->busy = BURN_DRIVE_GRABBING;
if (le)
d->load(d);
d->lock(d);
/* ts A61020 : this was BURN_DISC_BLANK as pure guess */
d->status = BURN_DISC_UNREADY;
if (d->mdata->cdr_write || d->mdata->cdrw_write ||
d->mdata->dvdr_write || d->mdata->dvdram_write) {
#define Libburn_knows_correct_state_after_loaD 1
#ifdef Libburn_knows_correct_state_after_loaD
#ifdef Libburn_grab_release_and_grab_agaiN
d->read_disc_info(d);
#else
/* ts A61227 : This repeated read_disc_info seems
to be obsoleted by above d->getcaps(d).
*/
/* ts A60908 */
/* Trying to stabilize the disc status after eventual load
without closing and re-opening the drive */
/* This seems to work for burn_disc_erasable() .
Speed values on RIP-14 and LITE-ON 48125S are stable
and false, nevertheless. */
int was_equal = 0, must_equal = 3, max_loop = 20;
int loop_count, old_speed = -1234567890, new_speed= -987654321;
int old_erasable = -1234567890, new_erasable = -987654321;
fprintf(stderr,"LIBBURN_DEBUG: read_disc_info()\n");
and false, nevertheless. So cdrskin -atip is still
forced to finish-initialize. */
/*
fprintf(stderr,"libburn: experimental: read_disc_info()\n");
*/
for (loop_count = 0; loop_count < max_loop; loop_count++){
old_speed = new_speed;
old_erasable = new_erasable;
@ -190,57 +201,18 @@ int burn_drive_inquire_media(struct burn_drive *d)
was_equal = 0;
/*
if (loop_count >= 1 && was_equal == 0)
*/
fprintf(stderr,"LIBBURN_DEBUG: %d : speed %d:%d erasable %d:%d\n",
fprintf(stderr,"libburn: experimental: %d : speed %d:%d erasable %d:%d\n",
loop_count,old_speed,new_speed,old_erasable,new_erasable);
*/
usleep(100000);
}
#endif /* ! Libburn_knows_correct_state_after_loaD */
#endif /* ! Libburn_grab_release_and_grab_agaiN */
} else {
if (d->current_profile == -1 || d->current_is_cd_profile)
d->read_toc(d);
d->read_toc(d);
}
return 1;
}
int burn_drive_grab(struct burn_drive *d, int le)
{
int errcode;
/* ts A61125 - A61202 */
int ret, sose;
if (!d->released) {
burn_print(1, "can't grab - already grabbed\n");
return 0;
}
d->status = BURN_DISC_UNREADY;
errcode = d->grab(d);
if (errcode == 0) {
burn_print(1, "low level drive grab failed\n");
return 0;
}
d->busy = BURN_DRIVE_GRABBING;
if (le)
d->load(d);
d->lock(d);
/* ts A61118 */
d->start_unit(d);
/* ts A61202 : gave bit1 of le a meaning */
sose = d->silent_on_scsi_error;
if (!le)
d->silent_on_scsi_error = 1;
/* ts A61125 : outsourced media state inquiry aspects */
ret = burn_drive_inquire_media(d);
d->silent_on_scsi_error = sose;
d->busy = BURN_DRIVE_IDLE;
return ret;
return 1;
}
struct burn_drive *burn_drive_register(struct burn_drive *d)
@ -338,28 +310,6 @@ struct burn_drive *burn_drive_finish_enum(struct burn_drive *d)
}
/* ts A61125 : model aspects of burn_drive_release */
int burn_drive_mark_unready(struct burn_drive *d)
{
/* ts A61020 : mark media info as invalid */
d->start_lba= -2000000000;
d->end_lba= -2000000000;
/* ts A61202 */
d->current_profile = -1;
d->status = BURN_DISC_UNREADY;
if (d->toc_entry != NULL)
free(d->toc_entry);
d->toc_entry = NULL;
d->toc_entries = 0;
if (d->disc != NULL) {
burn_disc_free(d->disc);
d->disc = NULL;
}
return 1;
}
void burn_drive_release(struct burn_drive *d, int le)
{
if (d->released) {
@ -383,14 +333,26 @@ void burn_drive_release(struct burn_drive *d, int le)
return;
}
/* ts A61020 : mark media info as invalid */
d->start_lba= -2000000000;
d->end_lba= -2000000000;
d->unlock(d);
if (le)
d->eject(d);
d->release(d);
d->released = 1;
/* ts A61125 : outsourced model aspects */
burn_drive_mark_unready(d);
d->release(d);
d->status = BURN_DISC_UNREADY;
d->released = 1;
if (d->toc_entry != NULL)
free(d->toc_entry);
d->toc_entry = NULL;
d->toc_entries = 0;
if (d->disc != NULL) {
burn_disc_free(d->disc);
d->disc = NULL;
}
}
@ -446,6 +408,11 @@ void burn_disc_erase_sync(struct burn_drive *d, int fast)
burn_print(1, "erasing drive %s %s\n", d->idata->vendor,
d->idata->product);
/* ts A60825 : allow on parole to blank appendable CDs */
if ( ! (d->status == BURN_DISC_FULL ||
(d->status == BURN_DISC_APPENDABLE &&
! libburn_back_hack_42) ) )
return;
d->cancel = 0;
d->busy = BURN_DRIVE_ERASING;
d->erase(d, fast);
@ -463,147 +430,12 @@ void burn_disc_erase_sync(struct burn_drive *d, int fast)
while (!d->test_unit_ready(d) && d->get_erase_progress(d) == 0)
sleep(1);
while ((d->progress.sector = d->get_erase_progress(d)) > 0 ||
!d->test_unit_ready(d))
!d->test_unit_ready(d))
sleep(1);
d->progress.sector = 0x10000;
/* ts A61125 : update media state records */
burn_drive_mark_unready(d);
burn_drive_inquire_media(d);
d->busy = BURN_DRIVE_IDLE;
}
/*
@param flag: bit0 = fill formatted size with zeros
bit1, bit2 , bit4, bit7 - bit15 are for d->format_unit()
*/
void burn_disc_format_sync(struct burn_drive *d, off_t size, int flag)
{
int ret, buf_secs, err, i, stages = 1, pbase, pfill, pseudo_sector;
off_t num_bufs;
char msg[80];
struct buffer buf;
/* reset the progress */
d->progress.session = 0;
d->progress.sessions = 1;
d->progress.track = 0;
d->progress.tracks = 1;
d->progress.index = 0;
d->progress.indices = 1;
d->progress.start_sector = 0;
d->progress.sectors = 0x10000;
d->progress.sector = 0;
stages = 1 + ((flag & 1) && size > 1024 * 1024);
d->cancel = 0;
d->busy = BURN_DRIVE_FORMATTING;
ret = d->format_unit(d, size, flag & 0xff96); /* forward bits */
if (ret <= 0)
d->cancel = 1;
while (!d->test_unit_ready(d) && d->get_erase_progress(d) == 0)
sleep(1);
while ((pseudo_sector = d->get_erase_progress(d)) > 0 ||
!d->test_unit_ready(d)) {
d->progress.sector = pseudo_sector / stages;
sleep(1);
}
d->sync_cache(d);
if (size <= 0)
goto ex;
/* update media state records */
burn_drive_mark_unready(d);
burn_drive_inquire_media(d);
if (flag & 1) {
/* write size in zeros */;
pbase = 0x8000 + 0x7fff * (stages == 1);
pfill = 0xffff - pbase;
buf_secs = 16; /* Must not be more than 16 */
num_bufs = size / buf_secs / 2048;
if (num_bufs > 0x7fffffff) {
d->cancel = 1;
goto ex;
}
/* <<< */
sprintf(msg,
"Writing %.f sectors of zeros to formatted media",
(double) num_bufs * (double) buf_secs);
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00000002,
LIBDAX_MSGS_SEV_DEBUG, LIBDAX_MSGS_PRIO_ZERO,
msg, 0, 0);
d->buffer = &buf;
memset(d->buffer, 0, sizeof(struct buffer));
d->buffer->bytes = buf_secs * 2048;
d->buffer->sectors = buf_secs;
d->busy = BURN_DRIVE_WRITING;
for (i = 0; i < num_bufs; i++) {
d->nwa = i * buf_secs;
err = d->write(d, d->nwa, d->buffer);
if (err == BE_CANCELLED || d->cancel) {
d->cancel = 1;
break;
}
d->progress.sector = pbase
+ pfill * ((double) i / (double) num_bufs);
}
d->sync_cache(d);
if (d->current_profile == 0x13 || d->current_profile == 0x1a) {
/* DVD-RW or DVD+RW */
d->busy = BURN_DRIVE_CLOSING_SESSION;
/* CLOSE SESSION, 010b */
d->close_track_session(d, 1, 0);
d->busy = BURN_DRIVE_WRITING;
}
}
ex:;
d->progress.sector = 0x10000;
d->busy = BURN_DRIVE_IDLE;
d->buffer = NULL;
}
/* ts A70112 API */
int burn_disc_get_formats(struct burn_drive *d, int *status, off_t *size,
unsigned *bl_sas, int *num_formats)
{
int ret;
*status = 0;
*size = 0;
*bl_sas = 0;
*num_formats = 0;
ret = d->read_format_capacities(d, 0x00);
if (ret <= 0)
return 0;
*status = d->format_descr_type;
*size = d->format_curr_max_size;
*bl_sas = d->format_curr_blsas;
*num_formats = d->num_format_descr;
return 1;
}
/* ts A70112 API */
int burn_disc_get_format_descr(struct burn_drive *d, int index,
int *type, off_t *size, unsigned *tdp)
{
*type = 0;
*size = 0;
*tdp = 0;
if (index < 0 || index >= d->num_format_descr)
return 0;
*type = d->format_descriptors[index].type;
*size = d->format_descriptors[index].size;
*tdp = d->format_descriptors[index].tdp;
return 1;
}
enum burn_disc_status burn_disc_get_status(struct burn_drive *d)
{
/* ts A61007 */
@ -746,10 +578,8 @@ int burn_drive_scan_sync(struct burn_drive_info *drives[],
#endif /* 0 */
/* refresh the lib's drives */
/* ts A61115 : formerly sg_enumerate(); ata_enumerate(); */
scsi_enumerate_drives();
sg_enumerate();
ata_enumerate();
count = burn_drive_count();
if (count)
*drives =
@ -1182,7 +1012,7 @@ int burn_drive_find_scsi_equiv(char *path, char adr[])
ret = burn_drive_obtain_scsi_adr(path, &bus_no, &host_no, &channel_no,
&target_no, &lun_no);
if(ret <= 0) {
sprintf(msg,"burn_drive_obtain_scsi_adr( %s ) returns %d",
sprintf(msg,"burn_drive_obtain_scsi_adr( %s ) returns %d\n",
path, ret);
burn_drive_adr_debug_msg(msg, NULL);
return 0;
@ -1269,7 +1099,7 @@ int burn_abort_pacifier(void *handle, int patience, int elapsed)
}
/** Abort any running drive operation and finish libburn.
/** Abort any running drive operation and finis libburn.
@param patience Maximum number of seconds to wait for drives to finish
@param pacifier_func Function to produce appeasing messages. See
burn_abort_pacifier() for an example.
@ -1346,16 +1176,6 @@ int burn_disc_pretend_blank(struct burn_drive *d)
return 1;
}
/* ts A61106 API function */
int burn_disc_pretend_full(struct burn_drive *d)
{
if (d->status != BURN_DISC_UNREADY &&
d->status != BURN_DISC_UNSUITABLE)
return 0;
d->status = BURN_DISC_FULL;
return 1;
}
/* ts A61021: new API function */
int burn_disc_read_atip(struct burn_drive *d)
{
@ -1367,165 +1187,8 @@ int burn_disc_read_atip(struct burn_drive *d)
0, 0);
return -1;
}
if (d->current_profile == -1 || d->current_is_cd_profile) {
d->read_atip(d);
/* >>> some control of success would be nice :) */
} else {
/* mmc5r03c.pdf 6.26.3.6.3 : ATIP is undefined for non-CD */;
return 0;
}
d->read_atip(d);
/* >>> some control of success would be nice :) */
return 1;
}
/* ts A61110 : new API function */
int burn_disc_track_lba_nwa(struct burn_drive *d, struct burn_write_opts *o,
int trackno, int *lba, int *nwa)
{
int ret;
if (burn_drive_is_released(d)) {
libdax_msgs_submit(libdax_messenger,
d->global_index, 0x0002011b,
LIBDAX_MSGS_SEV_FATAL, LIBDAX_MSGS_PRIO_HIGH,
"Attempt to read track info from ungrabbed drive",
0, 0);
return -1;
}
if (d->busy != BURN_DRIVE_IDLE) {
libdax_msgs_submit(libdax_messenger,
d->global_index, 0x0002011c,
LIBDAX_MSGS_SEV_FATAL, LIBDAX_MSGS_PRIO_HIGH,
"Attempt to read track info from busy drive",
0, 0);
return -1;
}
if (o!=NULL)
d->send_write_parameters(d, o);
ret = d->get_nwa(d, trackno, lba, nwa);
return ret;
}
/* ts A61202 : New API function */
int burn_disc_get_profile(struct burn_drive *d, int *pno, char name[80])
{
*pno = d->current_profile;
strcpy(name,d->current_profile_text);
return *pno >= 0;
}
/* ts A61223 : New API function */
int burn_drive_wrote_well(struct burn_drive *d)
{
return !d->cancel;
}
/* ts A61226 */
int burn_speed_descriptor_new(struct burn_speed_descriptor **s,
struct burn_speed_descriptor *prev,
struct burn_speed_descriptor *next, int flag)
{
struct burn_speed_descriptor *o;
(*s) = o = malloc(sizeof(struct burn_speed_descriptor));
if (o == NULL)
return -1;
o->source = 0;
o->profile_loaded = -2;
o->profile_name[0] = 0;
o->wrc = 0;
o->exact = 0;
o->mrw = 0;
o->end_lba = -1;
o->write_speed = 0;
o->read_speed = 0;
o->prev = prev;
if (prev != NULL) {
next = prev->next;
prev->next = o;
}
o->next = next;
if (next != NULL)
next->prev = o;
return 1;
}
/* ts A61226 */
/* @param flag bit0= destroy whole next-chain of descriptors */
int burn_speed_descriptor_destroy(struct burn_speed_descriptor **s, int flag)
{
struct burn_speed_descriptor *next, *o;
if ((*s) == NULL)
return 0;
if (flag&1)
for (o = (*s); o->prev != NULL; o = o->prev);
else
o = (*s);
next = o->next;
if (next != NULL)
next->prev = o->prev;
if (o->prev != NULL)
o->prev->next = next;
free((char *) (*s));
(*s) = NULL;
if (flag&1)
return burn_speed_descriptor_destroy(&next, flag&1);
return 1;
}
/* ts A61226 */
int burn_speed_descriptor_copy(struct burn_speed_descriptor *from,
struct burn_speed_descriptor *to, int flag)
{
to->source = from->source;
to->profile_loaded = from->profile_loaded;
strcpy(to->profile_name, from->profile_name);
to->wrc = from->wrc;
to->exact = from->exact;
to->mrw = from->mrw;
to->end_lba = from->end_lba;
to->write_speed = from->write_speed;
to->read_speed = from->read_speed;
return 1;
}
/* ts A61226 : free dynamically allocated sub data of struct scsi_mode_data */
int burn_mdata_free_subs(struct scsi_mode_data *m)
{
burn_speed_descriptor_destroy(&(m->speed_descriptors), 1);
return 1;
}
/* ts A61226 : API function */
int burn_drive_get_speedlist(struct burn_drive *d,
struct burn_speed_descriptor **speed_list)
{
int ret;
struct burn_speed_descriptor *sd, *csd = NULL;
(*speed_list) = NULL;
for (sd = d->mdata->speed_descriptors; sd != NULL; sd = sd->next) {
ret = burn_speed_descriptor_new(&csd, NULL, csd, 0);
if (ret <= 0)
return -1;
burn_speed_descriptor_copy(sd, csd, 0);
}
(*speed_list) = csd;
return (csd != NULL);
}
/* ts A61226 : API function */
int burn_drive_free_speedlist(struct burn_speed_descriptor **speed_list)
{
return burn_speed_descriptor_destroy(speed_list, 1);
}

View File

@ -10,8 +10,6 @@
struct burn_drive;
struct command;
struct mempage;
struct scsi_mode_data;
struct burn_speed_descriptor;
#define LEAD_IN 1
#define GAP 2
@ -71,28 +69,5 @@ int burn_setup_drive(struct burn_drive *d, char *fname);
*/
struct burn_drive *burn_drive_finish_enum(struct burn_drive *d);
/* ts A61125 : media status aspects of burn_drive_grab() */
int burn_drive_inquire_media(struct burn_drive *d);
/* ts A61125 : model aspects of burn_drive_release */
int burn_drive_mark_unready(struct burn_drive *d);
/* ts A61226 */
int burn_speed_descriptor_new(struct burn_speed_descriptor **s,
struct burn_speed_descriptor *prev,
struct burn_speed_descriptor *next, int flag);
/* ts A61226 */
/* @param flag bit0= destroy whole next-chain of descriptors */
int burn_speed_descriptor_destroy(struct burn_speed_descriptor **s, int flag);
/* ts A61226 : free dynamically allocated sub data of struct scsi_mode_data */
int burn_mdata_free_subs(struct scsi_mode_data *m);
/* ts A61230 */
void burn_disc_format_sync(struct burn_drive *d, off_t size, int flag);
#endif /* __DRIVE */

View File

@ -73,6 +73,9 @@ static off_t file_size(struct burn_source *source)
if (fstat(fs->datafd, &buf) == -1)
return (off_t) 0;
/* for now we keep it compatible to the old (int) return value */
if(buf.st_size >= 1308622848) /* 2 GB - 800 MB to prevent rollover */
return (off_t) 1308622848;
return (off_t) buf.st_size;
}
@ -123,6 +126,9 @@ static off_t fd_get_size(struct burn_source *source)
return fs->fixed_size;
if (fstat(fs->datafd, &buf) == -1)
return (off_t) 0;
/* for now we keep it compatible to the old (int) return value */
if (buf.st_size >= 1308622848) /* 2 GB - 800 MB to prevent rollover */
return (off_t) 1308622848;
return buf.st_size;
}

View File

@ -209,38 +209,8 @@ ex:
int burn_builtin_abort_handler(void *handle, int signum, int flag)
{
if(getpid() != abort_control_pid) {
#ifdef Not_yeT
pthread_t thread_id;
/* >>> need better handling of self-induced SIGs
like SIGSEGV or SIGFPE.
Like bonking the control thread if it did not show up
after a short while.
*/
/* >>> if this is a non-fatal signal : return -2 */
thread_id = pthread_self();
/* >>> find thread_id in worker list of async.c */
/* >>> if owning a drive : mark idle and canceled
(can't do anything more) */
usleep(1000000); /* calm down */
/* forward signal to control thread */
if (abort_control_pid>1)
kill(abort_control_pid, signum);
/* >>> ??? end thread */;
#else
usleep(1000000); /* calm down */
if(getpid() != abort_control_pid)
return -2;
#endif /* ! Not_yeT */
}
Cleanup_set_handlers(NULL, NULL, 2);
fprintf(stderr,"%sABORT : Trying to shut down drive and library\n",
abort_message_prefix);

View File

@ -7,7 +7,6 @@
expressing a file or stream size.
XXX we should enforce 64-bitness for off_t
ts A61101 : this is usually done by the build system (if it is not broken)
*/
#include <sys/types.h>
@ -41,10 +40,6 @@ struct burn_session;
/** References a single track on a disc */
struct burn_track;
/* ts A61111 */
/** References a set of write parameters */
struct burn_write_opts;
/** Session format for normal audio or data discs */
#define BURN_CDROM 0
/** Session format for obsolete CD-I discs */
@ -221,20 +216,7 @@ enum burn_drive_status
/** The drive is erasing a disc */
BURN_DRIVE_ERASING,
/** The drive is being grabbed */
BURN_DRIVE_GRABBING,
/* ts A61102 */
/** The drive gets written zeroes before the track payload data */
BURN_DRIVE_WRITING_PREGAP,
/** The drive is told to close a track (TAO only) */
BURN_DRIVE_CLOSING_TRACK,
/** The drive is told to close a session (TAO only) */
BURN_DRIVE_CLOSING_SESSION,
/* ts A61223 */
/** The drive is formatting media */
BURN_DRIVE_FORMATTING
BURN_DRIVE_GRABBING
};
/** Information about a track on a disc - this is from the q sub channel of the
@ -392,11 +374,9 @@ struct burn_progress {
int index;
/** The starting logical block address */
int start_sector;
/** On write: The number of sectors.
On blank: 0x10000 as upper limit for relative progress steps */
/** The number of sector */
int sectors;
/** On write: The current sector being processed.
On blank: Relative progress steps 0 to 0x10000 */
/** The current sector being processed */
int sector;
/* ts A61023 */
@ -404,65 +384,8 @@ struct burn_progress {
unsigned buffer_capacity;
/** The free space in the drive buffer (might be slightly outdated) */
unsigned buffer_available;
/* ts A61119 */
/** The number of bytes sent to the drive buffer */
off_t buffered_bytes;
/** The minimum number of buffered bytes. (Caution: Before surely
one buffer size of bytes was processed, this value is 0xffffffff.)
*/
unsigned buffer_min_fill;
};
/* ts A61226 */
/** Description of a speed capability as reported by the drive in conjunction
with eventually loaded media. There can be more than one such object per
drive. So they are chained via .next and .prev , where NULL marks the end
of the chain. This list is set up by burn_drive_scan() and gets updated
by burn_drive_grab().
A copy may be obtained by burn_drive_get_speedlist() and disposed by
burn_drive_free_speedlist().
For technical background info see SCSI specs MMC and SPC:
mode page 2Ah (from SPC 5Ah MODE SENSE) , mmc3r10g.pdf , 6.3.11 Table 364
ACh GET PERFORMANCE, Type 03h , mmc5r03c.pdf , 6.8.5.3 Table 312
*/
struct burn_speed_descriptor {
/** Where this info comes from :
0 = misc , 1 = mode page 2Ah , 2 = ACh GET PERFORMANCE */
int source;
/** The media type that was current at the time of report
-2 = state unknown, -1 = no media was loaded , else see
burn_disc_get_profile() */
int profile_loaded;
char profile_name[80];
/** The attributed capacity of appropriate media in logical block units
i.e. 2352 raw bytes or 2048 data bytes. -1 = capacity unknown. */
int end_lba;
/** Speed is given in 1000 bytes/s , 0 = invalid. The numbers
are supposed to be usable with burn_drive_set_speed() */
int write_speed;
int read_speed;
/** Expert info from ACh GET PERFORMANCE and/or mode page 2Ah.
Expect values other than 0 or 1 to get a meaning in future.*/
/* Rotational control: 0 = CLV/default , 1 = CAV */
int wrc;
/* 1 = drive promises reported performance over full media */
int exact;
/* 1 = suitable for mixture of read and write */
int mrw;
/** List chaining. Use .next until NULL to iterate over the list */
struct burn_speed_descriptor *prev;
struct burn_speed_descriptor *next;
};
/** Initialize the library.
This must be called before using any other functions in the library. It
may be called more than once with no effect.
@ -494,7 +417,6 @@ void burn_finish(void);
@param patience Maximum number of seconds to wait for drives to finish
@param pacifier_func If not NULL: a function to produce appeasing messages.
See burn_abort_pacifier() for an example.
@param handle Opaque handle to be used with pacifier_func
@return 1 ok, all went well
0 had to leave a drive in unclean state
<0 severe error, do no use libburn again
@ -669,11 +591,6 @@ int burn_drive_convert_fs_adr(char *path, char adr[]);
then it is not decisive and the first enumerated address which matches
the >= 0 parameters is taken as result.
Note: bus and (host,channel) are supposed to be redundant.
@param bus_no "Bus Number" (something like a virtual controller)
@param host_no "Host Number" (something like half a virtual controller)
@param channel_no "Channel Number" (other half of "Host Number")
@param target_no "Target Number" or "SCSI Id" (a device)
@param lun_no "Logical Unit Number" (a sub device)
@param adr An application provided array of at least BURN_DRIVE_ADR_LEN
characters size. The persistent address gets copied to it.
@return 1 = success , 0 = failure , -1 = severe error
@ -712,7 +629,7 @@ void burn_drive_release(struct burn_drive *drive, int eject);
/** Returns what kind of disc a drive is holding. This function may need to be
called more than once to get a proper status from it. See burn_disc_status
called more than once to get a proper status from it. See burn_status
for details.
@param drive The drive to query for a disc.
@return The status of the drive, or what kind of disc is in it.
@ -731,15 +648,6 @@ enum burn_disc_status burn_disc_get_status(struct burn_drive *drive);
int burn_disc_pretend_blank(struct burn_drive *drive);
/* ts A61106 */
/** WARNING: This overrides the safety measures against unsuitable media.
Sets the drive status to BURN_DISC_FULL if it is BURN_DISC_UNREADY
or BURN_DISC_UNSUITABLE. Thus marking media as blankable which actually
failed to declare themselves either blank or (partially) filled.
*/
int burn_disc_pretend_full(struct burn_drive *drive);
/* ts A61021 */
/** Reads ATIP information from inserted media. To be obtained via
burn_drive_get_write_speed(), burn_drive_get_min_write_speed(),
@ -764,34 +672,6 @@ int burn_disc_read_atip(struct burn_drive *drive);
int burn_drive_get_start_end_lba(struct burn_drive *drive,
int *start_lba, int *end_lba, int flag);
/* ts A61110 */
/** Read start lba and Next Writeable Address of a track from media.
Usually a track lba is obtained from the result of burn_track_get_entry().
This call retrieves an updated lba, eventual nwa, and can address the
invisible track to come.
The drive must be grabbed for this call. One may not issue this call
during ongoing burn_disc_write() or burn_disc_erase().
@param d The drive to query.
@param o If not NULL: write parameters to be set on drive before query
@param trackno 0=next track to come, >0 number of existing track
@param lba return value: start lba
@param nwa return value: Next Writeable Address
@return 1=nwa is valid , 0=nwa is not valid , -1=error
*/
int burn_disc_track_lba_nwa(struct burn_drive *d, struct burn_write_opts *o,
int trackno, int *lba, int *nwa);
/* ts A61202 */
/** Tells the MMC Profile identifier of the loaded media. The drive must be
grabbed in order to get a non-zero result.
libburn currently writes only to profiles 0x09 "CD-R", 0x0a "CD-RW",
0x12 "DVD-RAM", 0x13 "DVD-RW restricted overwrite" or 0x1a "DVD+RW".
@param d The drive where the media is inserted.
@param pno Profile Number as of mmc5r03c.pdf, table 89
@param name Profile Name (e.g "CD-RW", unknown profiles have empty name)
@return 1 profile is valid, 0 no profile info available
*/
int burn_disc_get_profile(struct burn_drive *d, int *pno, char name[80]);
/** Tells whether a disc can be erased or not
@return Non-zero means erasable
@ -819,7 +699,7 @@ struct burn_write_opts *burn_write_opts_new(struct burn_drive *drive);
void burn_write_opts_free(struct burn_write_opts *opts);
/** Creates a read_opts struct for reading from the specified drive
must be freed with burn_read_opts_free
must be freed with burn_write_opts_free
@param drive The drive to read from
@return The read_opts
*/
@ -841,79 +721,6 @@ void burn_read_opts_free(struct burn_read_opts *opts);
*/
void burn_disc_erase(struct burn_drive *drive, int fast);
/* ts A70101 - A70112 */
/** Format media for use with libburn. This currently applies to DVD-RW
in state "Sequential Recording" (profile 0014h) which get formatted to
state "Restricted Overwrite" (profile 0013h). DVD+RW can be "de-iced"
by setting bit2 of flag. Other media cannot be formatted yet.
@param drive The drive with the disc to format.
@param size The size in bytes to be used with the format command. It should
be divisible by 32*1024. The effect of this parameter may
depend on the media profile.
@param flag Bitfield for control purposes:
bit0= after formatting, write the given number of zero-bytes
to the media and eventually perform preliminary closing.
bit1= insist in size 0 even if there is a better default known
bit2= format to maximum available size
bit3= -reserved-
bit4= enforce re-format of (partly) formatted media
bit7= MMC expert application mode (else libburn tries to
choose a suitable format type):
bit8 to bit15 contain the index of the format to use. See
burn_disc_get_formats(), burn_disc_get_format_descr().
Acceptable types are: 0x00, 0x10, 0x11, 0x13, 0x15, 0x26.
If bit7 is set, bit4 is set automatically.
*/
void burn_disc_format(struct burn_drive *drive, off_t size, int flag);
/* ts A70112 */
/** Possible formatting status values */
#define BURN_FORMAT_IS_UNFORMATTED 1
#define BURN_FORMAT_IS_FORMATTED 2
#define BURN_FORMAT_IS_UNKNOWN 3
/** Inquire the formatting status, the associated sizes and the number of
available formats. The info is media specific and stems from MMC command
23h READ FORMAT CAPACITY. See mmc5r03c.pdf 6.24 for background details.
Media type can be determined via burn_disc_get_profile().
@param drive The drive with the disc to format.
@param status The current formatting status of the inserted media.
See BURN_FORMAT_IS_* macros. Note: "unknown" is the
legal status for quick formatted, yet unwritten DVD-RW.
@param size The size in bytes associated with status.
unformatted: the maximum achievable size of the media
formatted: the currently formatted capacity
unknown: maximum capacity of drive or of media
@param bl_sas Additional info "Block Length/Spare Area Size".
Expected to be constantly 2048 for non-BD media.
@param num_formats The number of available formats. To be used with
burn_disc_get_format_descr() to obtain such a format
and eventually with burn_disc_format() to select one.
@return 1 reply is valid , <=0 failure
*/
int burn_disc_get_formats(struct burn_drive *drive, int *status, off_t *size,
unsigned *bl_sas, int *num_formats);
/** Inquire parameters of an available media format.
@param drive The drive with the disc to format.
@param index The index of the format item. Beginning with 0 up to reply
parameter from burn_disc_get_formats() : num_formats - 1
@param type The format type. See mmc5r03c.pdf, 6.5, 04h FORMAT UNIT.
0x00=full, 0x10=CD-RW/DVD-RW full, 0x11=CD-RW/DVD-RW grow,
0x15=DVD-RW quick, 0x13=DVD-RW quick grow,
0x26=DVD+RW background
@param size The maximum size in bytes achievable with this format.
@param tdp Type Dependent Parameter. See mmc5r03c.pdf.
@return 1 reply is valid , <=0 failure
*/
int burn_disc_get_format_descr(struct burn_drive *drive, int index,
int *type, off_t *size, unsigned *tdp);
/* ts A61109 : this was and is defunct */
/** Read a disc from the drive and write it to an fd pair. The drive must be
grabbed successfully BEFORE calling this function. Always ensure that the
drive reports a status of BURN_DISC_FULL before calling this function.
@ -922,9 +729,10 @@ int burn_disc_get_format_descr(struct burn_drive *drive, int index,
*/
void burn_disc_read(struct burn_drive *drive, const struct burn_read_opts *o);
/** Write a disc in the drive. The drive must be grabbed successfully before
/** Write a disc in the drive. The drive must be grabbed successfully BEFORE
calling this function. Always ensure that the drive reports a status of
BURN_DISC_BLANK before calling this function.
BURN_DISC_BLANK or BURN_STATUS_FULL (to append a new session to the
disc) before calling this function.
@param o The options for the writing operation.
@param disc The struct burn_disc * that described the disc to be created
*/
@ -937,17 +745,6 @@ void burn_disc_write(struct burn_write_opts *o, struct burn_disc *disc);
*/
void burn_drive_cancel(struct burn_drive *drive);
/* ts A61223 */
/** Inquire wether the most recent write run was successful. Reasons for
non-success may be: rejection of burn parameters, abort during fatal errors
during write, a call to burn_drive_cancel() by the application thread.
@param d The drive to inquire.
@return 1=burn seems to have went well, 0=burn failed
*/
int burn_drive_wrote_well(struct burn_drive *d);
/** Convert a minute-second-frame (MSF) value to sector count
@param m Minute component
@param s Second component
@ -1120,19 +917,10 @@ struct burn_source *burn_file_source_new(const char *path,
*/
struct burn_source *burn_fd_source_new(int datafd, int subfd, off_t size);
/** Tells how long a track will be on disc
>>> NOTE: Not reliable with tracks of undefined length
*/
/** Tells how long a track will be on disc */
int burn_track_get_sectors(struct burn_track *);
/* ts A61101 */
/** Tells how many source bytes have been read and how many data bytes have
been written by the track during burn */
int burn_track_get_counters(struct burn_track *t,
off_t *read_bytes, off_t *written_bytes);
/** Sets drive read and write speed
@param d The drive to set speed for
@param read Read speed in k/s (0 is max)
@ -1197,28 +985,6 @@ void burn_write_opts_set_has_mediacatalog(struct burn_write_opts *opts, int has_
void burn_write_opts_set_mediacatalog(struct burn_write_opts *opts, unsigned char mediacatalog[13]);
/* ts A61106 */
/** Sets the multi flag which eventually marks the emerging session as not
being the last one and thus creating a BURN_DISC_APPENDABLE media.
@param multi 1=media will be appendable, 0=media will be closed (default)
*/
void burn_write_opts_set_multi(struct burn_write_opts *opts, int multi);
/* ts A61222 */
/** Sets a start address for writing to media and write modes which allow to
choose this address at all (DVD+RW, DVD-RAM, DVD-RW only for now). The
address is given in bytes. If it is not -1 then a write run will fail if
choice of start address is not supported or if the block alignment of the
address is not suitable for media and write mode. (Alignment to 32 kB
blocks is advised with DVD media.)
@param opts The write opts to change
@param value The address in bytes (-1 = start at default address)
*/
void burn_write_opts_set_start_byte(struct burn_write_opts *opts, off_t value);
/** Sets whether to read in raw mode or not
@param opts The read opts to change
@param raw_mode If non-zero, reading will be done in raw mode, so that everything in the data tracks on the
@ -1274,9 +1040,7 @@ void burn_read_opts_transfer_damaged_blocks(struct burn_read_opts *opts,
void burn_read_opts_set_hardware_error_retries(struct burn_read_opts *opts,
unsigned char hardware_error_retries);
/** Gets the maximum write speed for a drive and eventually loaded media.
The return value might change by the media type of already loaded media,
again by call burn_drive_grab() and again by call burn_disc_read_atip().
/** Gets the maximum write speed for a drive
@param d Drive to query
@return Maximum write speed in K/s
*/
@ -1284,9 +1048,8 @@ int burn_drive_get_write_speed(struct burn_drive *d);
/* ts A61021 */
/** Gets the minimum write speed for a drive and eventually loaded media.
The return value might change by the media type of already loaded media,
again by call burn_drive_grab() and again by call burn_disc_read_atip().
/** Gets the minimum write speed for a drive. This might differ from
burn_drive_get_write_speed() only after burn_disc_read_atip()
@param d Drive to query
@return Minimum write speed in K/s
*/
@ -1299,31 +1062,6 @@ int burn_drive_get_min_write_speed(struct burn_drive *d);
*/
int burn_drive_get_read_speed(struct burn_drive *d);
/* ts A61226 */
/** Obtain a copy of the current speed descriptor list. The drive's list gets
updated on various occasions such as burn_drive_grab() but the copy
obtained here stays untouched. It has to be disposed via
burn_drive_free_speedlist() when it is not longer needed. Speeds
may appear several times in the list. The list content depends much on
drive and media type. It seems that .source == 1 applies mostly to CD media
whereas .source == 2 applies to any media.
@param d Drive to query
@param speed_list The copy. If empty, *speed_list gets returned as NULL.
@return 1=success , 0=list empty , <0 severe error
*/
int burn_drive_get_speedlist(struct burn_drive *d,
struct burn_speed_descriptor **speed_list);
/* ts A61226 */
/** Dispose a speed descriptor list copy which was obtained by
burn_drive_get_speedlist().
@param speed_list The list copy. *speed_list gets set to NULL.
@return 1=list disposed , 0= *speedlist was already NULL
*/
int burn_drive_free_speedlist(struct burn_speed_descriptor **speed_list);
/** Gets a copy of the toc_entry structure associated with a track
@param t Track to get the entry from
@param entry Struct for the library to fill out
@ -1401,10 +1139,9 @@ int burn_msgs_set_severities(char *queue_severity,
/** Obtain the oldest pending libburn message from the queue which has at
least the given minimum_severity. This message and any older message of
lower severity will get discarded from the queue and is then lost forever.
@param minimum_severity may be one of "NEVER", "FATAL", "SORRY",
"WARNING", "HINT", "NOTE", "UPDATE", "DEBUG", "ALL".
To call with minimum_severity "NEVER" will discard the
whole queue.
Severity may be one of "NEVER", "FATAL", "SORRY", "WARNING", "HINT",
"NOTE", "UPDATE", "DEBUG", "ALL". To call with minimum_severity "NEVER"
will discard the whole queue.
@param error_code Will become a unique error code as liste in
libburn/libdax_msgs.h
@param msg_text Must provide at least BURN_MSGS_MESSAGE_LEN bytes.

View File

@ -303,7 +303,7 @@ Range "scdbackup" : 0x00020000 to 0x0002ffff
0x00020006 (FATAL,HIGH) = Too many scsi siblings
0x00020007 (NOTE,HIGH) = Closed O_EXCL scsi siblings
General library operations:
From the hunt on Assert:
0x00020101 (WARNING,HIGH) = Cannot find given worker item
0x00020102 (SORRY,HIGH) = A drive operation is still going on
@ -320,7 +320,7 @@ Range "scdbackup" : 0x00020000 to 0x0002ffff
0x0002010c (FATAL,HIGH) = Failed to transfer command to drive
0x0002010d (DEBUG,HIGH) = Could not inquire TOC
0x0002010e (FATAL,HIGH) = Attempt to read ATIP from ungrabbed drive
0x0002010f (DEBUG,HIGH) = SCSI error condition on command
0x0002010f
0x00020110 (FATAL,HIGH) = Persistent drive address too long
0x00020111 (FATAL,HIGH) = Could not allocate new auxiliary object
0x00020112 (SORRY,HIGH) = Bad combination of write_type and block_type
@ -329,27 +329,6 @@ Range "scdbackup" : 0x00020000 to 0x0002ffff
0x00020115 (SORRY,HIGH) = Attempt to set track mode to unusable value
0x00020116 (FATAL,HIGH) = Track mode has unusable value
0x00020117 (FATAL,HIGH) = toc_entry of drive is already in use
0x00020118 (DEBUG,HIGH) = Closing track
0x00020119 (DEBUG,HIGH) = Closing session
0x0002011a (NOTE,HIGH) = Padding up track to minimum size
0x0002011b (FATAL,HIGH) = Attempt to read track info from ungrabbed drive
0x0002011c (FATAL,HIGH) = Attempt to read track info from busy drive
0x0002011d (FATAL,HIGH) = SCSI error on write
0x0002011e (SORRY,HIGH) = Unsuitable media detected
0x0002011f (SORRY,HIGH) = Burning is restricted to a single track
0x00020120 (NOTE,HIGH) = FORMAT UNIT ignored
0x00020121 (FATAL,HIGH) = Write preparation setup failed
0x00020122 (FATAL,HIGH) = SCSI error on format_unit
0x00020123 (SORRY,HIGH) = DVD Media are unsuitable for desired track type
0x00020124 (SORRY,HIGH) = SCSI error on set_streaming
0x00020125 (SORRY,HIGH) = Write start address not supported
0x00020126 (SORRY,HIGH) = Write start address not properly aligned
0x00020127 (NOTE,HIGH) = Write start address is ...
0x00020128 (FATAL,HIGH) = Unsupported inquiry_type with mmc_get_performance
0x00020129 (SORRY,HIGH) = Will not format media type
0x00020130 (SORRY,HIGH) = Drive and media state unsuitable for blanking
0x00020131 (SORRY,HIGH) = No suitable formatting type offered by drive
libdax_audioxtr:
0x00020200 (SORRY,HIGH) = Cannot open audio source file

File diff suppressed because it is too large Load Diff

View File

@ -37,11 +37,7 @@ void mmc_set_speed(struct burn_drive *, int, int);
void mmc_read_lead_in(struct burn_drive *, struct buffer *);
void mmc_perform_opc(struct burn_drive *);
void mmc_get_configuration(struct burn_drive *);
/* ts A61110 : added parameters trackno, lba, nwa. Redefined return value.
@return 1=nwa is valid , 0=nwa is not valid , -1=error */
int mmc_get_nwa(struct burn_drive *d, int trackno, int *lba, int *nwa);
int mmc_get_nwa(struct burn_drive *);
void mmc_send_cue_sheet(struct burn_drive *, struct cue_sheet *);
/* ts A61023 : get size and free space of drive buffer */
@ -51,22 +47,4 @@ int mmc_read_buffer_capacity(struct burn_drive *d);
*/
int mmc_setup_drive(struct burn_drive *d);
/* ts A61219 : learned much from dvd+rw-tools-7.0: plus_rw_format()
and mmc5r03c.pdf, 6.5 FORMAT UNIT */
int mmc_format_unit(struct burn_drive *d, off_t size, int flag);
/* ts A61225 : obtain write speed descriptors via ACh GET PERFORMANCE */
int mmc_get_write_performance(struct burn_drive *d);
/* ts A61229 : outsourced from spc_select_write_params() */
/* Note: Page data is not zeroed here to allow preset defaults. Thus
memset(pd, 0, 2 + d->mdata->write_page_length);
is the eventual duty of the caller.
*/
int mmc_compose_mode_page_5(struct burn_drive *d,
const struct burn_write_opts *o,
unsigned char *pd);
#endif /*__MMC*/

View File

@ -32,9 +32,6 @@ struct burn_write_opts *burn_write_opts_new(struct burn_drive *drive)
opts->simulate = 0;
opts->underrun_proof = drive->mdata->underrun_proof;
opts->perform_opc = 1;
opts->obs = -1;
opts->obs_pad = 0;
opts->start_byte = -1;
opts->has_mediacatalog = 0;
opts->format = BURN_CDROM;
opts->multi = 0;
@ -155,21 +152,6 @@ void burn_write_opts_set_mediacatalog(struct burn_write_opts *opts,
memcpy(opts->mediacatalog, &mediacatalog, 13);
}
/* ts A61106 */
void burn_write_opts_set_multi(struct burn_write_opts *opts, int multi)
{
opts->multi = !!multi;
}
/* ts A61222 */
void burn_write_opts_set_start_byte(struct burn_write_opts *opts, off_t value)
{
opts->start_byte = value;
}
void burn_read_opts_set_raw(struct burn_read_opts *opts, int raw)
{
opts->raw = raw;
@ -216,4 +198,3 @@ void burn_read_opts_set_hardware_error_retries(struct burn_read_opts *opts,
{
opts->hardware_error_retries = hardware_error_retries;
}

View File

@ -32,15 +32,6 @@ struct burn_write_opts
/** Perform calibration of the drive's laser before beginning the
write. */
unsigned int perform_opc:1;
/* ts A61219 : Output block size to trigger buffer flush if hit.
-1 with CD, 32 kB with DVD */
int obs;
int obs_pad; /* 1=pad up last block to obs */
/* ts A61222 : Start address for media which allow a choice */
off_t start_byte;
/** A disc can have a media catalog number */
int has_mediacatalog;
unsigned char mediacatalog[13];

View File

@ -1,73 +0,0 @@
/* os-freebsd.h
Operating system specific libburn definitions and declarations. Included
by os.h in case of compilation for
FreeBSD with CAM
Copyright (C) 2006 Thomas Schmitt <scdbackup@gmx.net>, provided under GPL
*/
#ifndef BURN_OS_H_INCLUDED
#define BURN_OS_H_INCLUDED 1
/** List of all signals which shall be caught by signal handlers and trigger
a graceful abort of libburn. (See man 7 signal.)
*/
/* Once as system defined macros */
#define BURN_OS_SIGNAL_MACRO_LIST \
SIGHUP, SIGINT, SIGQUIT, SIGILL, SIGABRT, \
SIGFPE, SIGSEGV, SIGPIPE, SIGALRM, SIGTERM, \
SIGUSR1, SIGUSR2, SIGXCPU, SIGTSTP, SIGTTIN, \
SIGTTOU, \
SIGBUS, SIGPROF, SIGSYS, SIGTRAP, \
SIGVTALRM, SIGXCPU, SIGXFSZ
/* Once as text 1:1 list of strings for messages and interpreters */
#define BURN_OS_SIGNAL_NAME_LIST \
"SIGHUP", "SIGINT", "SIGQUIT", "SIGILL", "SIGABRT", \
"SIGFPE", "SIGSEGV", "SIGPIPE", "SIGALRM", "SIGTERM", \
"SIGUSR1", "SIGUSR2", "SIGXCPU", "SIGTSTP", "SIGTTIN", \
"SIGTTOU", \
"SIGBUS", "SIGPROF", "SIGSYS", "SIGTRAP", \
"SIGVTALRM", "SIGXCPU", "SIGXFSZ"
/* The number of above list items */
#define BURN_OS_SIGNAL_COUNT 23
/** To list all signals which shall surely not be caught */
#define BURN_OS_NON_SIGNAL_MACRO_LIST \
SIGKILL, SIGCHLD, SIGSTOP, SIGURG, SIGWINCH
/* The number of above list items */
#define BURN_OS_NON_SIGNAL_COUNT 5
/* The maximum size for a (SCSI) i/o transaction */
/* Important : MUST be at least 32768 ! */
#define BURN_OS_TRANSPORT_BUFFER_SIZE 32768
/** To hold all state information of BSD device enumeration
which are now local in sg_enumerate() . So that sg_give_next_adr()
can work in BSD and sg_enumerate() can use it.
*/
#define BURN_OS_DEFINE_DRIVE_ENUMERATOR_T \
struct burn_drive_enumeration_state { \
union ccb ccb; \
int bufsize, fd; \
unsigned int i; \
int skip_device; \
}; \
typedef struct burn_drive_enumeration_state burn_drive_enumerator_t;
/* The list of operating system dependent elements in struct burn_drive.
To be initialized and used within sg-*.c .
*/
#define BURN_OS_TRANSPORT_DRIVE_ELEMENTS \
struct cam_device* cam;
#endif /* ! BURN_OS_H_INCLUDED */

View File

@ -1,64 +0,0 @@
/* os-linux.h
Operating system specific libburn definitions and declarations. Included
by os.h in case of compilation for
Linux kernels 2.4 and 2.6 with Linux SCSI Generic (sg)
Copyright (C) 2006 Thomas Schmitt <scdbackup@gmx.net>, provided under GPL
*/
/** List of all signals which shall be caught by signal handlers and trigger
a graceful abort of libburn. (See man 7 signal.)
*/
/* Once as system defined macros */
#define BURN_OS_SIGNAL_MACRO_LIST \
SIGHUP, SIGINT, SIGQUIT, SIGILL, SIGABRT, \
SIGFPE, SIGSEGV, SIGPIPE, SIGALRM, SIGTERM, \
SIGUSR1, SIGUSR2, SIGXCPU, SIGTSTP, SIGTTIN, \
SIGTTOU, \
SIGBUS, SIGPOLL, SIGPROF, SIGSYS, SIGTRAP, \
SIGVTALRM, SIGXCPU, SIGXFSZ
/* Once as text 1:1 list of strings for messages and interpreters */
#define BURN_OS_SIGNAL_NAME_LIST \
"SIGHUP", "SIGINT", "SIGQUIT", "SIGILL", "SIGABRT", \
"SIGFPE", "SIGSEGV", "SIGPIPE", "SIGALRM", "SIGTERM", \
"SIGUSR1", "SIGUSR2", "SIGXCPU", "SIGTSTP", "SIGTTIN", \
"SIGTTOU", \
"SIGBUS", "SIGPOLL", "SIGPROF", "SIGSYS", "SIGTRAP", \
"SIGVTALRM", "SIGXCPU", "SIGXFSZ"
/* The number of above list items */
#define BURN_OS_SIGNAL_COUNT 24
/** To list all signals which shall surely not be caught */
#define BURN_OS_NON_SIGNAL_MACRO_LIST \
SIGKILL, SIGCHLD, SIGSTOP, SIGURG, SIGWINCH
/* The number of above list items */
#define BURN_OS_NON_SIGNAL_COUNT 5
/* The maximum size for a (SCSI) i/o transaction */
/* Important : MUST be at least 32768 ! */
#define BURN_OS_TRANSPORT_BUFFER_SIZE 65536
/* To hold the index number of the most recently delivered address from
device enumeration.
*/
#define BURN_OS_DEFINE_DRIVE_ENUMERATOR_T \
typedef int burn_drive_enumerator_t;
/* The list of operating system dependent elements in struct burn_drive.
Usually they are initialized in sg-*.c:enumerate_common().
*/
#define BURN_OS_TRANSPORT_DRIVE_ELEMENTS \
int fd; \
\
/* ts A60926 : trying to lock against growisofs /dev/srN, /dev/scdN */ \
int sibling_count; \
int sibling_fds[LIBBURN_SG_MAX_SIBLINGS];

View File

@ -1,34 +0,0 @@
/* os.h
Operating system specific libburn definitions and declarations.
The macros defined here are used by libburn modules in order to
avoid own system dependent case distinctions.
Copyright (C) 2006 Thomas Schmitt <scdbackup@gmx.net>, provided under GPL
*/
#ifndef BURN_OS_H_INCLUDED
#define BURN_OS_H_INCLUDED 1
/*
Operating system case distinction
*/
#ifdef __FreeBSD__
/* ----------------------------- FreeBSD with CAM -------------------------- */
#include "os-freebsd.h"
#else /* operating system case distinction */
/* --------- Linux kernels 2.4 and 2.6 with Linux SCSI Generic (sg) -------- */
#include "os-linux.h"
#endif /* End of operating system case distinction */
#endif /* ! BURN_OS_H_INCLUDED */

View File

@ -11,7 +11,6 @@
/* spc command set */
static char SBC_LOAD[] = { 0x1b, 0, 0, 0, 3, 0 };
static char SBC_UNLOAD[] = { 0x1b, 0, 0, 0, 2, 0 };
static char SBC_START_UNIT[] = { 0x1b, 0, 0, 0, 1, 0 };
void sbc_load(struct burn_drive *d)
{
@ -38,28 +37,12 @@ void sbc_eject(struct burn_drive *d)
d->issue_command(d, &c);
}
/* ts A61118 : is it necessary to tell the drive to get ready for use ? */
int sbc_start_unit(struct burn_drive *d)
{
struct command c;
memcpy(c.opcode, SBC_START_UNIT, sizeof(SBC_START_UNIT));
c.retry = 1;
c.oplen = sizeof(SBC_START_UNIT);
c.dir = NO_TRANSFER;
c.page = NULL;
d->issue_command(d, &c);
return (c.error==0);
}
/* ts A61021 : the sbc specific part of sg.c:enumerate_common()
*/
int sbc_setup_drive(struct burn_drive *d)
{
d->eject = sbc_eject;
d->load = sbc_load;
d->start_unit = sbc_start_unit;
return 1;
}

View File

@ -8,9 +8,6 @@ struct burn_drive;
void sbc_load(struct burn_drive *);
void sbc_eject(struct burn_drive *);
/* ts A61118 */
int sbc_start_unit(struct burn_drive *);
/* ts A61021 : the sbc specific part of sg.c:enumerate_common()
*/
int sbc_setup_drive(struct burn_drive *d);

View File

@ -19,15 +19,6 @@
#include "toc.h"
#include "write.h"
#ifdef Libburn_log_in_and_out_streaM
/* <<< ts A61031 */
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#endif /* Libburn_log_in_and_out_streaM */
/*static unsigned char isrc[] = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ";*/
#define sector_common(X) d->alba++; d->rlba X;
@ -53,8 +44,7 @@ static void uncook_subs(unsigned char *dest, unsigned char *source)
int sector_get_outmode(enum burn_write_types write_type,
enum burn_block_types block_type)
{
/* ts A61103 : extended SAO condition to TAO */
if (write_type == BURN_WRITE_SAO || write_type == BURN_WRITE_TAO)
if (write_type == BURN_WRITE_SAO)
return 0;
else
switch (block_type) {
@ -91,15 +81,6 @@ static void get_bytes(struct burn_track *track, int count, unsigned char *data)
{
int valid, shortage, curr, i, tr;
#ifdef Libburn_log_in_and_out_streaM
/* <<< ts A61031 */
static int tee_fd= -1;
if(tee_fd==-1)
tee_fd= open("/tmp/libburn_sg_readin",
O_WRONLY|O_CREAT|O_TRUNC,S_IRUSR|S_IWUSR);
#endif /* Libburn_log_in_and_out_streaM */
/* no track pointer means we're just generating 0s */
if (!track) {
memset(data, 0, count);
@ -126,18 +107,10 @@ static void get_bytes(struct burn_track *track, int count, unsigned char *data)
valid = track->source->read(track->source, data + curr, count - curr);
} else valid = 0;
if (valid <= 0) { /* ts A61031 : extended from (valid == -1) */
if (valid == -1) {
track->eos = 1;
valid = 0;
}
track->sourcecount += valid;
#ifdef Libburn_log_in_and_out_streaM
/* <<< ts A61031 */
if(tee_fd!=-1 && valid>0) {
write(tee_fd, data + curr, valid);
}
#endif /* Libburn_log_in_and_out_streaM */
curr += valid;
shortage = count - curr;
@ -161,12 +134,6 @@ static void get_bytes(struct burn_track *track, int count, unsigned char *data)
if (!shortage)
goto ex;
/* ts A61031 */
if (shortage >= count)
track->track_data_done = 1;
if (track->open_ended)
goto ex;
/* If we're still short, and there's a "next" pointer, we pull from that.
if that depletes, we'll just fill with 0s.
*/
@ -185,6 +152,15 @@ ex:;
if(shortage)
memset(data + curr, 0, shortage); /* this is old icculus.org */
if (track->swap_source_bytes == 1) {
/*
{ static int swapping_count= 0;
fprintf(stderr,"\rlibburn_debug: swapping #%d \r",
swapping_count);
swapping_count++;
}
*/
for (i = 1; i < count; i += 2) {
tr = data[i];
data[i] = data[i-1];
@ -195,9 +171,7 @@ ex:;
/* ts A61009 : seems to hand out sector start pointer in opts->drive->buffer
and to count hand outs as well as reserved bytes */
/* ts A61101 : added parameter track for counting written bytes */
static unsigned char *get_sector(struct burn_write_opts *opts,
struct burn_track *track, int inmode)
static unsigned char *get_sector(struct burn_write_opts *opts, int inmode)
{
struct burn_drive *d = opts->drive;
struct buffer *out = d->buffer;
@ -217,21 +191,11 @@ static unsigned char *get_sector(struct burn_write_opts *opts,
return NULL;
seclen += burn_subcode_length(outmode);
/* ts A61219 : opts->obs is eventually a 32k trigger for DVD */
if (out->bytes + (seclen) > BUFFER_SIZE || out->bytes == opts->obs) {
if (out->bytes + (seclen) >= BUFFER_SIZE) {
int err;
err = d->write(d, d->nwa, out);
if (err == BE_CANCELLED)
return NULL;
/* ts A61101 */
if(track != NULL) {
track->writecount += out->bytes;
track->written_sectors += out->sectors;
}
/* ts A61119 */
d->progress.buffered_bytes += out->bytes;
d->nwa += out->sectors;
out->bytes = 0;
out->sectors = 0;
@ -244,32 +208,6 @@ static unsigned char *get_sector(struct burn_write_opts *opts,
return ret;
}
/* ts A61031 */
/* Revoke the counting of the most recent sector handed out by get_sector() */
static void unget_sector(struct burn_write_opts *opts, int inmode)
{
struct burn_drive *d = opts->drive;
struct buffer *out = d->buffer;
int outmode;
int seclen;
outmode = get_outmode(opts);
if (outmode == 0)
outmode = inmode;
/* ts A61009 : react on eventual failure of burn_sector_length()
(should not happen if API tested properly).
Ensures out->bytes >= out->sectors */
seclen = burn_sector_length(outmode);
if (seclen <= 0)
return;
seclen += burn_subcode_length(outmode);
out->bytes -= seclen;
out->sectors--;
}
/* either inmode == outmode, or outmode == raw. anything else is bad news */
/* ts A61010 : changed type to int in order to propagate said bad news */
/** @return 1 is ok, <= 0 is failure */
@ -351,8 +289,6 @@ static void convert_subs(struct burn_write_opts *o, int inmode,
out[0] = ~out[0];
out[1] = ~out[1];
break;
/* ts A61119 : to silence compiler warnings */
default:;
}
}
@ -399,7 +335,7 @@ int sector_toc(struct burn_write_opts *o, int mode)
unsigned char *data;
unsigned char subs[96];
data = get_sector(o, NULL, mode);
data = get_sector(o, mode);
if (data == NULL)
return 0;
/* ts A61010 */
@ -419,7 +355,7 @@ int sector_pregap(struct burn_write_opts *o,
unsigned char *data;
unsigned char subs[96];
data = get_sector(o, NULL, mode);
data = get_sector(o, mode);
if (data == NULL)
return 0;
/* ts A61010 */
@ -439,7 +375,7 @@ int sector_postgap(struct burn_write_opts *o,
unsigned char subs[96];
unsigned char *data;
data = get_sector(o, NULL, mode);
data = get_sector(o, mode);
if (data == NULL)
return 0;
/* ts A61010 */
@ -611,7 +547,7 @@ int sector_lout(struct burn_write_opts *o, unsigned char control, int mode)
unsigned char subs[96];
unsigned char *data;
data = get_sector(o, NULL, mode);
data = get_sector(o, mode);
if (!data)
return 0;
/* ts A61010 */
@ -630,23 +566,14 @@ int sector_data(struct burn_write_opts *o, struct burn_track *t, int psub)
unsigned char subs[96];
unsigned char *data;
data = get_sector(o, t, t->mode);
data = get_sector(o, t->mode);
if (!data)
return 0;
/* ts A61010 */
if (convert_data(o, t, t->mode, data) <= 0)
return 0;
/* ts A61031 */
if (t->open_ended && t->track_data_done) {
unget_sector(o, t->mode);
return 2;
}
/* ts A61219 : allow track without .entry */
if (t->entry == NULL)
;
else if (!t->source->read_sub)
if (!t->source->read_sub)
subcode_user(o, subs, t->entry->point,
t->entry->control, 1, &t->isrc, psub);
else if (!t->source->read_sub(t->source, subs, 96))
@ -695,11 +622,6 @@ int sector_headers_is_ok(struct burn_write_opts *o, int mode)
return 1;
if (o->write_type == BURN_WRITE_SAO)
return 1;
/* ts A61031 */
if (o->write_type == BURN_WRITE_TAO)
return 1;
if (mode & BURN_MODE1)
return 2;
return 0;
@ -728,11 +650,6 @@ void sector_headers(struct burn_write_opts *o, unsigned char *out,
return;
if (o->write_type == BURN_WRITE_SAO)
return;
/* ts A61031 */
if (o->write_type == BURN_WRITE_TAO)
return;
if (mode & BURN_MODE1)
modebyte = 1;
@ -819,14 +736,7 @@ void process_q(struct burn_drive *d, unsigned char *q)
#endif
/* this needs more info. subs in the data? control/adr? */
/* ts A61119 : One should not use inofficial compiler extensions.
>>> Some day this function needs to be implemented. At least for now
the result does not match the "mode" of cdrecord -toc.
*/
/*
#warning sector_identify needs to be written
*/
int sector_identify(unsigned char *data)
{
scramble(data);

View File

@ -1,559 +0,0 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/*
This is the main operating system dependent SCSI part of libburn. It implements
the transport level aspects of SCSI control and command i/o.
Present implementation: FreeBSD CAM (untested)
PORTING:
Porting libburn typically will consist of adding a new operating system case
to the following switcher files:
os.h Operating system specific libburn definitions and declarations.
sg.c Operating system dependent transport level modules.
and of deriving the following system specific files from existing examples:
os-*.h Included by os.h. You will need some general system knowledge
about signals and knowledge about the storage object needs of your
transport level module sg-*.c.
sg-*.c This source module. You will need special system knowledge about
how to detect all potentially available drives, how to open them,
eventually how to exclusively reserve them, how to perform
SCSI transactions, how to inquire the (pseudo-)SCSI driver.
You will not need to care about CD burning, MMC or other high-level
SCSI aspects.
Said sg-*.c operations are defined by a public function interface, which has
to be implemented in a way that provides libburn with the desired services:
sg_give_next_adr() iterates over the set of potentially useful drive
address strings.
scsi_enumerate_drives() brings all available, not-whitelist-banned, and
accessible drives into libburn's list of drives.
sg_drive_is_open() tells wether libburn has the given drive in use.
sg_grab() opens the drive for SCSI commands and ensures
undisturbed access.
sg_release() closes a drive opened by sg_grab()
sg_issue_command() sends a SCSI command to the drive, receives reply,
and evaluates wether the command succeeded or shall
be retried or finally failed.
sg_obtain_scsi_adr() tries to obtain SCSI address parameters.
Porting hints are marked by the text "PORTING:".
Send feedback to libburn-hackers@pykix.org .
*/
/** PORTING : ------- OS dependent headers and definitions ------ */
#include <errno.h>
#include <unistd.h>
#include <stdio.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <sys/ioctl.h>
#include <stdlib.h>
#include <string.h>
#include <sys/poll.h>
#include <camlib.h>
#include <cam/scsi/scsi_message.h>
#include <cam/scsi/scsi_pass.h>
#include <err.h> /* XXX */
/** PORTING : ------ libburn portable headers and definitions ----- */
#include "transport.h"
#include "drive.h"
#include "sg.h"
#include "spc.h"
#include "mmc.h"
#include "sbc.h"
#include "debug.h"
#include "toc.h"
#include "util.h"
#include "libdax_msgs.h"
extern struct libdax_msgs *libdax_messenger;
/* is in portable part of libburn */
int burn_drive_is_banned(char *device_address);
/* ------------------------------------------------------------------------ */
/* ts A61115: Private functions. Port only if needed by public functions */
/* (Public functions are listed below) */
/* ------------------------------------------------------------------------ */
/* Helper function for scsi_give_next_adr() */
static int sg_init_enumerator(burn_drive_enumerator_t *idx)
{
idx->skip_device = 0;
if ((idx->fd = open(XPT_DEVICE, O_RDWR)) == -1) {
warn("couldn't open %s", XPT_DEVICE);
return -1;
}
bzero(&(idx->ccb), sizeof(union ccb));
idx->ccb.ccb_h.path_id = CAM_XPT_PATH_ID;
idx->ccb.ccb_h.target_id = CAM_TARGET_WILDCARD;
idx->ccb.ccb_h.target_lun = CAM_LUN_WILDCARD;
idx->ccb.ccb_h.func_code = XPT_DEV_MATCH;
idx->bufsize = sizeof(struct dev_match_result) * 100;
idx->ccb.cdm.match_buf_len = idx->bufsize;
idx->ccb.cdm.matches = (struct dev_match_result *)malloc(idx->bufsize);
if (idx->ccb.cdm.matches == NULL) {
warnx("can't malloc memory for matches");
close(idx->fd);
return -1;
}
idx->ccb.cdm.num_matches = 0;
idx->i = idx->ccb.cdm.num_matches; /* to trigger buffer load */
/*
* We fetch all nodes, since we display most of them in the default
* case, and all in the verbose case.
*/
idx->ccb.cdm.num_patterns = 0;
idx->ccb.cdm.pattern_buf_len = 0;
return 1;
}
/* Helper function for scsi_give_next_adr() */
static int sg_next_enumeration_buffer(burn_drive_enumerator_t *idx)
{
/*
* We do the ioctl multiple times if necessary, in case there are
* more than 100 nodes in the EDT.
*/
if (ioctl(idx->fd, CAMIOCOMMAND, &(idx->ccb)) == -1) {
warn("error sending CAMIOCOMMAND ioctl");
return -1;
}
if ((idx->ccb.ccb_h.status != CAM_REQ_CMP)
|| ((idx->ccb.cdm.status != CAM_DEV_MATCH_LAST)
&& (idx->ccb.cdm.status != CAM_DEV_MATCH_MORE))) {
warnx("got CAM error %#x, CDM error %d\n",
idx->ccb.ccb_h.status, idx->ccb.cdm.status);
return -1;
}
return 1;
}
static int sg_close_drive(struct burn_drive * d)
{
if (d->cam != NULL) {
cam_close_device(d->cam);
d->cam = NULL;
}
return 0;
}
/* ----------------------------------------------------------------------- */
/* PORTING: Private functions which contain publicly needed functionality. */
/* Their portable part must be performed. So it is probably best */
/* to replace the non-portable part and to call these functions */
/* in your port, too. */
/* ----------------------------------------------------------------------- */
/** Wraps a detected drive into libburn structures and hands it over to
libburn drive list.
*/
static void enumerate_common(char *fname, int bus_no, int host_no,
int channel_no, int target_no, int lun_no)
{
int ret;
struct burn_drive out;
/* General libburn drive setup */
burn_setup_drive(&out, fname);
/* This transport adapter uses SCSI-family commands and models
(seems the adapter would know better than its boss, if ever) */
ret = burn_scsi_setup_drive(&out, bus_no, host_no, channel_no,
target_no, lun_no, 0);
if (ret<=0)
return;
/* PORTING: ------------------- non portable part --------------- */
/* Operating system adapter is CAM */
/* Adapter specific handles and data */
out.cam = NULL;
/* PORTING: ---------------- end of non portable part ------------ */
/* Adapter specific functions with standardized names */
out.grab = sg_grab;
out.release = sg_release;
out.drive_is_open = sg_drive_is_open;
out.issue_command = sg_issue_command;
/* Finally register drive and inquire drive information */
burn_drive_finish_enum(&out);
}
/* ts A61115 */
/* ------------------------------------------------------------------------ */
/* PORTING: Public functions. These MUST be ported. */
/* ------------------------------------------------------------------------ */
/** Returns the next index number and the next enumerated drive address.
The enumeration has to cover all available and accessible drives. It is
allowed to return addresses of drives which are not available but under
some (even exotic) circumstances could be available. It is on the other
hand allowed, only to hand out addresses which can really be used right
in the moment of this call. (This implementation chooses the latter.)
@param idx An opaque handle. Make no own theories about it.
@param adr Takes the reply
@param adr_size Gives maximum size of reply including final 0
@param initialize 1 = start new,
0 = continue, use no other values for now
-1 = finish
@return 1 = reply is a valid address , 0 = no further address available
-1 = severe error (e.g. adr_size too small)
*/
int sg_give_next_adr(burn_drive_enumerator_t *idx,
char adr[], int adr_size, int initialize)
{
int ret;
if (initialize == 1) {
ret = sg_init_enumerator(idx);
if (ret<=0)
return ret;
} else if (initialize == -1) {
if(idx->fd != -1)
close(idx->fd);
idx->fd = -1;
return 0;
}
try_item:; /* This spaghetti loop keeps the number of tabs small */
/* Loop content from old scsi_enumerate_drives() */
while (idx->i >= idx->ccb.cdm.num_matches) {
ret = sg_next_enumeration_buffer(idx);
if (ret<=0)
return -1;
if (!((idx->ccb.ccb_h.status == CAM_REQ_CMP)
&& (idx->ccb.cdm.status == CAM_DEV_MATCH_MORE)) )
return 0;
idx->i = 0;
}
switch (idx->ccb.cdm.matches[idx->i].type) {
case DEV_MATCH_BUS:
break;
case DEV_MATCH_DEVICE: {
struct device_match_result* result;
result = &(idx->ccb.cdm.matches[i].result.device_result);
if (result->flags & DEV_RESULT_UNCONFIGURED)
idx->skip_device = 1;
else
idx->skip_device = 0;
break;
}
case DEV_MATCH_PERIPH: {
struct periph_match_result* result;
char buf[64];
result = &(idx->ccb.cdm.matches[i].result.periph_result);
if (idx->skip_device ||
strcmp(result->periph_name, "pass") == 0)
break;
snprintf(buf, sizeof (buf), "/dev/%s%d",
result->periph_name, result->unit_number);
if(adr_size <= strlen(buf)
return -1;
strcpy(adr, buf);
/* Found next enumerable address */
return 1;
}
default:
/* printf(stderr, "unknown match type\n"); */
break;
}
(idx->i)++;
goto try_item; /* Regular function exit is return 1 above */
}
/** Brings all available, not-whitelist-banned, and accessible drives into
libburn's list of drives.
*/
int scsi_enumerate_drives(void)
{
burn_drive_enumerator_t idx;
int initialize = 1;
char buf[64];
while(1) {
ret = sg_give_next_adr(&idx, buf, sizeof(buf), initialize);
initialize = 0;
if (ret <= 0)
break;
if (burn_drive_is_banned(buf))
continue;
enumerate_common(buf, idx.result->path_id, idx.result->path_id,
0, idx.result->target_id,
idx.result->target_lun);
}
sg_give_next_adr(&idx, buf, sizeof(buf), -1);
}
/** Tells wether libburn has the given drive in use or exclusively reserved.
If it is "open" then libburn will eventually call sg_release() on it when
it is time to give up usage resp. reservation.
*/
/** Published as burn_drive.drive_is_open() */
int sg_drive_is_open(struct burn_drive * d)
{
return (d->cam != NULL);
}
/** Opens the drive for SCSI commands and - if burn activities are prone
to external interference on your system - obtains an exclusive access lock
on the drive. (Note: this is not physical tray locking.)
A drive that has been opened with sg_grab() will eventually be handed
over to sg_release() for closing and unreserving.
*/
int sg_grab(struct burn_drive *d)
{
int count;
struct cam_device *cam;
if(d->cam != NULL)
return 0;
cam = cam_open_device(d->devname, O_RDWR);
if (cam == NULL) {
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020003,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
"Could not grab drive", 0/*os_errno*/, 0);
return 0;
}
d->cam = cam;
fcntl(cam->fd, F_SETOWN, getpid());
d->released = 0;
return 1;
}
/** PORTING: Is mainly about the call to sg_close_drive() and wether it
implements the demanded functionality.
*/
/** Gives up the drive for SCSI commands and releases eventual access locks.
(Note: this is not physical tray locking.)
*/
int sg_release(struct burn_drive *d)
{
if (d->cam == NULL) {
burn_print(1, "release an ungrabbed drive. die\n");
return 0;
}
sg_close_drive(d);
return 0;
}
/** Sends a SCSI command to the drive, receives reply and evaluates wether
the command succeeded or shall be retried or finally failed.
Returned SCSI errors shall not lead to a return value indicating failure.
The callers get notified by c->error. An SCSI failure which leads not to
a retry shall be notified via scsi_notify_error().
The Libburn_log_sg_commandS facility might be of help when problems with
a drive have to be examined. It shall stay disabled for normal use.
@return: 1 success , <=0 failure
*/
int sg_issue_command(struct burn_drive *d, struct command *c)
{
int done = 0;
int err;
union ccb *ccb;
if (d->cam == NULL) {
c->error = 0;
return 0;
}
c->error = 0;
ccb = cam_getccb(d->cam);
cam_fill_csio(&ccb->csio,
1, /* retries */
NULL, /* cbfncp */
CAM_DEV_QFRZDIS, /* flags */
MSG_SIMPLE_Q_TAG, /* tag_action */
NULL, /* data_ptr */
0, /* dxfer_len */
sizeof (ccb->csio.sense_data), /* sense_len */
0, /* cdb_len */
30*1000); /* timeout */
switch (c->dir) {
case TO_DRIVE:
ccb->csio.ccb_h.flags |= CAM_DIR_OUT;
break;
case FROM_DRIVE:
ccb->csio.ccb_h.flags |= CAM_DIR_IN;
break;
case NO_TRANSFER:
ccb->csio.ccb_h.flags |= CAM_DIR_NONE;
break;
}
ccb->csio.cdb_len = c->oplen;
memcpy(&ccb->csio.cdb_io.cdb_bytes, &c->opcode, c->oplen);
memset(&ccb->csio.sense_data, 0, sizeof (ccb->csio.sense_data));
if (c->page) {
ccb->csio.data_ptr = c->page->data;
if (c->dir == FROM_DRIVE) {
ccb->csio.dxfer_len = BUFFER_SIZE;
/* touch page so we can use valgrind */
memset(c->page->data, 0, BUFFER_SIZE);
} else {
/* ts A61115: removed a ssert() */
if(c->page->bytes <= 0)
return 0;
ccb->csio.dxfer_len = c->page->bytes;
}
} else {
ccb->csio.data_ptr = NULL;
ccb->csio.dxfer_len = 0;
}
do {
err = cam_send_ccb(d->cam, ccb);
if (err == -1) {
libdax_msgs_submit(libdax_messenger,
d->global_index, 0x0002010c,
LIBDAX_MSGS_SEV_FATAL, LIBDAX_MSGS_PRIO_HIGH,
"Failed to transfer command to drive",
errno, 0);
cam_freeccb(ccb);
sg_close_drive(d);
d->released = 1;
d->busy = BURN_DRIVE_IDLE;
c->error = 1;
return -1;
}
/* XXX */
memcpy(c->sense, &ccb->csio.sense_data, ccb->csio.sense_len);
if ((ccb->ccb_h.status & CAM_STATUS_MASK) != CAM_REQ_CMP) {
if (!c->retry) {
c->error = 1;
cam_freeccb(ccb);
return 1;
}
switch (scsi_error(d, c->sense, 0)) {
case RETRY:
done = 0;
break;
case FAIL:
done = 1;
c->error = 1;
break;
}
} else {
done = 1;
}
} while (!done);
cam_freeccb(ccb);
return 1;
}
/** Tries to obtain SCSI address parameters.
@return 1 is success , 0 is failure
*/
int sg_obtain_scsi_adr(char *path, int *bus_no, int *host_no, int *channel_no,
int *target_no, int *lun_no)
{
burn_drive_enumerator_t idx;
int initialize = 1;
char buf[64];
struct periph_match_result* result;
while(1) {
ret = sg_give_next_adr(&idx, buf, sizeof(buf), initialize);
initialize = 0;
if (ret <= 0)
break;
if (strcmp(adr, buf) != 0)
continue;
result = &(idx->ccb.cdm.matches[i].result.periph_result);
*bus_no = result->path_id;
*host_no = result->path_id;
*channel_no = 0;
*target_no = result->target_id
*lun_no = result->target_lun;
sg_give_next_adr(&idx, buf, sizeof(buf), -1);
return 1;
}
sg_give_next_adr(&idx, buf, sizeof(buf), -1);
return (0);
}
/** Tells wether a text is a persistent address as listed by the enumeration
functions.
*/
int sg_is_enumerable_adr(char* adr)
{
burn_drive_enumerator_t idx;
int initialize = 1;
char buf[64];
while(1) {
ret = sg_give_next_adr(&idx, buf, sizeof(buf), initialize);
initialize = 0;
if (ret <= 0)
break;
if (strcmp(adr, buf) == 0) {
sg_give_next_adr(&idx, buf, sizeof(buf), -1);
return 1;
}
}
sg_give_next_adr(&idx, buf, sizeof(buf), -1);
return (0);
}

View File

@ -1,17 +1,19 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* >>> ts A61021 : for testing the new arrangement of code
please outcomment these defines : */
/* Revives old enumerate_common(). New version delegates much work
/* Keeps alive old enumerate_common(). New version delegates much work
to methods in drive, mmc, spc, and sbc .
*/
#define Scsi_freebsd_make_own_enumeratE 1
/* Revives old scsi_enumerate_drives(). New version delegates most work to
/* Keeps alive old sg_enumerate(). New version delegates most work to
sg_give_next_adr().
*/
#define Scsi_freebsd_old_scsi_enumeratE 1
#define Scsi_freebsd_old_sg_enumeratE 1
#include <assert.h>
@ -57,7 +59,7 @@ int burn_drive_is_banned(char *device_address);
int mmc_function_spy(char * text);
#ifdef Scsi_freebsd_old_scsi_enumeratE
#ifdef Scsi_freebsd_old_sg_enumeratE
int sg_give_next_adr(burn_drive_enumerator_t *idx,
char adr[], int adr_size, int initialize)
@ -76,11 +78,10 @@ int sg_obtain_scsi_adr(char *path, int *bus_no, int *host_no, int *channel_no,
return (0);
}
#else /* Scsi_freebsd_old_scsi_enumeratE */
#else /* Scsi_freebsd_old_sg_enumeratE */
/* ts A61021 : Moved most code from scsi_enumerate_drives under
sg_give_next_adr() */
/* Some helper functions for scsi_give_next_adr() */
/* ts A61021 : Moved most code from sg_enumerate under sg_give_next_adr() */
/* Some helper functions for sg_give_next_adr() */
static int sg_init_enumerator(burn_drive_enumerator_t *idx)
{
@ -171,7 +172,7 @@ int sg_give_next_adr(burn_drive_enumerator_t *idx,
try_item:; /* This spaghetti loop keeps the number of tabs small */
/* Loop content from old scsi_enumerate_drives() */
/* Loop content from old sg_enumerate() */
while (idx->i >= idx->ccb.cdm.num_matches) {
ret = sg_next_enumeration_buffer(idx);
@ -276,7 +277,7 @@ int sg_obtain_scsi_adr(char *path, int *bus_no, int *host_no, int *channel_no,
return (0);
}
#endif /* ! Scsi_freebsd_old_scsi_enumeratE */
#endif /* ! Scsi_freebsd_old_sg_enumeratE */
int sg_close_drive(struct burn_drive * d)
@ -293,10 +294,20 @@ int sg_drive_is_open(struct burn_drive * d)
return (d->cam != NULL);
}
int scsi_enumerate_drives(void)
void ata_enumerate(void)
{
/* ts A61021: Only a dummy function is needed in FreeBSD */
/* The difference between sg and ata should be encapsulated
in sg-linux.c */
;
}
void sg_enumerate(void)
{
#ifdef Scsi_freebsd_old_scsi_enumeratE
#ifdef Scsi_freebsd_old_sg_enumeratE
union ccb ccb;
int bufsize, fd;
@ -393,7 +404,7 @@ int scsi_enumerate_drives(void)
close(fd);
#else /* Scsi_freebsd_old_scsi_enumeratE */
#else /* Scsi_freebsd_old_sg_enumeratE */
burn_drive_enumerator_t idx;
int initialize = 1;
@ -412,7 +423,7 @@ int scsi_enumerate_drives(void)
}
sg_give_next_adr(&idx, buf, sizeof(buf), -1);
#endif /* ! Scsi_freebsd_old_scsi_enumeratE */
#endif /* ! Scsi_freebsd_old_sg_enumeratE */
}
@ -470,7 +481,6 @@ static void enumerate_common(char *fname, int bus_no, int host_no,
out.get_nwa = mmc_get_nwa;
out.close_disc = mmc_close_disc;
out.close_session = mmc_close_session;
out.close_track_session = mmc_close;
out.read_buffer_capacity = mmc_read_buffer_capacity;
out.idata = malloc(sizeof(struct burn_scsi_inquiry_data));
out.idata->valid = 0;
@ -695,3 +705,64 @@ int sg_issue_command(struct burn_drive *d, struct command *c)
return 1;
}
enum response scsi_error(struct burn_drive *d, unsigned char *sense,
int senselen)
{
int key, asc, ascq;
senselen = senselen;
key = sense[2];
asc = sense[12];
ascq = sense[13];
burn_print(12, "CONDITION: 0x%x 0x%x 0x%x on %s %s\n",
key, asc, ascq, d->idata->vendor, d->idata->product);
switch (asc) {
case 0:
burn_print(12, "NO ERROR!\n");
return RETRY;
case 2:
burn_print(1, "not ready\n");
return RETRY;
case 4:
burn_print(1,
"logical unit is in the process of becoming ready\n");
return RETRY;
case 0x20:
if (key == 5)
burn_print(1, "bad opcode\n");
return FAIL;
case 0x21:
burn_print(1, "invalid address or something\n");
return FAIL;
case 0x24:
if (key == 5)
burn_print(1, "invalid field in cdb\n");
else
break;
return FAIL;
case 0x26:
if (key == 5)
burn_print( 1, "invalid field in parameter list\n" );
return FAIL;
case 0x28:
if (key == 6)
burn_print(1,
"Not ready to ready change, medium may have changed\n");
else
break;
return RETRY;
case 0x3A:
burn_print(12, "Medium not present in %s %s\n",
d->idata->vendor, d->idata->product);
d->status = BURN_DISC_EMPTY;
return FAIL;
}
burn_print(1, "unknown failure\n");
burn_print(1, "key:0x%x, asc:0x%x, ascq:0x%x\n", key, asc, ascq);
return FAIL;
}

File diff suppressed because it is too large Load Diff

View File

@ -1,9 +1,5 @@
/* sg.c
Switcher for operating system dependent transport level modules of libburn.
Copyright (C) 2006 Thomas Schmitt <scdbackup@gmx.net>, provided under GPL
*/
/* ts A61013 : It would be nice if autotools could do that job */
#ifdef __FreeBSD__

View File

@ -3,17 +3,45 @@
#ifndef __SG
#define __SG
#ifdef __FreeBSD__
#include "os.h"
/* >>> To hold all state information of BSD device enumeration
which are now local in sg_enumerate() . So that sg_give_next_adr()
can work in BSD and sg_enumerate() can use it. */
struct burn_drive_enumeration_state {
#ifdef Scsi_freebsd_old_sg_enumeratE
int dummy;
#else
union ccb ccb;
int bufsize, fd;
unsigned int i;
int skip_device;
#endif /* ! Scsi_freebsd_old_sg_enumeratE */
/* see os.h for name of particular os-*.h where this is defined */
BURN_OS_DEFINE_DRIVE_ENUMERATOR_T
};
typedef struct burn_drive_enumeration_state burn_drive_enumerator_t;
#else /* __FreeBSD__ */
/* <<< just for testing the C syntax */
struct burn_drive_enumeration_state {
int dummy;
};
typedef struct burn_drive_enumeration_state burn_drive_enumerator_tX;
typedef int burn_drive_enumerator_t;
#endif /* ! __FreeBSD__ */
struct burn_drive;
struct command;
enum response
{ RETRY, FAIL };
/* ts A60925 : ticket 74 */
int sg_close_drive_fd(char *fname, int driveno, int *fd, int sorry);
/* ts A60922 ticket 33 */
int sg_give_next_adr(burn_drive_enumerator_t *enm_context,
@ -22,13 +50,19 @@ int sg_is_enumerable_adr(char *adr);
int sg_obtain_scsi_adr(char *path, int *bus_no, int *host_no, int *channel_no,
int *target_no, int *lun_no);
/* ts A60926 : ticket 33 ++ */
int sg_open_scsi_siblings(char *fname, int driveno,
int sibling_fds[], int *sibling_count,
int host_no, int channel_no, int id_no, int lun_no);
int sg_release_siblings(int sibling_fds[], int *sibling_count);
int sg_close_drive(struct burn_drive *d);
void sg_enumerate(void);
void ata_enumerate(void);
int sg_grab(struct burn_drive *);
int sg_release(struct burn_drive *);
int sg_issue_command(struct burn_drive *, struct command *);
enum response scsi_error(struct burn_drive *, unsigned char *, int);
/* ts A61115 : formerly sg_enumerate();ata_enumerate() */
int scsi_enumerate_drives(void);
int sg_drive_is_open(struct burn_drive * d);
#endif /* __SG */

View File

@ -22,10 +22,6 @@ enum burn_source_status burn_track_set_source(struct burn_track *t,
return BURN_SOURCE_FAILED;
s->refcount++;
t->source = s;
/* ts A61031 */
t->open_ended = (s->get_size(s) <= 0);
return BURN_SOURCE_OK;
}

View File

@ -15,7 +15,6 @@
#include <stdlib.h>
#include "libburn.h"
#include "transport.h"
#include "spc.h"
#include "mmc.h"
@ -133,13 +132,9 @@ void spc_sense_caps(struct burn_drive *d)
{
struct buffer buf;
struct scsi_mode_data *m;
int size, page_length, num_write_speeds = 0, i, speed, ret;
int size;
unsigned char *page;
struct command c;
struct burn_speed_descriptor *sd;
/* ts A61225 : 1 = report about post-MMC-1 speed descriptors */
static int speed_debug = 0;
memcpy(c.opcode, SPC_MODE_SENSE, sizeof(SPC_MODE_SENSE));
c.retry = 1;
@ -155,17 +150,6 @@ void spc_sense_caps(struct burn_drive *d)
m = d->mdata;
page = c.page->data + 8;
/* ts A61225 :
Although MODE SENSE indeed belongs to SPC, the returned code page
2Ah is part of MMC-1 to MMC-3. In MMC-1 5.2.3.4. it has 22 bytes,
in MMC-3 6.3.11 there are at least 28 bytes plus a variable length
set of speed descriptors. In MMC-5 E.11 it is declared "legacy".
*/
page_length = page[1];
m->valid = 0;
burn_mdata_free_subs(m);
m->buffer_size = page[12] * 256 + page[13];
m->dvdram_read = page[2] & 32;
m->dvdram_write = page[3] & 32;
@ -178,83 +162,28 @@ void spc_sense_caps(struct burn_drive *d)
m->cdr_read = page[2] & 1;
m->cdr_write = page[3] & 1;
m->c2_pointers = page[5] & 16;
m->underrun_proof = page[4] & 128;
/* ts A61021 : these fields are marked obsolete in MMC 3 */
m->max_read_speed = page[8] * 256 + page[9];
m->cur_read_speed = page[14] * 256 + page[15];
/* in MMC-3 : see [30-31] and blocks beginning at [32] */
m->max_write_speed = page[18] * 256 + page[19];
m->cur_write_speed = page[20] * 256 + page[21];
/* ts A61021 : New field to be set by atip (or following MMC-3 info) */
/* New field to be set by atip */
m->min_write_speed = m->max_write_speed;
/* ts A61225 : for ACh GET PERFORMANCE, Type 03h */
m->min_end_lba = 0x7fffffff;
m->max_end_lba = 0;
/* in MMC-3 : [28-29] */
m->cur_write_speed = page[20] * 256 + page[21];
/* >>> ts A61021 : iterate over all speeds :
data[30-31]: number of speed performance descriptor blocks
data[32-35]: block 0 : [+2-3] speed in kbytes/sec
*/
m->c2_pointers = page[5] & 16;
m->valid = 1;
mmc_get_configuration(d);
/* ts A61225 : end of MMC-1 , begin of MMC-3 */
if (page_length < 32) /* no write speed descriptors ? */
goto try_mmc_get_performance;
m->cur_write_speed = page[28] * 256 + page[29];
if (speed_debug)
fprintf(stderr, "LIBBURN_DEBUG: cur_write_speed = %d\n",
m->cur_write_speed);
num_write_speeds = page[30] * 256 + page[31];
m->max_write_speed = m->min_write_speed = m->cur_write_speed;
for (i = 0; i < num_write_speeds; i++) {
speed = page[32 + 4*i + 2] * 256 + page[32 + 4*i + 3];
if (speed_debug)
fprintf(stderr,
"LIBBURN_DEBUG: write speed #%d = %d kB/s (rc %d)\n",
i, speed, page[32 + 4*i +1] & 7);
/* ts A61226 */
ret = burn_speed_descriptor_new(&(d->mdata->speed_descriptors),
NULL, d->mdata->speed_descriptors, 0);
if (ret > 0) {
sd = d->mdata->speed_descriptors;
sd->source = 1;
if (d->current_profile > 0) {
sd->profile_loaded = d->current_profile;
strcpy(sd->profile_name,
d->current_profile_text);
}
sd->wrc = (( page[32 + 4*i +1] & 7 ) == 1 );
sd->write_speed = speed;
}
if (speed > m->max_write_speed)
m->max_write_speed = speed;
if (speed < m->min_write_speed)
m->min_write_speed = speed;
}
if (speed_debug)
fprintf(stderr,
"LIBBURN_DEBUG: 5Ah,2Ah min_write_speed = %d , max_write_speed = %d\n",
m->min_write_speed, m->max_write_speed);
try_mmc_get_performance:;
ret = mmc_get_write_performance(d);
if (ret > 0 && speed_debug)
fprintf(stderr,
"LIBBURN_DEBUG: ACh min_write_speed = %d , max_write_speed = %d\n",
m->min_write_speed, m->max_write_speed);
m->underrun_proof = page[4] & 128;
}
void spc_sense_error_params(struct burn_drive *d)
{
struct buffer buf;
@ -345,35 +274,16 @@ void spc_sense_write_params(struct burn_drive *d)
mmc_read_disc_info(d);
}
/* ts A61229 */
#define Libburn_mmc_compose_mode_page_5 1
/* remark ts A61104 :
Although command MODE SELECT is SPC, the content of the
Write Parameters Mode Page (05h) is MMC (Table 108 in MMC-1).
Thus the filling of the mode page should be done by a mmc_ function.
*/
void spc_select_write_params(struct burn_drive *d,
const struct burn_write_opts *o)
{
struct buffer buf;
struct command c;
#ifndef Libburn_mmc_compose_mode_page_5
int bufe, sim;
#endif
/* ts A61007 : All current callers are safe. */
/* a ssert(o->drive == d); */
/* <<< A61030
fprintf(stderr,"libburn_debug: write_type=%d multi=%d control=%d\n",
o->write_type,o->multi,o->control);
fprintf(stderr,"libburn_debug: block_type=%d spc_block_type=%d\n",
o->block_type,spc_block_type(o->block_type));
*/
memcpy(c.opcode, SPC_MODE_SELECT, sizeof(SPC_MODE_SELECT));
c.retry = 1;
c.oplen = sizeof(SPC_MODE_SELECT);
@ -387,44 +297,21 @@ void spc_select_write_params(struct burn_drive *d,
memset(c.page->data, 0, 8 + 2 + d->mdata->write_page_length);
c.page->bytes = 8 + 2 + d->mdata->write_page_length;
burn_print(12, "using write page length %d (valid %d)\n",
d->mdata->write_page_length, d->mdata->write_page_valid);
#ifdef Libburn_mmc_compose_mode_page_5
/* ts A61229 */
if (mmc_compose_mode_page_5(d, o, c.page->data + 8) <= 0)
return;
#else
c.page->data[8] = 5;
c.page->data[9] = d->mdata->write_page_length;
burn_print(12, "using write page length %d (valid %d)\n",
d->mdata->write_page_length, d->mdata->write_page_valid);
bufe = o->underrun_proof;
sim = o->simulate;
c.page->data[10] = (bufe << 6)
+ (sim << 4)
+ o->write_type;
/* ts A61106 : MMC-1 table 110 : multi==0 or multi==3 */
c.page->data[11] = ((3 * !!o->multi) << 6) | o->control;
c.page->data[11] = (o->multi << 6) | o->control;
c.page->data[12] = spc_block_type(o->block_type);
/* ts A61104 */
if(!(o->control&4)) /* audio (MMC-1 table 61) */
if(o->write_type == BURN_WRITE_TAO) /* ??? for others too ? */
c.page->data[12] = 0; /* Data Block Type: Raw Data */
c.page->data[22] = 0;
c.page->data[23] = 150; /* audio pause length */
/*XXX need session format! */
#endif /* ! Libburn_mmc_compose_mode_page_5 */
c.dir = TO_DRIVE;
d->issue_command(d, &c);
}
@ -434,10 +321,6 @@ void spc_getcaps(struct burn_drive *d)
spc_inquiry(d);
spc_sense_caps(d);
spc_sense_error_params(d);
/* <<< for debugging. >>> ??? to be fixely included here ?
mmc_read_format_capacities(d, -1);
*/
}
/*
@ -478,10 +361,7 @@ void spc_probe_write_modes(struct burn_drive *d)
c.page->data[12] = try_block_type;
c.page->data[23] = 150;
c.dir = TO_DRIVE;
d->silent_on_scsi_error = 1;
d->issue_command(d, &c);
d->silent_on_scsi_error = 0;
key = c.sense[2];
asc = c.sense[12];
@ -525,8 +405,6 @@ void spc_probe_write_modes(struct burn_drive *d)
}
}
/* ( ts A61229 : shouldn't this go to mmc.c too ?) */
/** @return -1 = error */
int spc_block_type(enum burn_block_types b)
{
@ -592,15 +470,10 @@ int burn_scsi_setup_drive(struct burn_drive *d, int bus_no, int host_no,
d->channel = channel_no;
d->lun = lun_no;
/* ts A61106 */
d->silent_on_scsi_error = 0;
d->idata = malloc(sizeof(struct burn_scsi_inquiry_data));
d->idata->valid = 0;
d->mdata = malloc(sizeof(struct scsi_mode_data));
d->mdata->valid = 0;
d->mdata->speed_descriptors = NULL;
/* ts A61007 : obsolete Assert in drive_getcaps() */
if(d->idata == NULL || d->mdata == NULL) {
@ -622,131 +495,3 @@ int burn_scsi_setup_drive(struct burn_drive *d, int bus_no, int host_no,
}
return 1;
}
/* ts A61122 */
enum response scsi_error_msg(struct burn_drive *d, unsigned char *sense,
int senselen, char msg[161],
int *key, int *asc, int *ascq)
{
*key= *asc= *ascq= -1;
if (senselen<=0 || senselen>2)
*key = sense[2];
if (senselen<=0 || senselen>12)
*asc = sense[12];
if (senselen<=0 || senselen>13)
*ascq = sense[13];
burn_print(12, "CONDITION: 0x%x 0x%x 0x%x on %s %s\n",
*key, *asc, *ascq, d->idata->vendor, d->idata->product);
switch (*asc) {
case 0:
sprintf(msg, "(no error reported by SCSI transaction)");
return RETRY;
case 2:
sprintf(msg, "not ready");
return RETRY;
case 4:
sprintf(msg,
"logical unit is in the process of becoming ready");
return RETRY;
case 0x20:
if (*key == 5)
sprintf(msg, "bad opcode");
return FAIL;
case 0x21:
sprintf(msg, "invalid address");
return FAIL;
case 0x24:
if (*key == 5)
sprintf(msg, "invalid field in cdb");
else
break;
return FAIL;
case 0x26:
if (*key == 5 )
sprintf(msg, "invalid field in parameter list" );
return FAIL;
case 0x28:
if (*key == 6)
sprintf(msg, "Medium may have changed");
else
break;
return RETRY;
case 0x3A:
sprintf(msg, "Medium not present");
d->status = BURN_DISC_EMPTY;
return FAIL;
}
sprintf(msg,
"Failure. See mmc3r10g.pdf: Sense Key %X ASC %2.2X ASCQ %2.2X",
*key, *asc, *ascq);
return FAIL;
}
/* ts A61115 moved from sg-*.c */
/* ts A61122 made it frontend to scsi_error_msg() */
enum response scsi_error(struct burn_drive *d, unsigned char *sense,
int senselen)
{
int key, asc, ascq;
char msg[160];
enum response resp;
resp = scsi_error_msg(d, sense, senselen, msg, &key, &asc, &ascq);
if (asc == 0 || asc == 0x3A)
burn_print(12, "%s\n", msg);
else
burn_print(1, "%s\n", msg);
return resp;
}
/* ts A61030 - A61115 */
/* @param flag bit0=do report conditions which are considered not an error */
int scsi_notify_error(struct burn_drive *d, struct command *c,
unsigned char *sense, int senselen, int flag)
{
int key= -1, asc= -1, ascq= -1, ret;
char msg[320],scsi_msg[160];
if (d->silent_on_scsi_error)
return 1;
strcpy(scsi_msg, " \"");
scsi_error_msg(d, sense, senselen, scsi_msg + strlen(scsi_msg),
&key, &asc, &ascq);
strcat(scsi_msg, "\"");
if(!(flag & 1)) {
/* SPC : TEST UNIT READY command */
if (c->opcode[0] == 0)
return 1;
/* MMC : READ DISC INFORMATION command */
if (c->opcode[0] == 0x51)
if (key == 0x2 && asc == 0x3A &&
ascq>=0 && ascq <= 0x02) /* MEDIUM NOT PRESENT */
return 1;
}
sprintf(msg,"SCSI error condition on command %2.2Xh :", c->opcode[0]);
if (key>=0)
sprintf(msg+strlen(msg), " key=%Xh", key);
if (asc>=0)
sprintf(msg+strlen(msg), " asc=%2.2Xh", asc);
if (ascq>=0)
sprintf(msg+strlen(msg), " ascq=%2.2Xh", ascq);
ret = libdax_msgs_submit(libdax_messenger, d->global_index, 0x0002010f,
LIBDAX_MSGS_SEV_DEBUG, LIBDAX_MSGS_PRIO_HIGH, msg,0,0);
if (ret < 0)
return ret;
ret = libdax_msgs_submit(libdax_messenger, d->global_index, 0x0002010f,
LIBDAX_MSGS_SEV_DEBUG, LIBDAX_MSGS_PRIO_HIGH,
scsi_msg,0,0);
return ret;
}

View File

@ -33,18 +33,5 @@ int spc_setup_drive(struct burn_drive *d);
int burn_scsi_setup_drive(struct burn_drive *d, int bus_no, int host_no,
int channel_no, int target_no, int lun_no, int flag);
/* ts A61115 moved from sg-*.h */
enum response { RETRY, FAIL };
enum response scsi_error(struct burn_drive *, unsigned char *, int);
/* ts A61122 */
enum response scsi_error_msg(struct burn_drive *d, unsigned char *sense,
int senselen, char msg[161],
int *key, int *asc, int *ascq);
/* ts A61030 */
/* @param flag bit0=do report conditions which are considered not an error */
int scsi_notify_error(struct burn_drive *, struct command *c,
unsigned char *sense, int senselen, int flag);
#endif /*__SPC*/

View File

@ -115,17 +115,6 @@ struct burn_track *burn_track_create(void)
t->pad = 1;
t->entry = NULL;
t->source = NULL;
t->eos = 0;
/* ts A61101 */
t->sourcecount = 0;
t->writecount = 0;
t->written_sectors = 0;
/* ts A61031 */
t->open_ended = 0;
t->track_data_done = 0;
t->postgap = 0;
t->pregap1 = 0;
t->pregap2 = 0;
@ -317,7 +306,7 @@ void burn_track_clear_isrc(struct burn_track *t)
int burn_track_get_sectors(struct burn_track *t)
{
off_t size;
int size;
int sectors, seclen;
seclen = burn_sector_length(t->mode);
@ -329,31 +318,6 @@ int burn_track_get_sectors(struct burn_track *t)
return sectors;
}
/* ts A61031 */
int burn_track_is_open_ended(struct burn_track *t)
{
return !!t->open_ended;
}
/* ts A61101 : API function */
int burn_track_get_counters(struct burn_track *t,
off_t *read_bytes, off_t *written_bytes)
{
/*
fprintf(stderr, "libburn_experimental: sizeof(off_t)=%d\n",
sizeof(off_t));
*/
*read_bytes = t->sourcecount;
*written_bytes = t->writecount;
return 1;
}
/* ts A61031 */
int burn_track_is_data_done(struct burn_track *t)
{
return !!t->track_data_done;
}
int burn_track_get_shortage(struct burn_track *t)
{
int size;
@ -386,19 +350,13 @@ int burn_disc_get_sectors(struct burn_disc *d)
void burn_track_get_entry(struct burn_track *t, struct burn_toc_entry *entry)
{
if (t->entry == NULL)
memset(entry, 0, sizeof(struct burn_toc_entry));
else
memcpy(entry, t->entry, sizeof(struct burn_toc_entry));
memcpy(entry, t->entry, sizeof(struct burn_toc_entry));
}
void burn_session_get_leadout_entry(struct burn_session *s,
struct burn_toc_entry *entry)
{
if (s->leadout_entry == NULL)
memset(entry, 0, sizeof(struct burn_toc_entry));
else
memcpy(entry, s->leadout_entry, sizeof(struct burn_toc_entry));
memcpy(entry, s->leadout_entry, sizeof(struct burn_toc_entry));
}
struct burn_session **burn_disc_get_sessions(struct burn_disc *d, int *num)

View File

@ -31,18 +31,6 @@ struct burn_track
struct burn_source *source;
/** End of Source flag */
int eos;
/* ts A61101 */
off_t sourcecount;
off_t writecount;
off_t written_sectors;
/* ts A61031 */
/** Source is of undefined length */
int open_ended;
/** End of open ended track flag : offset+payload+tail are delivered */
int track_data_done;
/** The audio/data mode for the entry. Derived from control and
possibly from reading the track's first sector. */
int mode;
@ -83,10 +71,4 @@ struct burn_disc
int burn_track_get_shortage(struct burn_track *t);
/* ts A61031 : might go to libburn.h */
int burn_track_is_open_ended(struct burn_track *t);
int burn_track_is_data_done(struct burn_track *t);
#endif /* BURN__STRUCTURE_H */

View File

@ -4,16 +4,20 @@
#define __TRANSPORT
#include "libburn.h"
#include "os.h"
#include <pthread.h>
/* sg data structures */
#include <sys/types.h>
#ifdef __FreeBSD__
/* see os.h for name of particular os-*.h where this is defined */
#define BUFFER_SIZE BURN_OS_TRANSPORT_BUFFER_SIZE
#define BUFFER_SIZE 65536/2
#else /* __FreeBSD__ */
#define BUFFER_SIZE 65536
#endif /* ! __FreeBSD__ */
enum transfer_direction
{ TO_DRIVE, FROM_DRIVE, NO_TRANSFER };
@ -36,12 +40,7 @@ struct params
struct buffer
{
/* ts A61219:
Added 4096 bytes reserve against possible buffer overflows.
(Changed in sector.c buffer flush test from >= to > BUFFER_SIZE .
This can at most cause a 1 sector overlap. Sometimes an offset
of 16 byte is applied to the output data (in some RAW mode). ) */
unsigned char data[BUFFER_SIZE + 4096];
unsigned char data[BUFFER_SIZE];
int sectors;
int bytes;
};
@ -65,7 +64,6 @@ struct burn_scsi_inquiry_data
int valid;
};
struct scsi_mode_data
{
int buffer_size;
@ -85,12 +83,6 @@ struct scsi_mode_data
/* ts A61021 */
int min_write_speed;
/* ts A61225 : Results from ACh GET PERFORMANCE, Type 03h
Speed values go into *_*_speed */
int min_end_lba;
int max_end_lba;
struct burn_speed_descriptor *speed_descriptors;
int cur_read_speed;
int cur_write_speed;
int retry_page_length;
@ -103,20 +95,6 @@ struct scsi_mode_data
};
/* ts A70112 : represents a single Formattable Capacity Descriptor as of
mmc5r03c.pdf 6.24.3.3 . There can at most be 32 of them. */
struct burn_format_descr {
/* format type: e.g 0x00 is "Full", 0x15 is "Quick" */
int type;
/* the size in bytes derived from Number of Blocks */
off_t size;
/* the Type Dependent Parameter (usually the write alignment size) */
unsigned tdp;
};
#define LIBBURN_SG_MAX_SIBLINGS 16
/** Gets initialized in enumerate_common() and burn_drive_register() */
@ -129,10 +107,15 @@ struct burn_drive
int lun;
char *devname;
#if defined(__FreeBSD__)
struct cam_device* cam;
#else
int fd;
/* see os.h for name of particular os-*.h where this is defined */
BURN_OS_TRANSPORT_DRIVE_ELEMENTS
/* ts A60926 : trying to lock against growisofs /dev/srN, /dev/scdN */
int sibling_count;
int sibling_fds[LIBBURN_SG_MAX_SIBLINGS];
#endif
/* ts A60904 : ticket 62, contribution by elmom */
/**
@ -145,37 +128,7 @@ struct burn_drive
enum burn_disc_status status;
int erasable;
/* ts A61201 from 46h GET CONFIGURATION */
int current_profile;
char current_profile_text[80];
int current_is_cd_profile;
int current_is_supported_profile;
/* ts A70114 : wether a DVD-RW media holds an incomplete session
(which could need closing after write) */
int dvd_minus_rw_incomplete;
/* ts A61218 from 46h GET CONFIGURATION */
int bg_format_status; /* 0=needs format start, 1=needs format restart*/
/* ts A70108 from 23h READ FORMAT CAPACITY mmc5r03c.pdf 6.24 */
int format_descr_type; /* 1=unformatted, 2=formatted, 3=unclear */
off_t format_curr_max_size; /* meaning depends on format_descr_type */
unsigned format_curr_blsas; /* meaning depends on format_descr_type */
int best_format_type;
off_t best_format_size;
/* The complete list of format descriptors as read with 23h */
int num_format_descr;
struct burn_format_descr format_descriptors[32];
volatile int released;
/* ts A61106 */
int silent_on_scsi_error;
int nwa; /* next writeable address */
int alba; /* absolute lba */
int rlba; /* relative lba in section */
@ -211,7 +164,6 @@ struct burn_drive
void (*unlock) (struct burn_drive *);
void (*eject) (struct burn_drive *);
void (*load) (struct burn_drive *);
int (*start_unit) (struct burn_drive *);
void (*read_disc_info) (struct burn_drive *);
void (*read_sectors) (struct burn_drive *,
int start,
@ -226,7 +178,7 @@ struct burn_drive
void (*send_cue_sheet) (struct burn_drive *, struct cue_sheet *);
void (*sync_cache) (struct burn_drive *);
int (*get_erase_progress) (struct burn_drive *);
int (*get_nwa) (struct burn_drive *, int trackno, int *lba, int *nwa);
int (*get_nwa) (struct burn_drive *);
/* ts A61009 : removed d in favor of o->drive */
/* void (*close_disc) (struct burn_drive * d,
@ -237,10 +189,6 @@ struct burn_drive
void (*close_disc) (struct burn_write_opts * o);
void (*close_session) ( struct burn_write_opts * o);
/* ts A61029 */
void (*close_track_session) ( struct burn_drive *d,
int session, int track);
int (*test_unit_ready) (struct burn_drive * d);
void (*probe_write_modes) (struct burn_drive * d);
struct params params;
@ -252,13 +200,6 @@ struct burn_drive
/* ts A61023 : get size and free space of drive buffer */
int (*read_buffer_capacity) (struct burn_drive *d);
/* ts A61220 : format media (e.g. DVD+RW) */
int (*format_unit) (struct burn_drive *d, off_t size, int flag);
/* ts A70108 */
/* mmc5r03c.pdf 6.24 : get list of available formats */
int (*read_format_capacities) (struct burn_drive *d, int top_wanted);
};
/* end of generic 'drive' data structures */

View File

@ -7,18 +7,6 @@
/* #include <a ssert.h> */
/* ts A61106 : Deliberate defect provocation macros
DO NOT DEFINE THESE IF YOU WANT SUCCESSFUL TAO !
#define Libburn_experimental_no_close_tracK 1
#define Libburn_experimental_no_close_sessioN 1
*/
/* ts A61114 : Highly experimental : try to achieve SAO on appendables
THIS DOES NOT WORK YET !
#define Libburn_sao_can_appenD 1
*/
#include <stdio.h>
#include <string.h>
#include <ctype.h>
@ -111,7 +99,7 @@ void type_to_form(int mode, unsigned char *ctladr, int *form)
*form |= 0x40;
}
int burn_write_flush(struct burn_write_opts *o, struct burn_track *track)
int burn_write_flush(struct burn_write_opts *o)
{
struct burn_drive *d = o->drive;
@ -120,115 +108,13 @@ int burn_write_flush(struct burn_write_opts *o, struct burn_track *track)
err = d->write(d, d->nwa, d->buffer);
if (err == BE_CANCELLED)
return 0;
/* A61101 */
if(track != NULL) {
track->writecount += d->buffer->bytes;
track->written_sectors += d->buffer->sectors;
}
/* ts A61119 */
d->progress.buffered_bytes += d->buffer->bytes;
d->nwa += d->buffer->sectors;
d->buffer->bytes = 0;
d->buffer->sectors = 0;
}
d->sync_cache(d);
return 1;
}
/* ts A61030 */
int burn_write_close_track(struct burn_write_opts *o, struct burn_session *s,
int tnum)
{
char msg[81];
struct burn_drive *d;
struct burn_track *t;
int todo, step, cancelled, seclen;
/* ts A61106 */
#ifdef Libburn_experimental_no_close_tracK
return 1;
#endif
d = o->drive;
t = s->track[tnum];
/* ts A61103 : pad up track to minimum size of 600 sectors */
if (t->written_sectors < 300) {
todo = 300 - t->written_sectors;
sprintf(msg,"Padding up track to minimum size (+ %d sectors)",
todo);
libdax_msgs_submit(libdax_messenger, o->drive->global_index,
0x0002011a,
LIBDAX_MSGS_SEV_NOTE, LIBDAX_MSGS_PRIO_HIGH, msg,0,0);
step = BUFFER_SIZE / 4096; /* shall fit any sector size */
if (step <= 0)
step = 1;
seclen = burn_sector_length(t->mode);
if (seclen <= 0)
seclen = 2048;
memset(d->buffer, 0, sizeof(struct buffer));
cancelled = d->cancel;
for (; todo > 0; todo -= step) {
if (step > todo)
step = todo;
d->buffer->bytes = step*seclen;
d->buffer->sectors = step;
d->cancel = 0;
d->write(d, d->nwa, d->buffer);
d->nwa += d->buffer->sectors;
t->writecount += d->buffer->bytes;
t->written_sectors += d->buffer->sectors;
d->progress.buffered_bytes += d->buffer->bytes;
}
d->cancel = cancelled;
}
/* ts A61102 */
d->busy = BURN_DRIVE_CLOSING_TRACK;
sprintf(msg, "Closing track %2.2d", tnum+1);
libdax_msgs_submit(libdax_messenger, o->drive->global_index,0x00020119,
LIBDAX_MSGS_SEV_DEBUG, LIBDAX_MSGS_PRIO_HIGH, msg,0,0);
/* MMC-1 mentions track number 0xFF for "the incomplete track",
MMC-3 does not. I tried both. 0xFF was in effect when other
bugs finally gave up and made way for readable tracks. */
d->close_track_session(o->drive, 0, 0xff); /* tnum+1); */
/* ts A61102 */
d->busy = BURN_DRIVE_WRITING;
return 1;
}
/* ts A61030 */
int burn_write_close_session(struct burn_write_opts *o, struct burn_session *s)
{
/* ts A61106 */
#ifdef Libburn_experimental_no_close_sessioN
return 1;
#endif
libdax_msgs_submit(libdax_messenger, o->drive->global_index,0x00020119,
LIBDAX_MSGS_SEV_DEBUG, LIBDAX_MSGS_PRIO_HIGH,
"Closing session", 0, 0);
/* ts A61102 */
o->drive->busy = BURN_DRIVE_CLOSING_SESSION;
o->drive->close_track_session(o->drive, 1, 0);
/* ts A61102 */
o->drive->busy = BURN_DRIVE_WRITING;
return 1;
}
/* ts A60819:
This is unused since about Feb 2006, icculus.org/burn CVS.
The compiler complains. We shall please our compiler.
@ -293,10 +179,8 @@ static int add_cue(struct cue_sheet *sheet, unsigned char ctladr,
return 1;
}
/* ts A61114: added parameter nwa */
struct cue_sheet *burn_create_toc_entries(struct burn_write_opts *o,
struct burn_session *session,
int nwa)
struct burn_session *session)
{
int i, m, s, f, form, pform, runtime = -150, ret;
unsigned char ctladr;
@ -309,11 +193,6 @@ struct cue_sheet *burn_create_toc_entries(struct burn_write_opts *o,
d = o->drive;
#ifdef Libburn_sao_can_appenD
if (d->status == BURN_DISC_APPENDABLE)
runtime = nwa-150;
#endif
sheet = malloc(sizeof(struct cue_sheet));
/* ts A61009 : react on failures of malloc(), add_cue_sheet()
@ -349,20 +228,11 @@ struct cue_sheet *burn_create_toc_entries(struct burn_write_opts *o,
/* ts A61009 */
/* a ssert(d->toc_entry == NULL); */
if (d->toc_entry != NULL) {
/* ts A61109 : this happens with appendable CDs
>>> Open question: is the existing TOC needed ? */
/* ts A61109 : for non-SAO, this sheet is thrown away later */
free((char *) d->toc_entry);
/*
libdax_msgs_submit(libdax_messenger,
d->global_index, 0x00020117,
LIBDAX_MSGS_SEV_FATAL, LIBDAX_MSGS_PRIO_HIGH,
"toc_entry of drive is already in use", 0, 0);
goto failed;
*/
}
d->toc_entry = calloc(d->toc_entries, sizeof(struct burn_toc_entry));
@ -432,17 +302,8 @@ XXX this is untested :)
*/
if (!tar[i]->pad) {
rem += burn_track_get_shortage(tar[i]);
/* ts A61101 : I doubt that linking would yield a
desireable effect. With TAO it is
counterproductive in any way.
*/
if (o->write_type == BURN_WRITE_TAO)
tar[i]->source->next = NULL;
else
if (i +1 != ntr)
tar[i]->source->next = tar[i+1]->source;
if (i +1 != ntr)
tar[i]->source->next = tar[i+1]->source;
} else if (rem) {
rem = 0;
runtime++;
@ -556,7 +417,7 @@ int burn_write_session(struct burn_write_opts *o, struct burn_session *s)
{
struct burn_drive *d = o->drive;
struct burn_track *prev = NULL, *next = NULL;
int i, ret;
int i;
d->rlba = 0;
burn_print(1, " writing a session\n");
@ -569,57 +430,23 @@ int burn_write_session(struct burn_write_opts *o, struct burn_session *s)
next = NULL;
if (!burn_write_track(o, s, i))
{ ret = 0; goto ex; }
return 0;
}
/* ts A61103 */
ret = 1;
ex:;
if (o->write_type == BURN_WRITE_TAO)
burn_write_close_session(o, s);
return ret;
}
/* ts A61218 : outsourced from burn_write_track() */
int burn_disc_init_track_status(struct burn_write_opts *o,
struct burn_session *s, int tnum, int sectors)
{
struct burn_drive *d = o->drive;
/* Update progress */
d->progress.start_sector = d->nwa;
d->progress.sectors = sectors;
d->progress.sector = 0;
/* ts A60831: added tnum-line, extended print message on proposal
by bonfire-app@wanadoo.fr in http://libburn.pykix.org/ticket/58 */
d->progress.track = tnum;
/* ts A61102 */
d->busy = BURN_DRIVE_WRITING;
return 1;
}
int burn_write_track(struct burn_write_opts *o, struct burn_session *s,
int tnum)
{
struct burn_track *t = s->track[tnum];
struct burn_drive *d = o->drive;
int i, tmp = 0, open_ended = 0, ret= 0, nwa, lba;
int i, tmp = 0;
int sectors;
char msg[80];
d->rlba = -150;
/* XXX for tao, we don't want the pregaps but still want post? */
if (o->write_type != BURN_WRITE_TAO) {
/* ts A61102 */
d->busy = BURN_DRIVE_WRITING_PREGAP;
if (t->pregap1)
d->rlba += 75;
if (t->pregap2)
@ -635,43 +462,36 @@ int burn_write_track(struct burn_write_opts *o, struct burn_session *s,
for (i = 0; i < 75; i++)
if (!sector_pregap(o, t->entry->point,
pt->entry->control, pt->mode))
{ ret = 0; goto ex; }
return 0;
}
if (t->pregap2)
for (i = 0; i < 150; i++)
if (!sector_pregap(o, t->entry->point,
t->entry->control, t->mode))
{ ret = 0; goto ex; }
return 0;
} else {
o->control = t->entry->control;
d->send_write_parameters(d, o);
/* ts A61103 */
ret = d->get_nwa(d, -1, &lba, &nwa);
sprintf(msg,
"pre-track %2.2d : get_nwa(%d), ret= %d , d->nwa= %d\n",
tnum+1, nwa, ret, d->nwa);
libdax_msgs_submit(libdax_messenger, d->global_index, 0x000002,
LIBDAX_MSGS_SEV_DEBUG, LIBDAX_MSGS_PRIO_ZERO,
msg,0,0);
if (nwa > d->nwa)
d->nwa = nwa;
}
/* user data */
sectors = burn_track_get_sectors(t);
open_ended = burn_track_is_open_ended(t);
burn_disc_init_track_status(o, s, tnum, sectors);
/* Update progress */
d->progress.start_sector = d->nwa;
d->progress.sectors = sectors;
d->progress.sector = 0;
/* ts A60831: added tnum-line, extended print message on proposal
by bonfire-app@wanadoo.fr in http://libburn.pykix.org/ticket/58 */
d->progress.track = tnum;
burn_print(12, "track %d is %d sectors long\n", tnum, sectors);
/* ts A61030 : this cannot happen. tnum is always < s->tracks */
if (tnum == s->tracks)
tmp = sectors > 150 ? 150 : sectors;
for (i = 0; open_ended || i < sectors - tmp; i++) {
for (i = 0; i < sectors - tmp; i++) {
/* ts A61023 : http://libburn.pykix.org/ticket/14
From time to time inquire drive buffer */
@ -679,24 +499,12 @@ int burn_write_track(struct burn_write_opts *o, struct burn_session *s,
d->read_buffer_capacity(d);
if (!sector_data(o, t, 0))
{ ret = 0; goto ex; }
/* ts A61031 */
if (open_ended) {
d->progress.sectors = sectors = i;
if (burn_track_is_data_done(t))
break;
}
return 0;
/* update current progress */
d->progress.sector++;
}
for (; i < sectors; i++) {
/* ts A61030: program execution never gets to this point */
fprintf(stderr,"LIBBURN_DEBUG: TNUM=%d TRACKS=%d TMP=%d\n",
tnum, s->tracks, tmp);
burn_print(1, "last track, leadout prep\n");
/* ts A61023 */
@ -704,7 +512,7 @@ int burn_write_track(struct burn_write_opts *o, struct burn_session *s,
d->read_buffer_capacity(d);
if (!sector_data(o, t, 1))
{ ret = 0; goto ex; }
return 0;
/* update progress */
d->progress.sector++;
@ -714,43 +522,23 @@ int burn_write_track(struct burn_write_opts *o, struct burn_session *s,
for (i = 0; i < 150; i++)
if (!sector_postgap(o, t->entry->point, t->entry->control,
t->mode))
{ ret = 0; goto ex; }
return 0;
i = t->offset;
if (o->write_type == BURN_WRITE_SAO) {
if (d->buffer->bytes) {
int err;
err = d->write(d, d->nwa, d->buffer);
if (err == BE_CANCELLED)
{ ret = 0; goto ex; }
/* A61101 : probably this is not all payload data */
/* A61108 : but audio count is short without this */
t->writecount += d->buffer->bytes;
t->written_sectors += d->buffer->sectors;
d->progress.buffered_bytes += d->buffer->bytes;
return 0;
d->nwa += d->buffer->sectors;
d->buffer->bytes = 0;
d->buffer->sectors = 0;
}
}
/* ts A61103 */
ret = 1;
ex:;
if (o->write_type == BURN_WRITE_TAO) {
/* ts A61103 */
/* >>> if cancelled: ensure that at least 600 kB get written */
if (!burn_write_flush(o, t))
ret = 0;
/* ts A61030 */
if (burn_write_close_track(o, s, tnum) <= 0)
ret = 0;
}
return ret;
if (o->write_type == BURN_WRITE_TAO)
if (!burn_write_flush(o))
return 0;
return 1;
}
/* ts A61009 */
@ -774,14 +562,38 @@ bad_track_mode_found:;
return 0;
}
/* ts A61218 : outsourced from burn_disc_write_sync() */
int burn_disc_init_write_status(struct burn_write_opts *o,
struct burn_disc *disc)
void burn_disc_write_sync(struct burn_write_opts *o, struct burn_disc *disc)
{
struct cue_sheet *sheet;
struct burn_drive *d = o->drive;
struct buffer buf;
struct burn_track *lt;
int first = 1, i;
int res;
d->cancel = 0;
/* ts A60924 : libburn/message.c gets obsoleted
burn_message_clear_queue();
*/
burn_print(1, "sync write of %d sessions\n", disc->sessions);
d->buffer = &buf;
memset(d->buffer, 0, sizeof(struct buffer));
d->rlba = -150;
d->toc_temp = 9;
/* Apparently some drives require this command to be sent, and a few drives
return crap. so we send the command, then ignore the result.
*/
res = d->get_nwa(d);
/* printf("ignored nwa: %d\n", res);*/
d->alba = d->start_lba;
d->nwa = d->alba;
if (o->write_type != BURN_WRITE_TAO)
d->send_write_parameters(d, o);
/* init progress before showing the state */
d->progress.session = 0;
@ -796,448 +608,19 @@ int burn_disc_init_write_status(struct burn_write_opts *o,
d->progress.start_sector = 0;
d->progress.sectors = 0;
d->progress.sector = 0;
d->progress.track = 0;
/* ts A61023 */
d->progress.buffer_capacity = 0;
d->progress.buffer_available = 0;
d->progress.buffered_bytes = 0;
d->progress.buffer_min_fill = 0xffffffff;
d->busy = BURN_DRIVE_WRITING;
return 1;
}
/* ts A61218 */
int burn_dvd_write_track(struct burn_write_opts *o,
struct burn_session *s, int tnum)
{
struct burn_track *t = s->track[tnum];
struct burn_drive *d = o->drive;
struct buffer *out = d->buffer;
int sectors;
int i, open_ended = 0, ret= 0;
sectors = burn_track_get_sectors(t);
open_ended = burn_track_is_open_ended(t);
/* >>> any type specific track preparations */;
burn_disc_init_track_status(o, s, tnum, sectors);
for (i = 0; open_ended || i < sectors; i++) {
/* From time to time inquire drive buffer */
if ((i%256)==0)
d->read_buffer_capacity(d);
/* transact a (CD sized) sector */
if (!sector_data(o, t, 0))
{ ret = 0; goto ex; }
if (open_ended) {
d->progress.sectors = sectors = i;
if (burn_track_is_data_done(t))
break;
}
/* update current progress */
d->progress.sector++;
}
/* Pad up buffer to next full 32 kB */
if (o->obs_pad && out->bytes > 0 && out->bytes < o->obs) {
memset(out->data + out->bytes, 0, o->obs - out->bytes);
out->sectors += (o->obs - out->bytes) / 2048;
out->bytes = o->obs;
}
ret = burn_write_flush(o, t);
if (ret <= 0)
goto ex;
/* >>> any other normal track finalizing */;
ret = 1;
ex:;
if (ret<=0) {
d->sync_cache(d); /* burn_write_flush() was not called */
}
return ret;
}
/* ts A61219 */
int burn_disc_close_session_dvd_plus_rw(struct burn_write_opts *o,
struct burn_session *s)
{
struct burn_drive *d = o->drive;
d->busy = BURN_DRIVE_CLOSING_SESSION;
/* This seems to be a quick end : "if (!dvd_compat)" */
/* >>> Stop de-icing (ongoing background format) quickly
by mmc_close() (but with opcode[2]=0).
Wait for unit to get ready.
return 1;
*/
/* Else: end eventual background format in a "DVD-RO" compatible way */
d->close_track_session(d, 1, 0); /* same as CLOSE SESSION for CD */
d->busy = BURN_DRIVE_WRITING;
return 1;
}
/* ts A61228 */
int burn_disc_close_session_dvd_minus_rw(struct burn_write_opts *o,
struct burn_session *s)
{
struct burn_drive *d = o->drive;
d->busy = BURN_DRIVE_CLOSING_SESSION;
if (d->current_profile == 0x13) {
d->close_track_session(d, 1, 0); /* CLOSE SESSION, 010b */
/* ??? under what circumstances to use close functiom 011b
"Finalize disc" ? */
}
d->busy = BURN_DRIVE_WRITING;
return 1;
}
/* ts A61218 */
int burn_dvd_write_session(struct burn_write_opts *o,
struct burn_session *s)
{
int i,ret;
struct burn_drive *d = o->drive;
for (i = 0; i < s->tracks; i++) {
ret = burn_dvd_write_track(o, s, i);
if (ret <= 0)
break;
}
if (d->current_profile == 0x1a) {
/* DVD+RW */
ret = burn_disc_close_session_dvd_plus_rw(o, s);
if (ret <= 0)
return 0;
} else if (d->current_profile == 0x13) {
/* DVD-RW restricted overwrite */
if (d->dvd_minus_rw_incomplete) {
ret = burn_disc_close_session_dvd_minus_rw(o, s);
if (ret <= 0)
return 0;
}
} else if (d->current_profile == 0x12) {
/* DVD-RAM */
/* ??? any finalization needed ? */;
}
return 1;
}
/* ts A61218 : learned much from dvd+rw-tools-7.0/growisofs_mmc.cpp */
int burn_disc_setup_dvd_plus_rw(struct burn_write_opts *o,
struct burn_disc *disc)
{
struct burn_drive *d = o->drive;
int ret;
char msg[160];
if (d->bg_format_status==0 || d->bg_format_status==1) {
d->busy = BURN_DRIVE_FORMATTING;
/* start or re-start dvd_plus_rw formatting */
ret = d->format_unit(d, (off_t) 0, 0);
if (ret <= 0)
return 0;
d->busy = BURN_DRIVE_WRITING;
}
d->nwa = 0;
if (o->start_byte >= 0) {
d->nwa = o->start_byte / 2048;
sprintf(msg, "Write start address is %d * 2048", d->nwa);
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020127,
LIBDAX_MSGS_SEV_NOTE, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
}
/* >>> perform OPC if needed */;
/* >>> ? what else ? */;
return 1;
}
/* ts A61228 : learned much from dvd+rw-tools-7.0/growisofs_mmc.cpp */
int burn_disc_setup_dvd_minus_rw(struct burn_write_opts *o,
struct burn_disc *disc)
{
struct burn_drive *d = o->drive;
char msg[160];
int ret;
d->nwa = 0;
if (o->start_byte >= 0) {
d->nwa = o->start_byte / 32768; /* align to 32 kB */
sprintf(msg, "Write start address is %d * 32768", d->nwa);
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020127,
LIBDAX_MSGS_SEV_NOTE, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
d->nwa *= 16; /* convert to 2048 block units */
}
if (d->current_profile == 0x13) { /* DVD-RW restricted overwrite */
/* ??? mmc5r03c.pdf 7.5.2 :
"For DVD-RW media ... If a medium is in Restricted overwrite
mode, this mode page shall not be used."
But growisofs composes a page 5 and sends it.
mmc5r03c.pdf 5.3.16 , table 127 specifies that mode page 5
shall be supported with feature 0026h Restricted Overwrite.
5.3.22 describes a feature 002Ch Rigid Restrictive Overwrite
which seems to apply to DVD-RW and does not mention page 5.
5.4.14 finally states that profile 0013h includes feature
002Ch rather than 0026h.
d->send_write_parameters(d, o);
*/
d->busy = BURN_DRIVE_FORMATTING;
/* "quick grow" to at least byte equivalent of d->nwa */
ret = d->format_unit(d, (off_t) d->nwa * (off_t) 2048,
(d->nwa > 0) << 3);
if (ret <= 0)
return 0;
d->busy = BURN_DRIVE_WRITING;
} else {
sprintf(msg, "Unsuitable media detected. Profile %4.4Xh %s",
d->current_profile, d->current_profile_text);
libdax_msgs_submit(libdax_messenger, d->global_index,
0x0002011e,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
return 0;
}
/* >>> perform OPC if needed */;
/* >>> */;
return 1;
}
/* ts A61218 */
int burn_dvd_write_sync(struct burn_write_opts *o,
struct burn_disc *disc)
{
int i, ret, sx, tx, mode, exotic_track = 0;
struct burn_drive *d = o->drive;
char msg[160];
for (sx = 0; sx < disc->sessions; sx++)
for (tx = 0 ; tx < disc->session[sx]->tracks; tx++) {
mode = disc->session[sx]->track[tx]->mode;
if (disc->session[sx]->track[tx]->mode != BURN_MODE1)
exotic_track = 1;
}
if (exotic_track) {
sprintf(msg,"DVD Media are unsuitable for desired track type");
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020123,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
goto early_failure;
}
if (d->current_profile == 0x1a || d->current_profile == 0x13 ||
d->current_profile == 0x12) {
/* DVD+RW , DVD-RW Restricted Overwrite , DVD-RAM */
if (disc->sessions!=1 || disc->session[0]->tracks>1
|| o->multi ) {
sprintf(msg,
"Burning is restricted to a single track and no multi-session on %s",
d->current_profile_text);
libdax_msgs_submit(libdax_messenger, d->global_index,
0x0002011f,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
goto early_failure;
}
}
if (d->current_profile == 0x1a || d->current_profile == 0x12) {
/* DVD+RW , DVD-RAM */
if (o->start_byte >= 0 && (o->start_byte % 2048)) {
sprintf(msg,
"Write start address not properly aligned to 2048");
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020125,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
goto early_failure;
}
ret = 1;
if (d->current_profile == 0x1a)
ret = burn_disc_setup_dvd_plus_rw(o, disc);
if (ret <= 0) {
sprintf(msg,
"Write preparation setup failed for DVD+RW");
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020121,
LIBDAX_MSGS_SEV_FATAL, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
goto early_failure;
}
o->obs_pad = 0; /* no filling-up of track's last 32k buffer */
} else if (d->current_profile == 0x13) { /* DVD-RW Rest. Overwrite */
if (o->start_byte >= 0 && (o->start_byte % 32768)) {
sprintf(msg,
"Write start address not properly aligned to 32K");
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020125,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
goto early_failure;
}
ret = burn_disc_setup_dvd_minus_rw(o, disc);
if (ret <= 0) {
sprintf(msg,
"Write preparation setup failed for DVD-RW");
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020121,
LIBDAX_MSGS_SEV_FATAL, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
goto early_failure;
}
/* ??? is this necessary ? */
o->obs_pad = 1; /* fill-up track's last 32k buffer */
} else {
sprintf(msg, "Unsuitable media detected. Profile %4.4Xh %s",
d->current_profile, d->current_profile_text);
libdax_msgs_submit(libdax_messenger, d->global_index,
0x0002011e,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
goto early_failure;
}
o->obs = 32*1024; /* buffer flush trigger for sector.c:get_sector() */
for (i = 0; i < disc->sessions; i++) {
/* update progress */
d->progress.session = i;
d->progress.tracks = disc->session[i]->tracks;
ret = burn_dvd_write_session(o, disc->session[i]);
if (ret <= 0)
goto ex;
/* XXX: currently signs an end of session */
d->progress.sector = 0;
d->progress.start_sector = 0;
d->progress.sectors = 0;
}
/* >>> eventual normal finalization measures */
ret = 1;
ex:;
/* >>> eventual emergency finalization measures */
/* update media state records */
burn_drive_mark_unready(d);
burn_drive_inquire_media(d);
d->busy = BURN_DRIVE_IDLE;
return ret;
early_failure:;
return 0;
}
void burn_disc_write_sync(struct burn_write_opts *o, struct burn_disc *disc)
{
struct cue_sheet *sheet;
struct burn_drive *d = o->drive;
struct buffer buf;
struct burn_track *lt;
int first = 1, i, ret, lba, nwa = 0;
char msg[80];
/* ts A60924 : libburn/message.c gets obsoleted
burn_message_clear_queue();
*/
/* ts A61224 */
burn_disc_init_write_status(o, disc); /* must be done very early */
d->buffer = &buf;
memset(d->buffer, 0, sizeof(struct buffer));
d->rlba = -150;
d->toc_temp = 9;
/* ts A61218 */
if (! d->current_is_cd_profile) {
ret = burn_dvd_write_sync(o, disc);
if (ret <= 0)
goto fail_wo_sync;
return;
}
if (o->start_byte >= 0) {
sprintf(msg, "Write start address not supported");
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00020124,
LIBDAX_MSGS_SEV_SORRY, LIBDAX_MSGS_PRIO_HIGH,
msg, 0,0);
goto fail_wo_sync;
}
burn_print(1, "sync write of %d CD sessions\n", disc->sessions);
/* Apparently some drives require this command to be sent, and a few drives
return crap. so we send the command, then ignore the result.
*/
/* ts A61107 : moved up send_write_parameters because LG GSA-4082B
seems to dislike get_nwa() in advance */
d->alba = d->start_lba; /* ts A61114: this looks senseless */
d->nwa = d->alba;
if (o->write_type == BURN_WRITE_TAO) {
nwa = 0; /* get_nwa() will be called in burn_track() */
} else {
d->send_write_parameters(d, o);
ret = d->get_nwa(d, -1, &lba, &nwa);
sprintf(msg, "Inquired nwa: %d (ret=%d)", nwa, ret);
libdax_msgs_submit(libdax_messenger, d->global_index,
0x00000002,
LIBDAX_MSGS_SEV_DEBUG, LIBDAX_MSGS_PRIO_ZERO,
msg,0,0);
}
for (i = 0; i < disc->sessions; i++) {
/* update progress */
d->progress.session = i;
d->progress.tracks = disc->session[i]->tracks;
/* ts A61114: added parameter nwa */
sheet = burn_create_toc_entries(o, disc->session[i], nwa);
sheet = burn_create_toc_entries(o, disc->session[i]);
/* ts A61009 */
if (sheet == NULL)
@ -1253,39 +636,8 @@ return crap. so we send the command, then ignore the result.
goto fail;
} else {
if (first) {
/* ts A61030 : 0 made the burner take data. */
/* ts A61103 : Meanwhile d->nwa is updated in
burn_write_track() */
if(o->write_type == BURN_WRITE_TAO) {
d->nwa= d->alba = 0;
} else {
#ifdef Libburn_sao_can_appenD
/* ts A61114: address for d->write() */
if (d->status == BURN_DISC_APPENDABLE
&& o->write_type == BURN_WRITE_SAO) {
d->nwa = d->alba = nwa-150;
sprintf(msg,
"SAO appendable d->nwa= %d\n", d->nwa);
libdax_msgs_submit(
libdax_messenger, d->global_index, 0x000002,
LIBDAX_MSGS_SEV_DEBUG, LIBDAX_MSGS_PRIO_ZERO,
msg,0,0);
} else {
d->nwa = -150;
d->alba = -150;
}
#else
d->nwa = -150;
d->alba = -150;
#endif /* ! Libburn_sao_can_appenD */
}
d->nwa = -150;
d->alba = -150;
} else {
d->nwa += 4500;
d->alba += 4500;
@ -1300,13 +652,8 @@ return crap. so we send the command, then ignore the result.
lt->mode))
goto fail;
} else {
/* ts A61030 */
if (o->write_type != BURN_WRITE_TAO)
if (!burn_write_flush(o, NULL))
goto fail;
if (!burn_write_flush(o))
goto fail;
d->nwa += first ? 6750 : 2250;
d->alba += first ? 6750 : 2250;
}
@ -1318,33 +665,23 @@ return crap. so we send the command, then ignore the result.
d->progress.start_sector = 0;
d->progress.sectors = 0;
}
/* ts A61030: extended skipping of flush to TAO: session is closed */
if (o->write_type != BURN_WRITE_SAO && o->write_type != BURN_WRITE_TAO)
if (!burn_write_flush(o, NULL))
if (o->write_type != BURN_WRITE_SAO)
if (!burn_write_flush(o))
goto fail;
sleep(1);
/* ts A61125 : update media state records */
burn_drive_mark_unready(d);
burn_drive_inquire_media(d);
burn_print(1, "done\n");
d->busy = BURN_DRIVE_IDLE;
/* ts A61012 : This return was traditionally missing. I suspect this
/* ts A61012 : This return was traditionally missing, a suspect this
to have caused Cdrskin_eject() failures */
return;
fail:
d->sync_cache(d);
fail_wo_sync:;
usleep(500001); /* ts A61222: to avoid a warning from remove_worker()*/
burn_print(1, "done - failed\n");
libdax_msgs_submit(libdax_messenger, d->global_index, 0x0002010b,
LIBDAX_MSGS_SEV_FATAL, LIBDAX_MSGS_PRIO_HIGH,
"Burn run failed", 0, 0);
d->cancel = 1;
d->busy = BURN_DRIVE_IDLE;
}

View File

@ -9,8 +9,7 @@ struct burn_write_opts;
struct burn_disc;
struct cue_sheet *burn_create_toc_entries(struct burn_write_opts *o,
struct burn_session *session,
int nwa);
struct burn_session *session);
int burn_sector_length(int trackmode);
int burn_subcode_length(int trackmode);
@ -25,11 +24,6 @@ int burn_write_leadout(struct burn_write_opts *o,
int burn_write_session(struct burn_write_opts *o, struct burn_session *s);
int burn_write_track(struct burn_write_opts *o, struct burn_session *s,
int tnum);
int burn_write_flush(struct burn_write_opts *o, struct burn_track *track);
/* ts A61030 : necessary for TAO */
int burn_write_close_track(struct burn_write_opts *o, struct burn_session *s,
int tnum);
int burn_write_close_session(struct burn_write_opts *o,struct burn_session *s);
int burn_write_flush(struct burn_write_opts *o);
#endif /* BURN__WRITE_H */

11
libisofs-1.pc.in Normal file
View File

@ -0,0 +1,11 @@
prefix=@prefix@
exec_prefix=@exec_prefix@
libdir=@libdir@
includedir=@includedir@
Name: libisofs
Description: ISO9660 filesystem creation library
Version: @VERSION@
Requires:
Libs: -L${libdir} -lisofs
Cflags: -I${includedir}/libburn

4
libisofs/Makefile Executable file
View File

@ -0,0 +1,4 @@
all clean:
$(MAKE) -C .. -$(MAKEFLAGS) $@
.PHONY: all clean

49
libisofs/Makefile.am Executable file
View File

@ -0,0 +1,49 @@
pkgconfigdir=$(libdir)/pkgconfig
libincludedir=$(includedir)/libburn
##bin_PROGRAMS = test
lib_LTLIBRARIES = libisofs.la
libisofs_la_SOURCES = \
tree.h \
tree.c \
volume.h \
volume.c \
util.h \
util.c \
ecma119.c \
ecma119.h \
ecma119_tree.c \
ecma119_tree.h \
susp.h \
susp.c \
rockridge.h \
rockridge.c \
joliet.c \
joliet.h
libinclude_HEADERS = libisofs.h
##test_SOURCES = test.c
##test_LDADD = libisofs.la
##noinst_PROGRAMS = test
##test_SOURCES = test.c
##test_LDADD = $(libisofs_la_OBJECTS)
##INCLUDES = -I../burn/libburn
## ========================================================================= ##
indent_files = $(libisofs_la_SOURCES)
indent: $(indent_files)
indent -bad -bap -nbbb -nbbo -nbc -bli0 -br -bls \
-cdw -ce -cli0 -ncs -nbfda -i8 -l79 -lc79 \
-lp -saf -sai -nprs -npsl -saw -sob -ss -ut \
-sbi0 -nsc -ts8 -npcs -ncdb -fca \
$^
.PHONY: indent
## ========================================================================= ##

694
libisofs/ecma119.c Executable file
View File

@ -0,0 +1,694 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set noet ts=8 sts=8 sw=8 : */
#include <string.h>
#include <wchar.h>
#include <stdlib.h>
#include <time.h>
#include <assert.h>
#include <err.h>
#include "ecma119.h"
#include "ecma119_tree.h"
#include "susp.h"
#include "rockridge.h"
#include "joliet.h"
#include "volume.h"
#include "tree.h"
#include "util.h"
#include "libisofs.h"
#include "libburn/libburn.h"
/* burn-source compatible stuff */
static int
bs_read(struct burn_source *bs, unsigned char *buf, int size);
static off_t
bs_get_size(struct burn_source *bs);
static void
bs_free_data(struct burn_source *bs);
typedef void (*write_fn)(struct ecma119_write_target*, uint8_t*);
/* return true if the given state is only required for Joliet volumes */
static int
is_joliet_state(enum ecma119_write_state);
static void
next_state(struct ecma119_write_target *t);
/* write t->state_data to the buf, one block at a time */
static void
write_data_chunk(struct ecma119_write_target *t, uint8_t *buf);
/* writing functions. All these functions assume the buf is large enough */
static void
write_pri_vol_desc(struct ecma119_write_target *t, uint8_t *buf);
static void
write_vol_desc_terminator(struct ecma119_write_target *t, uint8_t *buf);
static void
write_path_table(struct ecma119_write_target *t, int l_type, uint8_t *buf);
static void
write_l_path_table(struct ecma119_write_target *t, uint8_t *buf);
static void
write_m_path_table(struct ecma119_write_target *t, uint8_t *buf);
static void
write_one_dir_record(struct ecma119_write_target *t,
struct ecma119_tree_node *dir,
int file_id,
uint8_t *buf);
static void
write_one_dir(struct ecma119_write_target *t,
struct ecma119_tree_node *dir,
uint8_t *buf);
static void
write_dirs(struct ecma119_write_target *t, uint8_t *buf);
/* wrapper functions for writing */
static void wr_system_area(struct ecma119_write_target*, uint8_t*);
static void wr_pri_vol_desc(struct ecma119_write_target*, uint8_t*);
static void wr_vol_desc_term(struct ecma119_write_target*, uint8_t*);
static void wr_l_path_table(struct ecma119_write_target*, uint8_t*);
static void wr_m_path_table(struct ecma119_write_target*, uint8_t*);
static void wr_dir_records(struct ecma119_write_target*, uint8_t*);
static void wr_files(struct ecma119_write_target*, uint8_t*);
static const write_fn writers[] =
{
NULL,
wr_system_area,
wr_pri_vol_desc,
joliet_wr_sup_vol_desc,
wr_vol_desc_term,
wr_l_path_table,
wr_m_path_table,
joliet_wr_l_path_table,
joliet_wr_m_path_table,
wr_dir_records,
joliet_wr_dir_records,
wr_files
};
/* When a writer is created, we
* 1) create an ecma119 tree
* 2) add SUSP fields (if necessary)
* 3) calculate the size and position of all nodes in the tree
* 4) finalize SUSP fields (if necessary)
*/
static void
add_susp_fields_rec(struct ecma119_write_target *t,
struct ecma119_tree_node *node)
{
size_t i;
if (!node->iso_self)
return;
rrip_add_PX(t, node);
rrip_add_NM(t, node);
rrip_add_TF(t, node);
if (node->iso_self->attrib.st_rdev)
rrip_add_PN(t, node);
if (S_ISLNK(node->iso_self->attrib.st_mode))
rrip_add_SL(t, node);
if (node->type == ECMA119_FILE && node->file.real_me)
rrip_add_CL(t, node);
if (node->type == ECMA119_DIR
&& node->dir.real_parent != node->parent) {
rrip_add_RE(t, node);
rrip_add_PL(t, node);
}
susp_add_CE(t, node);
if (node->type == ECMA119_DIR) {
for (i = 0; i < node->dir.nchildren; i++) {
add_susp_fields_rec(t, node->dir.children[i]);
}
}
}
static void
add_susp_fields(struct ecma119_write_target *t)
{
susp_add_SP(t, t->root);
rrip_add_ER(t, t->root);
add_susp_fields_rec(t, t->root);
}
/**
* Fill out the dir.len and dir.CE_len fields for each
* ecma119_tree_node that is a directory. Also calculate the total number of
* directories and the number of files for which we need to write out data.
* (dirlist_len and filelist_len)
*/
static void
calc_dir_size(struct ecma119_write_target *t,
struct ecma119_tree_node *dir)
{
size_t i;
assert(dir->type == ECMA119_DIR);
t->dirlist_len++;
dir->dir.len = 34 + dir->dir.self_susp.non_CE_len
+ 34 + dir->dir.parent_susp.non_CE_len;
dir->dir.CE_len = dir->dir.self_susp.CE_len
+ dir->dir.parent_susp.CE_len;
for (i = 0; i < dir->dir.nchildren; i++) {
struct ecma119_tree_node *ch = dir->dir.children[i];
dir->dir.len += ch->dirent_len + ch->susp.non_CE_len;
dir->dir.CE_len += ch->susp.CE_len;
}
t->total_dir_size += round_up(dir->dir.len + dir->dir.CE_len,
t->block_size);
for (i = 0; i < dir->dir.nchildren; i++) {
struct ecma119_tree_node *ch = dir->dir.children[i];
struct iso_tree_node *iso = ch->iso_self;
if (ch->type == ECMA119_DIR) {
calc_dir_size(t, ch);
} else if (iso && iso->attrib.st_size
&& iso->loc.type == LIBISO_FILESYS
&& iso->loc.path) {
t->filelist_len++;
}
}
}
/**
* Fill out the block field in each ecma119_tree_node that is a directory and
* fill out t->dirlist.
*/
static void
calc_dir_pos(struct ecma119_write_target *t,
struct ecma119_tree_node *dir)
{
size_t i;
assert(dir->type == ECMA119_DIR);
/* we don't need to set iso_self->block since each tree writes
* its own directories */
dir->block = t->curblock;
t->curblock += div_up(dir->dir.len + dir->dir.CE_len, t->block_size);
t->dirlist[t->curfile++] = dir;
for (i = 0; i < dir->dir.nchildren; i++) {
struct ecma119_tree_node *ch = dir->dir.children[i];
if (ch->type == ECMA119_DIR)
calc_dir_pos(t, ch);
}
/* reset curfile when we're finished */
if (!dir->parent) {
t->curfile = 0;
}
}
/**
* Fill out the block field for each ecma119_tree_node that is a file and fill
* out t->filelist.
*/
static void
calc_file_pos(struct ecma119_write_target *t,
struct ecma119_tree_node *dir)
{
size_t i;
assert(dir->type == ECMA119_DIR);
for (i = 0; i < dir->dir.nchildren; i++) {
struct ecma119_tree_node *ch = dir->dir.children[i];
if (ch->type == ECMA119_FILE && ch->iso_self) {
struct iso_tree_node *iso = ch->iso_self;
off_t size = iso->attrib.st_size;
iso->block = ch->block = t->curblock;
t->curblock += div_up(size, t->block_size);
if (size && iso->loc.type == LIBISO_FILESYS
&& iso->loc.path)
t->filelist[t->curfile++] = ch;
}
}
for (i = 0; i < dir->dir.nchildren; i++) {
struct ecma119_tree_node *ch = dir->dir.children[i];
if (ch->type == ECMA119_DIR)
calc_file_pos(t, ch);
}
/* reset curfile when we're finished */
if (!dir->parent) {
t->curfile = 0;
}
}
struct ecma119_write_target*
ecma119_target_new(struct iso_volset *volset,
int volnum,
int level,
int flags)
{
struct ecma119_write_target *t =
calloc(1, sizeof(struct ecma119_write_target));
size_t i, j, cur;
struct iso_tree_node *iso_root = volset->volume[volnum]->root;
volset->refcount++;
t->root = ecma119_tree_create(t, iso_root);
t->joliet = (flags & ECMA119_JOLIET) ? 1 : 0;
if (t->joliet)
t->joliet_root = joliet_tree_create(t, iso_root);
t->volset = volset;
t->volnum = volnum;
t->now = time(NULL);
t->rockridge = (flags & ECMA119_ROCKRIDGE) ? 1 : 0;
t->iso_level = level;
t->block_size = 2048;
if (t->rockridge)
add_susp_fields(t);
calc_dir_size(t, t->root);
if (t->joliet) {
joliet_calc_dir_size(t, t->joliet_root);
t->pathlist_joliet = calloc(1, sizeof(void*) * t->dirlist_len);
t->dirlist_joliet = calloc(1, sizeof(void*) * t->dirlist_len);
}
t->dirlist = calloc(1, sizeof(void*) * t->dirlist_len);
t->pathlist = calloc(1, sizeof(void*) * t->dirlist_len);
t->filelist = calloc(1, sizeof(void*) * t->filelist_len);
/* fill out the pathlist */
t->pathlist[0] = t->root;
t->path_table_size = 10; /* root directory record */
cur = 1;
for (i = 0; i < t->dirlist_len; i++) {
struct ecma119_tree_node *dir = t->pathlist[i];
for (j = 0; j < dir->dir.nchildren; j++) {
struct ecma119_tree_node *ch = dir->dir.children[j];
if (ch->type == ECMA119_DIR) {
size_t len = 8 + strlen(ch->name);
t->pathlist[cur++] = ch;
t->path_table_size += len + len % 2;
}
}
}
t->curblock = 16 /* system area */
+ 1 /* volume desc */
+ 1; /* volume desc terminator */
if (t->joliet) /* supplementary vol desc */
t->curblock += div_up (2048, t->block_size);
t->l_path_table_pos = t->curblock;
t->curblock += div_up(t->path_table_size, t->block_size);
t->m_path_table_pos = t->curblock;
t->curblock += div_up(t->path_table_size, t->block_size);
if (t->joliet) {
joliet_prepare_path_tables(t);
t->l_path_table_pos_joliet = t->curblock;
t->curblock += div_up(t->path_table_size_joliet, t->block_size);
t->m_path_table_pos_joliet = t->curblock;
t->curblock += div_up(t->path_table_size_joliet, t->block_size);
}
calc_dir_pos(t, t->root);
if (t->joliet)
joliet_calc_dir_pos(t, t->joliet_root);
calc_file_pos(t, t->root);
if (t->joliet)
joliet_update_file_pos (t, t->joliet_root);
if (t->rockridge) {
susp_finalize(t, t->root);
rrip_finalize(t, t->root);
}
t->total_size = t->curblock * t->block_size;
t->vol_space_size = t->curblock;
/* prepare for writing */
t->curblock = 0;
t->state = ECMA119_WRITE_SYSTEM_AREA;
return t;
}
static int
is_joliet_state(enum ecma119_write_state state)
{
return state == ECMA119_WRITE_SUP_VOL_DESC_JOLIET
|| state == ECMA119_WRITE_L_PATH_TABLE_JOLIET
|| state == ECMA119_WRITE_M_PATH_TABLE_JOLIET
|| state == ECMA119_WRITE_DIR_RECORDS_JOLIET;
}
static void
next_state(struct ecma119_write_target *t)
{
t->state++;
while (!t->joliet && is_joliet_state(t->state))
t->state++;
printf ("now in state %d, curblock=%d\n", (int)t->state, (int)t->curblock);
}
static void
wr_system_area(struct ecma119_write_target *t, uint8_t *buf)
{
memset(buf, 0, t->block_size);
if (t->curblock == 15) {
next_state(t);
}
}
static void
wr_pri_vol_desc(struct ecma119_write_target *t, uint8_t *buf)
{
ecma119_start_chunking(t, write_pri_vol_desc, 2048, buf);
}
static void
wr_vol_desc_term(struct ecma119_write_target *t, uint8_t *buf)
{
ecma119_start_chunking(t, write_vol_desc_terminator, 2048, buf);
}
static void
wr_l_path_table(struct ecma119_write_target *t, uint8_t *buf)
{
ecma119_start_chunking(t, write_l_path_table, t->path_table_size, buf);
}
static void
wr_m_path_table(struct ecma119_write_target *t, uint8_t *buf)
{
ecma119_start_chunking(t, write_m_path_table, t->path_table_size, buf);
}
static void
wr_dir_records(struct ecma119_write_target *t, uint8_t *buf)
{
ecma119_start_chunking(t, write_dirs, t->total_dir_size, buf);
}
static void
wr_files(struct ecma119_write_target *t, uint8_t *buf)
{
struct state_files *f_st = &t->state_files;
size_t nread;
struct ecma119_tree_node *f = t->filelist[f_st->file];
const char *path = f->iso_self->loc.path;
if (!f_st->fd) {
f_st->data_len = f->iso_self->attrib.st_size;
f_st->fd = fopen(path, "r");
if (!f_st->fd)
err(1, "couldn't open %s for reading", path);
assert(t->curblock == f->block);
}
nread = fread(buf, 1, t->block_size, f_st->fd);
f_st->pos += t->block_size;
if (nread < 0)
warn("problem reading from %s", path);
else if (nread != t->block_size && f_st->pos < f_st->data_len)
warnx("incomplete read from %s", path);
if (f_st->pos >= f_st->data_len) {
fclose(f_st->fd);
f_st->fd = 0;
f_st->pos = 0;
f_st->file++;
if (f_st->file >= t->filelist_len)
next_state(t);
}
}
static void
write_pri_vol_desc(struct ecma119_write_target *t, uint8_t *buf)
{
struct ecma119_pri_vol_desc *vol = (struct ecma119_pri_vol_desc*)buf;
struct iso_volume *volume = t->volset->volume[t->volnum];
char *vol_id = str2ascii(volume->volume_id);
char *pub_id = str2ascii(volume->publisher_id);
char *data_id = str2ascii(volume->data_preparer_id);
char *volset_id = str2ascii(t->volset->volset_id);
vol->vol_desc_type[0] = 1;
memcpy(vol->std_identifier, "CD001", 5);
vol->vol_desc_version[0] = 1;
memcpy(vol->system_id, "SYSID", 5);
if (vol_id)
strncpy((char*)vol->volume_id, vol_id, 32);
iso_bb(vol->vol_space_size, t->vol_space_size, 4);
iso_bb(vol->vol_set_size, t->volset->volset_size, 2);
iso_bb(vol->vol_seq_number, t->volnum + 1, 2);
iso_bb(vol->block_size, t->block_size, 2);
iso_bb(vol->path_table_size, t->path_table_size, 4);
iso_lsb(vol->l_path_table_pos, t->l_path_table_pos, 4);
iso_msb(vol->m_path_table_pos, t->m_path_table_pos, 4);
write_one_dir_record(t, t->root, 3, vol->root_dir_record);
strncpy((char*)vol->vol_set_id, volset_id, 128);
strncpy((char*)vol->publisher_id, pub_id, 128);
strncpy((char*)vol->data_prep_id, data_id, 128);
strncpy((char*)vol->application_id, "APPID", 128);
iso_datetime_17(vol->vol_creation_time, t->now);
iso_datetime_17(vol->vol_modification_time, t->now);
iso_datetime_17(vol->vol_effective_time, t->now);
vol->file_structure_version[0] = 1;
free(vol_id);
free(volset_id);
free(pub_id);
free(data_id);
}
static void
write_vol_desc_terminator(struct ecma119_write_target *t, uint8_t *buf)
{
struct ecma119_vol_desc_terminator *vol =
(struct ecma119_vol_desc_terminator*) buf;
vol->vol_desc_type[0] = 255;
memcpy(vol->std_identifier, "CD001", 5);
vol->vol_desc_version[0] = 1;
}
static void
write_path_table(struct ecma119_write_target *t, int l_type, uint8_t *buf)
{
void (*write_int)(uint8_t*, uint32_t, int) = l_type ? iso_lsb
: iso_msb;
size_t i;
struct ecma119_path_table_record *rec;
struct ecma119_tree_node *dir;
int parent = 0;
for (i = 0; i < t->dirlist_len; i++) {
dir = t->pathlist[i];
while ((i) && t->pathlist[parent] != dir->parent)
parent++;
assert(parent < i || i == 0);
rec = (struct ecma119_path_table_record*) buf;
rec->len_di[0] = dir->parent ? (uint8_t) strlen(dir->name) : 1;
rec->len_xa[0] = 0;
write_int(rec->block, dir->block, 4);
write_int(rec->parent, parent + 1, 2);
if (dir->parent)
memcpy(rec->dir_id, dir->name, rec->len_di[0]);
buf += 8 + rec->len_di[0] + (rec->len_di[0] % 2);
}
}
static void
write_l_path_table(struct ecma119_write_target *t, uint8_t *buf)
{
write_path_table(t, 1, buf);
}
static void
write_m_path_table(struct ecma119_write_target *t, uint8_t *buf)
{
write_path_table(t, 0, buf);
}
/* if file_id is >= 0, we use it instead of the filename. As a magic number,
* file_id == 3 means that we are writing the root directory record (in order
* to distinguish it from the "." entry in the root directory) */
static void
write_one_dir_record(struct ecma119_write_target *t,
struct ecma119_tree_node *node,
int file_id,
uint8_t *buf)
{
uint8_t len_dr = (file_id >= 0) ? 34 : node->dirent_len;
uint8_t len_fi = (file_id >= 0) ? 1 : strlen(node->name);
uint8_t f_id = (uint8_t) ((file_id == 3) ? 0 : file_id);
uint8_t *name = (file_id >= 0) ? &f_id : (uint8_t*)node->name;
uint32_t len = (node->type == ECMA119_DIR) ? node->dir.len
: node->file.real_me ? 0 : node->iso_self->attrib.st_size;
struct ecma119_dir_record *rec = (struct ecma119_dir_record*)buf;
/* we don't write out susp fields for the root node */
if (t->rockridge) {
if (file_id == 0) {
susp_write(t, &node->dir.self_susp, &buf[len_dr]);
len_dr += node->dir.self_susp.non_CE_len;
} else if (file_id == 1) {
susp_write(t, &node->dir.parent_susp, &buf[len_dr]);
len_dr += node->dir.parent_susp.non_CE_len;
} else if (file_id < 0) {
susp_write(t, &node->susp, &buf[len_dr]);
len_dr += node->susp.non_CE_len;
}
}
if (file_id == 1 && node->parent)
node = node->parent;
rec->len_dr[0] = len_dr;
iso_bb(rec->block, node->block, 4);
iso_bb(rec->length, len, 4);
iso_datetime_7(rec->recording_time, t->now);
rec->flags[0] = (node->type == ECMA119_DIR) ? 2 : 0;
iso_bb(rec->vol_seq_number, t->volnum + 1, 2);
rec->len_fi[0] = len_fi;
memcpy(rec->file_id, name, len_fi);
}
static void
write_one_dir(struct ecma119_write_target *t,
struct ecma119_tree_node *dir,
uint8_t *buf)
{
size_t i;
uint8_t *orig_buf = buf;
assert(dir->type == ECMA119_DIR);
/* write the "." and ".." entries first */
write_one_dir_record(t, dir, 0, buf);
buf += ((struct ecma119_dir_record*) buf)->len_dr[0];
write_one_dir_record(t, dir, 1, buf);
buf += ((struct ecma119_dir_record*) buf)->len_dr[0];
for (i = 0; i < dir->dir.nchildren; i++) {
write_one_dir_record(t, dir->dir.children[i], -1, buf);
buf += ((struct ecma119_dir_record*) buf)->len_dr[0];
}
/* write the susp continuation areas */
if (t->rockridge) {
susp_write_CE(t, &dir->dir.self_susp, buf);
buf += dir->dir.self_susp.CE_len;
susp_write_CE(t, &dir->dir.parent_susp, buf);
buf += dir->dir.parent_susp.CE_len;
for (i = 0; i < dir->dir.nchildren; i++) {
susp_write_CE(t, &dir->dir.children[i]->susp, buf);
buf += dir->dir.children[i]->susp.CE_len;
}
}
assert (buf - orig_buf == dir->dir.len + dir->dir.CE_len);
}
static void
write_dirs(struct ecma119_write_target *t, uint8_t *buf)
{
size_t i;
struct ecma119_tree_node *dir;
for (i = 0; i < t->dirlist_len; i++) {
dir = t->dirlist[i];
write_one_dir(t, dir, buf);
buf += round_up(dir->dir.len + dir->dir.CE_len, t->block_size);
}
}
void
ecma119_start_chunking(struct ecma119_write_target *t,
write_fn writer,
off_t data_size,
uint8_t *buf)
{
if (data_size != t->state_data_size) {
data_size = round_up(data_size, t->block_size);
t->state_data = realloc(t->state_data, data_size);
t->state_data_size = data_size;
}
memset(t->state_data, 0, t->state_data_size);
t->state_data_off = 0;
t->state_data_valid = 1;
writer(t, t->state_data);
write_data_chunk(t, buf);
}
static void
write_data_chunk(struct ecma119_write_target *t, uint8_t *buf)
{
memcpy(buf, t->state_data + t->state_data_off, t->block_size);
t->state_data_off += t->block_size;
if (t->state_data_off >= t->state_data_size) {
assert (t->state_data_off <= t->state_data_size);
t->state_data_valid = 0;
next_state(t);
}
}
static int
bs_read(struct burn_source *bs, unsigned char *buf, int size)
{
struct ecma119_write_target *t = (struct ecma119_write_target*)bs->data;
if (size != t->block_size) {
warnx("you must read data in block-sized chunks (%d bytes)",
(int)t->block_size);
return 0;
} else if (t->curblock >= t->vol_space_size) {
return 0;
}
if (t->state_data_valid)
write_data_chunk(t, buf);
else
writers[t->state](t, buf);
t->curblock++;
return size;
}
static off_t
bs_get_size(struct burn_source *bs)
{
struct ecma119_write_target *t = (struct ecma119_write_target*)bs->data;
return t->total_size;
}
static void
bs_free_data(struct burn_source *bs)
{
struct ecma119_write_target *t = (struct ecma119_write_target*)bs->data;
ecma119_tree_free(t->root);
free(t->dirlist);
free(t->pathlist);
free(t->dirlist_joliet);
free(t->pathlist_joliet);
free(t->filelist);
free(t->state_data);
if (t->state_files.fd)
fclose(t->state_files.fd);
}
struct burn_source *iso_source_new_ecma119(struct iso_volset *volset,
int volnum,
int level,
int flags)
{
struct burn_source *ret = calloc(1, sizeof(struct burn_source));
ret->refcount = 1;
ret->read = bs_read;
ret->get_size = bs_get_size;
ret->free_data = bs_free_data;
ret->data = ecma119_target_new(volset, volnum, level, flags);
return ret;
}

267
libisofs/ecma119.h Executable file
View File

@ -0,0 +1,267 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set noet ts=8 sts=8 sw=8 : */
/**
* \file ecma119.h
*
* Structures and definitions used for writing an emca119 (ISO9660) compatible
* volume.
*/
#ifndef LIBISO_ECMA119_H
#define LIBISO_ECMA119_H
#include <sys/time.h>
#include <stdint.h>
#include <stdio.h> /* for FILE */
#include <sys/types.h>
#include "susp.h"
struct ecma119_tree_node;
struct joliet_tree_node;
/**
* The possible states that the ecma119 writer can be in.
*/
enum ecma119_write_state
{
ECMA119_WRITE_BEFORE,
ECMA119_WRITE_SYSTEM_AREA,
ECMA119_WRITE_PRI_VOL_DESC,
ECMA119_WRITE_SUP_VOL_DESC_JOLIET,
ECMA119_WRITE_VOL_DESC_TERMINATOR,
ECMA119_WRITE_L_PATH_TABLE,
ECMA119_WRITE_M_PATH_TABLE,
ECMA119_WRITE_L_PATH_TABLE_JOLIET,
ECMA119_WRITE_M_PATH_TABLE_JOLIET,
ECMA119_WRITE_DIR_RECORDS,
ECMA119_WRITE_DIR_RECORDS_JOLIET,
ECMA119_WRITE_FILES,
ECMA119_WRITE_DONE
};
/**
* Data describing the state of the ecma119 writer. Everything here should be
* considered private!
*/
struct ecma119_write_target
{
struct ecma119_tree_node *root;
struct joliet_tree_node *joliet_root;
struct iso_volset *volset;
int volnum;
time_t now; /**< Time at which writing began. */
off_t total_size; /**< Total size of the output. This only
* includes the current volume. */
uint32_t vol_space_size;
unsigned int rockridge:1;
unsigned int joliet:1;
unsigned int iso_level:2;
int curblock;
uint16_t block_size;
uint32_t path_table_size;
uint32_t path_table_size_joliet;
uint32_t l_path_table_pos;
uint32_t m_path_table_pos;
uint32_t l_path_table_pos_joliet;
uint32_t m_path_table_pos_joliet;
uint32_t total_dir_size;
uint32_t total_dir_size_joliet;
struct ecma119_tree_node **dirlist;
/**< A pre-order list of directories
* (this is the order in which we write
* out directory records).
*/
struct ecma119_tree_node **pathlist;
/**< A breadth-first list of
* directories. This is used for
* writing out the path tables.
*/
size_t dirlist_len; /**< The length of the previous 2 lists.
*/
struct ecma119_tree_node **filelist;
/**< A pre-order list of files with
* non-NULL paths and non-zero sizes.
*/
size_t filelist_len; /* Length of the previous list. */
int curfile; /**< Used as a helper field for writing
out filelist and dirlist */
/* Joliet versions of the above lists. Since Joliet doesn't require
* directory relocation, the order of these lists might be different
* from the lists above (but they will be the same length).
*/
struct joliet_tree_node **dirlist_joliet;
struct joliet_tree_node **pathlist_joliet;
enum ecma119_write_state state; /* The current state of the writer. */
/* Most writers work by
* 1) making sure state_data is big enough for their data
* 2) writing _all_ their data into state_data
* 3) relying on write_data_chunk to write the data block
* by block.
*/
uint8_t *state_data;
off_t state_data_size;
off_t state_data_off;
int state_data_valid;
/* for writing out files */
struct state_files {
off_t pos; /* The number of bytes we have written
* so far in the current file.
*/
off_t data_len;/* The number of bytes in the currently
* open file.
*/
FILE *fd; /* The currently open file. */
int file; /* The index in filelist that we are
* currently writing (or about to write). */
} state_files;
};
/**
* Create a new ecma119_write_target from the given volume number of the
* given volume set.
*
* \pre \p volnum is less than \p volset-\>volset_size.
* \post For each node in the tree, writer_data has been allocated.
* \post The directory heirarchy has been reorganised to be ecma119-compatible.
*/
struct ecma119_write_target *ecma119_target_new(struct iso_volset *volset,
int volnum,
int level,
int flags);
#define BP(a,b) [(b) - (a) + 1]
struct ecma119_pri_vol_desc
{
uint8_t vol_desc_type BP(1, 1);
uint8_t std_identifier BP(2, 6);
uint8_t vol_desc_version BP(7, 7);
uint8_t unused1 BP(8, 8);
uint8_t system_id BP(9, 40);
uint8_t volume_id BP(41, 72);
uint8_t unused2 BP(73, 80);
uint8_t vol_space_size BP(81, 88);
uint8_t unused3 BP(89, 120);
uint8_t vol_set_size BP(121, 124);
uint8_t vol_seq_number BP(125, 128);
uint8_t block_size BP(129, 132);
uint8_t path_table_size BP(133, 140);
uint8_t l_path_table_pos BP(141, 144);
uint8_t opt_l_path_table_pos BP(145, 148);
uint8_t m_path_table_pos BP(149, 152);
uint8_t opt_m_path_table_pos BP(153, 156);
uint8_t root_dir_record BP(157, 190);
uint8_t vol_set_id BP(191, 318);
uint8_t publisher_id BP(319, 446);
uint8_t data_prep_id BP(447, 574);
uint8_t application_id BP(575, 702);
uint8_t copyright_file_id BP(703, 739);
uint8_t abstract_file_id BP(740, 776);
uint8_t bibliographic_file_id BP(777, 813);
uint8_t vol_creation_time BP(814, 830);
uint8_t vol_modification_time BP(831, 847);
uint8_t vol_expiration_time BP(848, 864);
uint8_t vol_effective_time BP(865, 881);
uint8_t file_structure_version BP(882, 882);
uint8_t reserved1 BP(883, 883);
uint8_t app_use BP(884, 1395);
uint8_t reserved2 BP(1396, 2048);
};
struct ecma119_sup_vol_desc
{
uint8_t vol_desc_type BP(1, 1);
uint8_t std_identifier BP(2, 6);
uint8_t vol_desc_version BP(7, 7);
uint8_t vol_flags BP(8, 8);
uint8_t system_id BP(9, 40);
uint8_t volume_id BP(41, 72);
uint8_t unused2 BP(73, 80);
uint8_t vol_space_size BP(81, 88);
uint8_t esc_sequences BP(89, 120);
uint8_t vol_set_size BP(121, 124);
uint8_t vol_seq_number BP(125, 128);
uint8_t block_size BP(129, 132);
uint8_t path_table_size BP(133, 140);
uint8_t l_path_table_pos BP(141, 144);
uint8_t opt_l_path_table_pos BP(145, 148);
uint8_t m_path_table_pos BP(149, 152);
uint8_t opt_m_path_table_pos BP(153, 156);
uint8_t root_dir_record BP(157, 190);
uint8_t vol_set_id BP(191, 318);
uint8_t publisher_id BP(319, 446);
uint8_t data_prep_id BP(447, 574);
uint8_t application_id BP(575, 702);
uint8_t copyright_file_id BP(703, 739);
uint8_t abstract_file_id BP(740, 776);
uint8_t bibliographic_file_id BP(777, 813);
uint8_t vol_creation_time BP(814, 830);
uint8_t vol_modification_time BP(831, 847);
uint8_t vol_expiration_time BP(848, 864);
uint8_t vol_effective_time BP(865, 881);
uint8_t file_structure_version BP(882, 882);
uint8_t reserved1 BP(883, 883);
uint8_t app_use BP(884, 1395);
uint8_t reserved2 BP(1396, 2048);
};
struct ecma119_vol_desc_terminator
{
uint8_t vol_desc_type BP(1, 1);
uint8_t std_identifier BP(2, 6);
uint8_t vol_desc_version BP(7, 7);
uint8_t reserved BP(8, 2048);
};
struct ecma119_dir_record
{
uint8_t len_dr BP(1, 1);
uint8_t len_xa BP(2, 2);
uint8_t block BP(3, 10);
uint8_t length BP(11, 18);
uint8_t recording_time BP(19, 25);
uint8_t flags BP(26, 26);
uint8_t file_unit_size BP(27, 27);
uint8_t interleave_gap_size BP(28, 28);
uint8_t vol_seq_number BP(29, 32);
uint8_t len_fi BP(33, 33);
uint8_t file_id BP(34, 34); /* 34 to 33+len_fi */
/* padding field (if len_fi is even) */
/* system use (len_dr - len_su + 1 to len_dr) */
};
struct ecma119_path_table_record
{
uint8_t len_di BP(1, 1);
uint8_t len_xa BP(2, 2);
uint8_t block BP(3, 6);
uint8_t parent BP(7, 8);
uint8_t dir_id BP(9, 9); /* 9 to 8+len_di */
/* padding field (if len_di is odd) */
};
/**
* A utility function for writers that want to write their data all at once
* rather than block-by-block. This creates a buffer of size \p size, passes
* it to the given writer, then hands out block-sized chunks.
*/
void
ecma119_start_chunking(struct ecma119_write_target *t,
void (*)(struct ecma119_write_target*, uint8_t*),
off_t size,
uint8_t *buf);
#endif /* LIBISO_ECMA119_H */

312
libisofs/ecma119_tree.c Normal file
View File

@ -0,0 +1,312 @@
/* vim: set noet ts=8 sts=8 sw=8 : */
#include <string.h>
#include <wchar.h>
#include <stdlib.h>
#include <assert.h>
#include "ecma119.h"
#include "ecma119_tree.h"
#include "tree.h"
#include "util.h"
static size_t calc_dirent_len(struct ecma119_tree_node *n)
{
int ret = n->name ? strlen(n->name) + 33 : 34;
if (ret % 2) ret++;
return ret;
}
static struct ecma119_tree_node*
create_dir(struct ecma119_write_target *t,
struct ecma119_tree_node *parent,
struct iso_tree_node *iso)
{
struct ecma119_tree_node *ret;
assert(t && (!parent || parent->type == ECMA119_DIR)
&& iso && S_ISDIR(iso->attrib.st_mode));
ret = calloc(1, sizeof(struct ecma119_tree_node));
ret->name = iso->name ? ((t->iso_level == 1) ? iso_1_dirid(iso->name)
: iso_2_dirid(iso->name))
: NULL;
ret->dirent_len = calc_dirent_len(ret);
ret->iso_self = iso;
ret->target = t;
ret->type = ECMA119_DIR;
ret->parent = ret->dir.real_parent = parent;
ret->dir.depth = parent ? parent->dir.depth + 1 : 1;
ret->dir.nchildren = iso->nchildren;
ret->dir.children = calloc(1, sizeof(void*) * iso->nchildren);
return ret;
}
static struct ecma119_tree_node*
create_file(struct ecma119_write_target *t,
struct ecma119_tree_node *parent,
struct iso_tree_node *iso)
{
struct ecma119_tree_node *ret;
assert(t && iso && parent && parent->type == ECMA119_DIR);
ret = calloc(1, sizeof(struct ecma119_tree_node));
ret->name = iso->name ? ((t->iso_level == 1) ? iso_1_fileid(iso->name)
: iso_2_fileid(iso->name))
: NULL;
ret->dirent_len = calc_dirent_len(ret);
ret->parent = parent;
ret->iso_self = iso;
ret->target = t;
ret->type = ECMA119_FILE;
return ret;
}
static struct ecma119_tree_node*
create_tree(struct ecma119_write_target *t,
struct ecma119_tree_node *parent,
struct iso_tree_node *iso)
{
struct ecma119_tree_node *ret;
size_t i;
assert(t && iso);
ret = (S_ISDIR(iso->attrib.st_mode) ? create_dir : create_file)
(t, parent, iso);
for (i = 0; i < iso->nchildren; i++) {
ret->dir.children[i] = create_tree(t, ret, iso->children[i]);
}
return ret;
}
void
ecma119_tree_free(struct ecma119_tree_node *root)
{
size_t i;
if (root->type == ECMA119_DIR) {
for (i=0; i < root->dir.nchildren; i++) {
ecma119_tree_free(root->dir.children[i]);
}
free(root->dir.children);
}
free(root->name);
free(root);
}
static size_t
max_child_name_len(struct ecma119_tree_node *root)
{
size_t ret = 0, i;
assert(root->type == ECMA119_DIR);
for (i=0; i < root->dir.nchildren; i++) {
size_t len = strlen(root->dir.children[i]->name);
ret = MAX(ret, len);
}
return ret;
}
static void
reparent(struct ecma119_tree_node *child,
struct ecma119_tree_node *parent)
{
int found = 0;
size_t i;
struct ecma119_tree_node *placeholder;
assert(child && parent && parent->type == ECMA119_DIR && child->parent);
/* replace the child in the original parent with a placeholder */
for (i=0; i < child->parent->dir.nchildren; i++) {
if (child->parent->dir.children[i] == child) {
placeholder = create_file(child->target,
child->parent,
child->iso_self);
placeholder->file.real_me = child;
child->parent->dir.children[i] = placeholder;
found = 1;
break;
}
}
assert(found);
/* add the child to its new parent */
child->parent = parent;
parent->dir.nchildren++;
parent->dir.children = realloc( parent->dir.children,
sizeof(void*) * parent->dir.nchildren );
parent->dir.children[parent->dir.nchildren-1] = child;
}
/**
* Reorder the tree, if necessary, to ensure that
* - the depth is at most 8
* - each path length is at most 255 characters
*/
static void
reorder_tree(struct ecma119_tree_node *root,
struct ecma119_tree_node *cur)
{
size_t max_path;
assert(root && cur && cur->type == ECMA119_DIR);
cur->dir.depth = cur->parent ? cur->parent->dir.depth + 1 : 1;
cur->dir.path_len = cur->parent ? cur->parent->dir.path_len
+ strlen(cur->name) : 0;
max_path = cur->dir.path_len + cur->dir.depth + max_child_name_len(cur);
if (cur->dir.depth > 8 || max_path > 255) {
reparent(cur, root);
/* we are appended to the root's children now, so there is no
* need to recurse (the root will hit us again) */
} else {
size_t i;
for (i=0; i < cur->dir.nchildren; i++) {
if (cur->dir.children[i]->type == ECMA119_DIR)
reorder_tree(root, cur->dir.children[i]);
}
}
}
static int
cmp_node(const void *f1, const void *f2)
{
struct ecma119_tree_node *f = *((struct ecma119_tree_node**)f1);
struct ecma119_tree_node *g = *((struct ecma119_tree_node**)f2);
return strcmp(f->name, g->name);
}
static void
sort_tree(struct ecma119_tree_node *root)
{
size_t i;
assert(root && root->type == ECMA119_DIR);
qsort(root->dir.children, root->dir.nchildren, sizeof(void*), cmp_node);
for (i=0; i < root->dir.nchildren; i++) {
if (root->dir.children[i]->type == ECMA119_DIR)
sort_tree(root->dir.children[i]);
}
}
/**
* Change num_change characters of the given filename in order to ensure the
* name is unique. If the name is short enough (depending on the ISO level),
* we can append the characters instead of changing them.
*
* \p seq_num is the index of this file in the sequence of identical filenames.
*
* For example, seq_num=3, num_change=2, name="HELLOTHERE.TXT" changes name to
* "HELLOTHE03.TXT"
*/
static void
mangle_name(char **name, int num_change, int level, int seq_num)
{
char *dot = strrchr(*name, '.');
char *semi = strrchr(*name, ';');
size_t len = strlen(*name);
char base[len+1], ext[len+1];
char fmt[12];
size_t baselen, extlen;
if (num_change >= len) {
return;
}
strncpy(base, *name, len+1);
if (dot) {
base[dot - *name] = '\0';
strncpy(ext, dot+1, len+1);
if (semi) {
ext[semi - dot - 1] = '\0';
}
} else {
base[len-2] = '\0';
ext[0] = '\0';
}
baselen = strlen(base);
extlen = strlen(ext);
if (level == 1 && baselen + num_change > 8) {
base[8 - num_change] = '\0';
} else if (level != 1 && baselen + extlen + num_change > 30) {
base[30 - extlen - num_change] = '\0';
}
sprintf(fmt, "%%s%%0%1dd.%%s;1", num_change);
*name = realloc(*name, baselen + extlen + num_change + 4);
sprintf(*name, fmt, base, seq_num, ext);
}
static void
mangle_all(struct ecma119_tree_node *dir)
{
size_t i, j, k;
struct ecma119_dir_info d = dir->dir;
size_t n_change;
int changed;
assert(dir->type == ECMA119_DIR);
do {
changed = 0;
for (i=0; i < d.nchildren; i++) {
/* find the number of consecutive equal names */
j = 1;
while ( i+j < d.nchildren &&
!strcmp(d.children[i]->name,
d.children[i+j]->name) )
j++;
if (j == 1) continue;
/* mangle the names */
changed = 1;
n_change = j / 10 + 1;
for (k=0; k < j; k++) {
mangle_name(&(d.children[i+k]->name),
n_change,
dir->target->iso_level,
k);
d.children[i+k]->dirent_len =
calc_dirent_len(d.children[i+k]);
}
/* skip ahead by the number of mangled names */
i += j - 1;
}
} while (changed);
for (i=0; i < d.nchildren; i++) {
if (d.children[i]->type == ECMA119_DIR)
mangle_all(d.children[i]);
}
}
struct ecma119_tree_node*
ecma119_tree_create(struct ecma119_write_target *t,
struct iso_tree_node *iso_root)
{
t->root = create_tree(t, NULL, iso_root);
reorder_tree(t->root, t->root);
sort_tree(t->root);
mangle_all(t->root);
return t->root;
}
void
ecma119_tree_print(struct ecma119_tree_node *root, int spaces)
{
size_t i;
char sp[spaces+1];
memset(sp, ' ', spaces);
sp[spaces] = '\0';
printf("%s%s\n", sp, root->name);
if (root->type == ECMA119_DIR)
for (i=0; i < root->dir.nchildren; i++)
ecma119_tree_print(root->dir.children[i], spaces+2);
}

95
libisofs/ecma119_tree.h Normal file
View File

@ -0,0 +1,95 @@
/* vim: set noet ts=8 sts=8 sw=8 : */
/**
* \file ecma119_tree.h
*
* Declarations for creating, modifying and printing filesystem trees that
* are compatible with ecma119.
*/
#ifndef LIBISO_ECMA119_TREE_H
#define LIBISO_ECMA119_TREE_H
struct ecma119_write_target;
enum {
ECMA119_FILE,
ECMA119_DIR
};
struct ecma119_dir_info {
struct susp_info self_susp; /**< susp entries for "." */
struct susp_info parent_susp; /**< susp entries for ".." */
size_t len; /**< sum of the lengths of children's
* Directory Records (including SU) */
size_t CE_len; /**< sum of the lengths of children's
* SUSP CE areas */
int depth;
size_t path_len; /**< The length of a path up to, and
* including, this directory. This
* cannot exceed 255. */
size_t nchildren;
struct ecma119_tree_node **children;
struct ecma119_tree_node *real_parent;
/**< The parent before relocation */
};
struct ecma119_file_info
{
struct ecma119_tree_node *real_me;
/**< If this is non-NULL, the file is
* a placeholder for a relocated
* directory and this field points to
* that relocated directory.
*/
};
/**
* A node for a tree containing all the information necessary for writing
* an ISO9660 volume.
*/
struct ecma119_tree_node
{
char *name; /**< in ASCII, conforming to the
* current ISO level. */
size_t dirent_len; /**< Length of the directory record,
* not including SU. */
size_t block;
struct ecma119_tree_node *parent;
struct iso_tree_node *iso_self;
struct ecma119_write_target *target;
struct susp_info susp;
int type; /**< file or directory */
/* union {*/
struct ecma119_dir_info dir;
struct ecma119_file_info file;
/* };*/
};
/**
* Create a new ecma119_tree that corresponds to the tree represented by
* \p iso_root.
*/
struct ecma119_tree_node*
ecma119_tree_create(struct ecma119_write_target *target,
struct iso_tree_node *iso_root);
/**
* Free an ecma119 tree.
*/
void
ecma119_tree_free(struct ecma119_tree_node *root);
/**
* Print an ecma119 tree.
*/
void
ecma119_tree_print(struct ecma119_tree_node *root, int spaces);
#endif /* LIBISO_ECMA119_TREE_H */

42
libisofs/exclude.c Normal file
View File

@ -0,0 +1,42 @@
#include "hash.h"
#include "exclude.h"
static struct iso_hash_node *table[HASH_NODES]={0,};
static int num=0;
void
iso_exclude_add_path(const char *path)
{
if (!path)
return;
num += iso_hash_insert(table, path);
}
void
iso_exclude_remove_path(const char *path)
{
if (!num || !path)
return;
num -= iso_hash_remove(table, path);
}
void
iso_exclude_empty(void)
{
if (!num)
return;
iso_hash_empty(table);
num=0;
}
int
iso_exclude_lookup(const char *path)
{
if (!num || !path)
return 0;
return iso_hash_lookup(table, path);
}

12
libisofs/exclude.h Normal file
View File

@ -0,0 +1,12 @@
#ifndef ISO_EXCLUDE_H
#define ISO_EXCLUDE_H
/**
* Add a path to ignore when adding a directory recursively.
*
* \param path The path, on the local filesystem, of the file.
*/
int
iso_exclude_lookup(const char *path);
#endif /* ISO_EXCLUDE */

158
libisofs/hash.c Normal file
View File

@ -0,0 +1,158 @@
#include <stdlib.h>
#include <string.h>
#include "hash.h"
static unsigned int
iso_hash_path(const char *path)
{
unsigned int hash_num=0;
const char *c;
c=path;
while(*c)
hash_num = (hash_num << 15) + (hash_num << 3) + (hash_num >> 3) + *c++;
return hash_num % HASH_NODES;
}
int
iso_hash_lookup(struct iso_hash_node **table, const char *path)
{
struct iso_hash_node *node;
unsigned int hash_num;
hash_num = iso_hash_path(path);
node=table[hash_num];
if (!node)
return 0;
if (!strcmp(path, node->path))
return 1;
while (node->next) {
node=node->next;
if (!strcmp(path, node->path))
return 1;
}
return 0;
}
static struct iso_hash_node*
iso_hash_node_new (const char *path)
{
struct iso_hash_node *node;
/*create an element to be inserted in the hash table */
node=malloc(sizeof(struct iso_hash_node));
node->path=strdup(path);
node->next=NULL;
return node;
}
int
iso_hash_insert(struct iso_hash_node **table, const char *path)
{
struct iso_hash_node *node;
unsigned int hash_num;
/* find the hash number */
hash_num = iso_hash_path(path);
/* insert it */
node = table[hash_num];
/* unfortunately, we can't safely consider that a path
* won't be twice in the hash table so make sure it
* doesn't already exists */
if (!node) {
table[hash_num]=iso_hash_node_new(path);
return 1;
}
/* if it's already in, we don't do anything */
if (!strcmp(path, node->path))
return 0;
while (node->next) {
node = node->next;
/* if it's already in, we don't do anything */
if (!strcmp (path, node->path))
return 0;
}
node->next = iso_hash_node_new(path);
return 1;
}
static void
iso_hash_node_free(struct iso_hash_node *node)
{
free(node->path);
free(node);
}
int
iso_hash_remove(struct iso_hash_node **table, const char *path)
{
unsigned int hash_num;
struct iso_hash_node *node;
hash_num = iso_hash_path(path);
node=table[hash_num];
if (!node)
return 0;
if (!strcmp(path, node->path)) {
table[hash_num]=node->next;
iso_hash_node_free(node);
return 1;
}
while (node->next) {
struct iso_hash_node *prev;
prev = node;
node = node->next;
if (!strcmp (path, node->path)) {
prev->next=node->next;
iso_hash_node_free(node);
return 1;
}
}
return 0;
}
void
iso_hash_empty(struct iso_hash_node **table)
{
int i;
for (i=0; i < HASH_NODES; i++) {
struct iso_hash_node *node;
node=table[i];
if (!node)
continue;
table[i]=NULL;
do {
struct iso_hash_node *next;
next=node->next;
iso_hash_node_free(node);
node=next;
} while (node);
}
}

46
libisofs/hash.h Normal file
View File

@ -0,0 +1,46 @@
#ifndef ISO_HASH_H
#define ISO_HASH_H
struct iso_hash_node {
struct iso_hash_node *next;
char *path;
};
#define HASH_NODES 128
/**
* Searches in the hash table if the path exists.
*
* \param table The hash table.
* \param path The path of the file to look for.
*
* \return 1 if the path exists in the hash table, 0 otherwise.
*/
int iso_hash_lookup(struct iso_hash_node **table, const char *path);
/**
* Insert a new path in the hash table.
*
* \param table The hash table.
* \param path The path of a file to add to the hash table.
*
* \return 1 if the file wasn't already in the hash table, 0 otherwise.
*/
int iso_hash_insert(struct iso_hash_node **table, const char *path);
/**
* Remove a path from the hash table.
*
* \param table The hash table.
* \param path The path of a file to remove from the hash table.
*
* \return 1 if the file was found and removed, 0 otherwise.
*/
int iso_hash_remove(struct iso_hash_node **table, const char *path);
/**
* Empty the hash table.
*/
void iso_hash_empty(struct iso_hash_node **table);
#endif /* ISO_HASH_H */

379
libisofs/joliet.c Normal file
View File

@ -0,0 +1,379 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set noet ts=8 sts=8 sw=8 : */
#include "joliet.h"
#include "ecma119.h"
#include "ecma119_tree.h"
#include "tree.h"
#include "util.h"
#include "volume.h"
#include <assert.h>
#include <string.h>
static struct joliet_tree_node*
create_node(struct ecma119_write_target *t,
struct joliet_tree_node *parent,
struct iso_tree_node *iso)
{
struct joliet_tree_node *ret =
calloc(1, sizeof(struct joliet_tree_node));
ret->name = iso_j_id(iso->name);
ret->dirent_len = 34 + (ret->name ? ucslen(ret->name) * 2 : 0);
ret->len = iso->attrib.st_size; /* for dirs, we'll change this */
ret->block = iso->block; /* only actually for files, not dirs */
ret->parent = parent;
ret->iso_self = iso;
ret->target = t;
ret->nchildren = iso->nchildren;
if (ret->nchildren)
ret->children = calloc(sizeof(void*), ret->nchildren);
return ret;
}
static struct joliet_tree_node*
create_tree(struct ecma119_write_target *t,
struct joliet_tree_node *parent,
struct iso_tree_node *iso_root)
{
struct joliet_tree_node *root = create_node(t, parent, iso_root);
size_t i;
for (i = 0; i < root->nchildren; i++) {
struct iso_tree_node *iso_ch = iso_root->children[i];
if (ISO_ISDIR(iso_ch))
root->children[i] = create_tree(t, root, iso_ch);
else
root->children[i] = create_node(t, root, iso_ch);
}
return root;
}
static int
cmp_node(const void *f1, const void *f2)
{
struct joliet_tree_node *f = *((struct joliet_tree_node**)f1);
struct joliet_tree_node *g = *((struct joliet_tree_node**)f2);
return ucscmp(f->name, g->name);
}
static void
sort_tree(struct joliet_tree_node *root)
{
size_t i;
assert(root && ISO_ISDIR(root->iso_self));
qsort(root->children, root->nchildren, sizeof(void*), cmp_node);
for (i = 0; i < root->nchildren; i++)
if (ISO_ISDIR(root->children[i]->iso_self))
sort_tree(root->children[i]);
}
void
joliet_prepare_path_tables(struct ecma119_write_target *t)
{
size_t cur, i, j;
t->pathlist_joliet[0] = t->joliet_root;
t->path_table_size_joliet = 10; /* root directory record */
cur = 1;
for (i = 0; i < t->dirlist_len; i++) {
struct joliet_tree_node *dir = t->pathlist_joliet[i];
for (j = 0; j < dir->nchildren; j++) {
struct joliet_tree_node *ch = dir->children[j];
if (ISO_ISDIR(ch->iso_self)) {
size_t len = 8 + ucslen(ch->name)*2;
t->pathlist_joliet[cur++] = ch;
t->path_table_size_joliet += len;
}
}
}
}
/**
* Calculate the size of each directory.
*/
void
joliet_calc_dir_size(struct ecma119_write_target *t,
struct joliet_tree_node *root)
{
size_t i;
struct joliet_tree_node *ch;
assert(root && ISO_ISDIR(root->iso_self));
root->len = 68; /* for "." and ".." entries */
for (i = 0; i < root->nchildren; i++) {
ch = root->children[i];
root->len += ch->dirent_len;
if (ISO_ISDIR(ch->iso_self))
joliet_calc_dir_size(t, ch);
}
t->total_dir_size_joliet += round_up (root->len, t->block_size);
}
/**
* Calculate the position of each directory. Also fill out t->dirlist_joliet.
*/
void
joliet_calc_dir_pos(struct ecma119_write_target *t,
struct joliet_tree_node *root)
{
size_t i;
struct joliet_tree_node *ch;
assert(root && ISO_ISDIR(root->iso_self));
root->block = t->curblock;
t->curblock += div_up(root->len, t->block_size);
t->dirlist_joliet[t->curfile++] = root;
for (i = 0; i < root->nchildren; i++) {
ch = root->children[i];
if (ISO_ISDIR(ch->iso_self))
joliet_calc_dir_pos(t, ch);
}
/* reset curfile when we're finished */
if (!root->parent)
t->curfile = 0;
}
void
joliet_update_file_pos(struct ecma119_write_target *t,
struct joliet_tree_node *dir)
{
size_t i;
assert(dir && ISO_ISDIR(dir->iso_self));
for (i = 0; i < dir->nchildren; i++) {
struct joliet_tree_node *ch;
ch = dir->children[i];
if (!ISO_ISDIR (ch->iso_self)) {
struct iso_tree_node *iso = ch->iso_self;
ch->block = iso->block;
}
else
joliet_update_file_pos(t, ch);
}
/* reset curfile when we're finished */
if (!dir->parent)
t->curfile = 0;
}
struct joliet_tree_node*
joliet_tree_create(struct ecma119_write_target *t,
struct iso_tree_node *iso_root)
{
struct joliet_tree_node *root = create_tree(t, NULL, iso_root);
sort_tree(root);
return root;
}
/* ugh. this is mostly C&P */
static void
write_path_table(struct ecma119_write_target *t,
int l_type,
uint8_t *buf)
{
void (*write_int)(uint8_t*, uint32_t, int) = l_type ?
iso_lsb : iso_msb;
size_t i;
struct ecma119_path_table_record *rec;
struct joliet_tree_node *dir;
int parent = 0;
assert (t->joliet);
for (i = 0; i < t->dirlist_len; i++) {
dir = t->pathlist_joliet[i];
while ((i) && t->pathlist_joliet[parent] != dir->parent)
parent++;
assert(parent < i || i == 0);
rec = (struct ecma119_path_table_record*) buf;
rec->len_di[0] = dir->parent ?
(uint8_t) ucslen(dir->name) * 2 : 1;
rec->len_xa[0] = 0;
write_int(rec->block, dir->block, 4);
write_int(rec->parent, parent + 1, 2);
if (dir->parent)
memcpy(rec->dir_id, dir->name, rec->len_di[0]);
buf += 8 + rec->len_di[0] + (rec->len_di[0] % 2);
}
}
/* if file_id is >= 0, we use it instead of the filename. As a magic number,
* file_id == 3 means that we are writing the root directory record (in order
* to distinguish it from the "." entry in the root directory) */
static void
write_one_dir_record(struct ecma119_write_target *t,
struct joliet_tree_node *node,
int file_id,
uint8_t *buf)
{
uint8_t len_dr = (file_id >= 0) ? 34 : node->dirent_len;
uint8_t len_fi = (file_id >= 0) ? 1 : ucslen(node->name) * 2;
uint8_t f_id = (uint8_t) ((file_id == 3) ? 0 : file_id);
uint8_t *name = (file_id >= 0) ? &f_id : (uint8_t*)node->name;
struct ecma119_dir_record *rec = (struct ecma119_dir_record*)buf;
if (file_id == 1 && node->parent)
node = node->parent;
rec->len_dr[0] = len_dr;
iso_bb(rec->block, node->block, 4);
iso_bb(rec->length, node->len, 4);
iso_datetime_7(rec->recording_time, t->now);
rec->flags[0] = ISO_ISDIR(node->iso_self) ? 2 : 0;
iso_bb(rec->vol_seq_number, t->volnum + 1, 2);
rec->len_fi[0] = len_fi;
memcpy(rec->file_id, name, len_fi);
}
static void
write_l_path_table(struct ecma119_write_target *t, uint8_t *buf)
{
write_path_table (t, 1, buf);
}
static void
write_m_path_table(struct ecma119_write_target *t, uint8_t *buf)
{
write_path_table (t, 0, buf);
}
static void
write_sup_vol_desc(struct ecma119_write_target *t, uint8_t *buf)
{
struct ecma119_sup_vol_desc *vol = (struct ecma119_sup_vol_desc*)buf;
struct iso_volume *volume = t->volset->volume[t->volnum];
uint16_t *vol_id = str2ucs(volume->volume_id);
uint16_t *pub_id = str2ucs(volume->publisher_id);
uint16_t *data_id = str2ucs(volume->data_preparer_id);
uint16_t *volset_id = str2ucs(t->volset->volset_id);
int vol_id_len = MIN(32, ucslen(vol_id) * 2);
int pub_id_len = MIN(128, ucslen(pub_id) * 2);
int data_id_len = MIN(128, ucslen(data_id) * 2);
int volset_id_len = MIN(128, ucslen(volset_id) * 2);
vol->vol_desc_type[0] = 2;
memcpy(vol->std_identifier, "CD001", 5);
vol->vol_desc_version[0] = 1;
memcpy(vol->system_id, "SYSID", 5);
if (vol_id)
memcpy(vol->volume_id, vol_id, vol_id_len);
memcpy(vol->esc_sequences, "%/E", 3);
iso_bb(vol->vol_space_size, t->vol_space_size, 4);
iso_bb(vol->vol_set_size, t->volset->volset_size, 2);
iso_bb(vol->vol_seq_number, t->volnum + 1, 2);
iso_bb(vol->block_size, t->block_size, 2);
iso_bb(vol->path_table_size, t->path_table_size_joliet, 4);
iso_lsb(vol->l_path_table_pos, t->l_path_table_pos_joliet, 4);
iso_msb(vol->m_path_table_pos, t->m_path_table_pos_joliet, 4);
write_one_dir_record(t, t->joliet_root, 3, vol->root_dir_record);
memcpy(vol->vol_set_id, volset_id, volset_id_len);
memcpy(vol->publisher_id, pub_id, pub_id_len);
memcpy(vol->data_prep_id, data_id, data_id_len);
/*memcpy(vol->application_id, "APPID", app_id_len);*/
iso_datetime_17(vol->vol_creation_time, t->now);
iso_datetime_17(vol->vol_modification_time, t->now);
iso_datetime_17(vol->vol_effective_time, t->now);
vol->file_structure_version[0] = 1;
free(vol_id);
free(volset_id);
free(pub_id);
free(data_id);
}
static void
write_one_dir(struct ecma119_write_target *t,
struct joliet_tree_node *dir,
uint8_t *buf)
{
size_t i;
uint8_t *orig_buf = buf;
assert(ISO_ISDIR (dir->iso_self));
/* write the "." and ".." entries first */
write_one_dir_record(t, dir, 0, buf);
buf += ((struct ecma119_dir_record*) buf)->len_dr[0];
write_one_dir_record(t, dir, 1, buf);
buf += ((struct ecma119_dir_record*) buf)->len_dr[0];
for (i = 0; i < dir->nchildren; i++) {
write_one_dir_record(t, dir->children[i], -1, buf);
buf += ((struct ecma119_dir_record*) buf)->len_dr[0];
}
assert (buf - orig_buf == dir->len);
}
static void
write_dirs(struct ecma119_write_target *t, uint8_t *buf)
{
size_t i;
struct joliet_tree_node *dir;
assert (t->curblock == t->dirlist_joliet[0]->block);
for (i = 0; i < t->dirlist_len; i++) {
dir = t->dirlist_joliet[i];
write_one_dir(t, dir, buf);
buf += round_up(dir->len, t->block_size);
}
}
void
joliet_wr_sup_vol_desc(struct ecma119_write_target *t,
uint8_t *buf)
{
ecma119_start_chunking(t,
write_sup_vol_desc,
2048,
buf);
}
void
joliet_wr_l_path_table(struct ecma119_write_target *t,
uint8_t *buf)
{
ecma119_start_chunking(t,
write_l_path_table,
t->path_table_size_joliet,
buf);
}
void
joliet_wr_m_path_table(struct ecma119_write_target *t,
uint8_t *buf)
{
ecma119_start_chunking(t,
write_m_path_table,
t->path_table_size_joliet,
buf);
}
void
joliet_wr_dir_records(struct ecma119_write_target *t,
uint8_t *buf)
{
ecma119_start_chunking(t,
write_dirs,
t->total_dir_size_joliet,
buf);
}

84
libisofs/joliet.h Normal file
View File

@ -0,0 +1,84 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set noet ts=8 sts=8 sw=8 : */
/**
* \file joliet.h
*
* Declare the filesystems trees that are Joliet-compatible and the public
* functions for tying them into an ecma119 volume.
*/
#ifndef LIBISO_JOLIET_H
#define LIBISO_JOLIET_H
#include <stdint.h>
#include <stdlib.h>
struct ecma119_write_target;
struct iso_tree_node;
struct joliet_tree_node
{
uint16_t *name; /**< In UCS-2BE. */
size_t dirent_len;
size_t len;
size_t block;
struct joliet_tree_node *parent;
struct iso_tree_node *iso_self;
struct ecma119_write_target *target;
struct joliet_tree_node **children;
size_t nchildren;
};
/**
* Create a new joliet_tree that corresponds to the tree represented by
* \p iso_root.
*/
struct joliet_tree_node*
joliet_tree_create(struct ecma119_write_target *target,
struct iso_tree_node *iso_root);
/**
* Calculate the size of each directory in the joliet heirarchy.
*/
void
joliet_calc_dir_size(struct ecma119_write_target *t, struct joliet_tree_node*);
/**
* Calculate the position of each directory in the joliet heirarchy.
*/
void
joliet_calc_dir_pos(struct ecma119_write_target *t, struct joliet_tree_node*);
/**
* Update the position of each file in the joliet hierarchy (to be called
* AFTER the file positions in the iso tree have been set).
*/
void
joliet_update_file_pos(struct ecma119_write_target *t, struct joliet_tree_node*);
/**
* Calculate the size of the joliet path table and fill in the list of
* directories.
*/
void
joliet_prepare_path_tables(struct ecma119_write_target *t);
void
joliet_tree_free(struct joliet_tree_node *root);
void
joliet_wr_sup_vol_desc(struct ecma119_write_target *t, uint8_t *buf);
void
joliet_wr_l_path_table(struct ecma119_write_target *t, uint8_t *buf);
void
joliet_wr_m_path_table(struct ecma119_write_target *t, uint8_t *buf);
void
joliet_wr_dir_records(struct ecma119_write_target *t, uint8_t *buf);
#endif /* LIBISO_JOLIET_H */

225
libisofs/libisofs.h Executable file
View File

@ -0,0 +1,225 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set noet ts=8 sts=8 sw=8 : */
/**
* Create an ISO-9660 data volume with Rock Ridge and Joliet extensions.
* Usage is easy:
* - Create a new volume.
* - Add files and directories.
* - Write the volume to a file or create a burn source for use with Libburn.
*/
#ifndef LIBISO_LIBISOFS_H
#define LIBISO_LIBISOFS_H
/* #include <libburn.h> */
struct burn_source;
/**
* Data volume.
* @see volume.h for details.
*/
struct iso_volume;
/**
* A set of data volumes.
* @see volume.h for details.
*/
struct iso_volset;
/**
* A node in the filesystem tree.
* \see tree.h
*/
struct iso_tree_node;
enum ecma119_extension_flag {
ECMA119_ROCKRIDGE = (1<<0),
ECMA119_JOLIET = (1<<1)
};
/**
* Create a new volume.
* The parameters can be set to NULL if you wish to set them later.
*/
struct iso_volume *iso_volume_new(const char *volume_id,
const char *publisher_id,
const char *data_preparer_id);
struct iso_volume *iso_volume_new_with_root(const char *volume_id,
const char *publisher_id,
const char *data_preparer_id,
struct iso_tree_node *root);
/**
* Free a volume.
*/
void iso_volume_free(struct iso_volume *volume);
/**
* Free a set of data volumes.
*/
void iso_volset_free(struct iso_volset *volume);
/**
* Get the root directory for a volume.
*/
struct iso_tree_node *iso_volume_get_root(const struct iso_volume *volume);
/**
* Fill in the volume identifier for a volume.
*/
void iso_volume_set_volume_id(struct iso_volume *volume,
const char *volume_id);
/**
* Fill in the publisher for a volume.
*/
void iso_volume_set_publisher_id(struct iso_volume *volume,
const char *publisher_id);
/**
* Fill in the data preparer for a volume.
*/
void iso_volume_set_data_preparer_id(struct iso_volume *volume,
const char *data_preparer_id);
/**
* Locate a node by its path on disc.
*
* \param volume The volume to search in.
* \param path The path, in the image, of the file.
*
* \return The node found or NULL.
*
*/
struct iso_tree_node *iso_tree_volume_path_to_node(struct iso_volume *volume, const char *path);
/**
* Add a file or a directory (recursively) to a volume by specifying its path on the volume.
*
* \param volume The volume to add the file to.
* \param disc_path The path on the disc at which to add the disc.
* \param path The path, on the local filesystem, of the file.
*
* \return The node for the file or NULL if the parent doesn't exists on the disc.
*/
struct iso_tree_node *iso_tree_volume_add_path(struct iso_volume *volume,
const char *disc_path,
const char *path);
/**
* Creates a new, empty directory on the volume.
*
* \param volume The volume to add the directory to.
* \param disc_path The path on the volume at which to add the directory.
*
* \return A pointer to the newly created directory.
*/
struct iso_tree_node *iso_tree_volume_add_new_dir(struct iso_volume *volume,
const char *disc_path);
/**
* Create a new Volume Set consisting of only one volume.
* @param volume The first and only volume for the volset to contain.
* @param volset_id The Volume Set ID.
* @return A new iso_volset.
*/
struct iso_volset *iso_volset_new(struct iso_volume *volume,
const char *volset_id);
/**
* Add a file to a directory.
*
* \param path The path, on the local filesystem, of the file.
*
* \pre \p parent is NULL or is a directory.
* \pre \p path is non-NULL and is a valid path to a non-directory on the local
* filesystem.
* \return An iso_tree_node whose path is \p path and whose parent is \p parent.
*/
struct iso_tree_node *iso_tree_add_node(struct iso_tree_node *parent,
const char *path);
/**
* Recursively add an existing directory to the tree.
* Warning: when using this, you'll lose pointers to files or subdirectories.
* If you want to have pointers to all files and directories,
* use iso_tree_add_file and iso_tree_add_dir.
*
* \param path The path, on the local filesystem, of the directory to add.
*
* \pre \p parent is NULL or is a directory.
* \pre \p path is non-NULL and is a valid path to a directory on the local
* filesystem.
* \return a pointer to the newly created directory.
*/
struct iso_tree_node *iso_tree_radd_dir(struct iso_tree_node *parent,
const char *path);
/**
* Add the path of a file or directory to ignore when adding a directory recursively.
*
* \param path The path, on the local filesystem, of the file.
*/
void iso_exclude_add_path(const char *path);
/**
* Remove a path that was set to be ignored when adding a directory recusively.
*
* \param path The path, on the local filesystem, of the file.
*/
void iso_exclude_remove_path(const char *path);
/**
* Remove all paths that were set to be ignored when adding a directory recusively.
*/
void iso_exclude_empty(void);
/**
* Creates a new, empty directory on the volume.
*
* \pre \p parent is NULL or is a directory.
* \pre \p name is unique among the children and files belonging to \p parent.
* Also, it doesn't contain '/' characters.
*
* \post \p parent contains a child directory whose name is \p name and whose
* POSIX attributes are the same as \p parent's.
* \return a pointer to the newly created directory.
*/
struct iso_tree_node *iso_tree_add_new_dir(struct iso_tree_node *parent,
const char *name);
/**
* Set the name of a file (using the current locale).
*/
void iso_tree_node_set_name(struct iso_tree_node *file, const char *name);
/**
* Recursively print a directory to stdout.
* \param spaces The initial number of spaces on the left. Set to 0 if you
* supply a root directory.
*/
void iso_tree_print(const struct iso_tree_node *root, int spaces);
/** Create a burn_source which can be used as a data source for a track
*
* The volume set used to create the libburn_source can _not_ be modified
* until the libburn_source is freed.
*
* \param volumeset The volume set from which you want to write
* \param volnum The volume in the set which you want to write (usually 0)
* \param level ISO level to write at.
* \param flags Which extensions to support.
*
* \pre \p volumeset is non-NULL
* \pre \p volnum is less than \p volset->volset_size.
* \return A burn_source to be used for the data source for a track
*/
struct burn_source* iso_source_new_ecma119 (struct iso_volset *volumeset,
int volnum,
int level,
int flags);
#endif /* LIBISO_LIBISOFS_H */

300
libisofs/rockridge.c Executable file
View File

@ -0,0 +1,300 @@
/* vim: set noet ts=8 sts=8 sw=8 : */
#include "rockridge.h"
#include "util.h"
#include "ecma119.h"
#include "ecma119_tree.h"
#include "tree.h"
#include "susp.h"
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include <errno.h>
#include <unistd.h>
#include <sys/stat.h>
/* create a PX field from the permissions on the current node. */
uint8_t *rrip_make_PX(struct ecma119_write_target *t,
struct ecma119_tree_node *node)
{
uint8_t *PX = malloc(44);
PX[0] = 'P';
PX[1] = 'X';
PX[2] = 44;
PX[3] = 1;
iso_bb(&PX[4], node->iso_self->attrib.st_mode, 4);
iso_bb(&PX[12], node->iso_self->attrib.st_nlink, 4);
iso_bb(&PX[20], node->iso_self->attrib.st_uid, 4);
iso_bb(&PX[28], node->iso_self->attrib.st_gid, 4);
iso_bb(&PX[36], node->iso_self->attrib.st_ino, 4);
return PX;
}
/** See IEEE 1282 4.1.1 */
void rrip_add_PX(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
susp_append(t, &node->susp, rrip_make_PX(t, node));
if (node->type == ECMA119_DIR) {
susp_append(t, &node->dir.self_susp, rrip_make_PX(t, node));
susp_append(t, &node->dir.parent_susp, rrip_make_PX(t, node));
}
}
void rrip_add_PN(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
uint8_t *PN = malloc(20);
PN[0] = 'P';
PN[1] = 'N';
PN[2] = 20;
PN[3] = 1;
iso_bb(&PN[4], node->iso_self->attrib.st_dev >> 32, 4);
iso_bb(&PN[12], node->iso_self->attrib.st_dev & 0xffffffff, 4);
susp_append(t, &node->susp, PN);
}
static void rrip_SL_append_comp(int *n, uint8_t ***comps,
char *s, int size, char fl)
{
uint8_t *comp = malloc(size + 2);
(*n)++;
comp[0] = fl;
comp[1] = size;
*comps = realloc(*comps, (*n) * sizeof(void*));
(*comps)[(*n) - 1] = comp;
if (size) {
memcpy(&comp[2], s, size);
}
}
static void rrip_SL_add_component(char *prev, char *cur, int *n_comp,
uint8_t ***comps)
{
int size = cur - prev;
if (size == 0) {
rrip_SL_append_comp(n_comp, comps, prev, 0, 1 << 3);
return;
}
if (size == 1 && prev[0] == '.') {
rrip_SL_append_comp(n_comp, comps, prev, 0, 1 << 1);
return;
}
if (size == 2 && !strncmp(prev, "..", 2)) {
rrip_SL_append_comp(n_comp, comps, prev, 0, 1 << 2);
return;
}
/* we can't make a component any bigger than 250 (is this really a
problem)? because then it won't fit inside the SL field */
while (size > 248) {
size -= 248;
rrip_SL_append_comp(n_comp, comps, prev, 248, 1 << 0);
}
rrip_SL_append_comp(n_comp, comps, prev, size, 0);
}
void rrip_add_SL(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
int ret, pathsize = 0;
char *path = NULL, *cur, *prev;
int i, j;
uint8_t **comp = NULL;
int n_comp = 0;
int total_comp_len = 0;
int written = 0, pos;
uint8_t *SL;
do {
pathsize += 128;
path = realloc(path, pathsize);
/* FIXME: what if the file is not on the local fs? */
ret = readlink(node->iso_self->loc.path, path, pathsize);
} while (ret == pathsize);
if (ret == -1) {
fprintf(stderr, "Error: couldn't read symlink: %s\n",
strerror(errno));
return;
}
path[ret] = '\0';
prev = path;
for (cur = strchr(path, '/'); cur && *cur; cur = strchr(cur, '/')) {
rrip_SL_add_component(prev, cur, &n_comp, &comp);
cur++;
prev = cur;
}
/* if there was no trailing '/', we need to add the last component. */
if (prev == path || prev != &path[ret - 1]) {
rrip_SL_add_component(prev, &path[ret], &n_comp, &comp);
}
for (i = 0; i < n_comp; i++) {
total_comp_len += comp[i][1] + 2;
if (total_comp_len > 250) {
total_comp_len -= comp[i][1] + 2;
SL = malloc(total_comp_len + 5);
SL[0] = 'S';
SL[1] = 'L';
SL[2] = total_comp_len + 5;
SL[3] = 1;
SL[4] = 1; /* CONTINUE */
pos = 5;
for (j = written; j < i; j++) {
memcpy(&SL[pos], comp[j], comp[j][2]);
pos += comp[j][2];
}
susp_append(t, &node->susp, SL);
written = i - 1;
total_comp_len = comp[i][1];
}
}
SL = malloc(total_comp_len + 5);
SL[0] = 'S';
SL[1] = 'L';
SL[2] = total_comp_len + 5;
SL[3] = 1;
SL[4] = 0;
pos = 5;
for (j = written; j < n_comp; j++) {
memcpy(&SL[pos], comp[j], comp[j][1] + 2);
pos += comp[j][1] + 2;
}
susp_append(t, &node->susp, SL);
free(path);
/* free the components */
for (i = 0; i < n_comp; i++) {
free(comp[i]);
}
free(comp);
}
static void rrip_add_NM_single(struct ecma119_write_target *t,
struct susp_info *susp,
char *name, int size, int flags)
{
uint8_t *NM = malloc(size + 5);
NM[0] = 'N';
NM[1] = 'M';
NM[2] = size + 5;
NM[3] = 1;
NM[4] = flags;
if (size) {
memcpy(&NM[5], name, size);
}
susp_append(t, susp, NM);
}
void
rrip_add_NM(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
char *name = iso_p_fileid(node->iso_self->name);
int len = name ? strlen(name) : 0;
char *pos = name;
if (!len)
return;
if (node->type == ECMA119_DIR) {
rrip_add_NM_single(t, &node->dir.self_susp, pos, 0, 1 << 1);
rrip_add_NM_single(t, &node->dir.parent_susp, pos, 0, 1 << 2);
}
while (len > 250) {
rrip_add_NM_single(t, &node->susp, pos, 250, 1);
len -= 250;
pos += 250;
}
rrip_add_NM_single(t, &node->susp, pos, len, 0);
}
void rrip_add_CL(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
uint8_t *CL = calloc(1, 12);
CL[0] = 'C';
CL[1] = 'L';
CL[2] = 12;
CL[3] = 1;
susp_append(t, &node->susp, CL);
}
void
rrip_add_PL(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
uint8_t *PL = calloc(1, 12);
PL[0] = 'P';
PL[1] = 'L';
PL[2] = 12;
PL[3] = 1;
susp_append(t, &node->dir.parent_susp, PL);
}
void
rrip_add_RE(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
uint8_t *RE = malloc(4);
RE[0] = 'R';
RE[1] = 'E';
RE[2] = 4;
RE[3] = 1;
susp_append(t, &node->susp, RE);
}
void
rrip_add_TF(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
uint8_t *TF = malloc(5 + 3 * 7);
TF[0] = 'T';
TF[1] = 'F';
TF[2] = 5 + 3 * 7;
TF[3] = 1;
TF[4] = (1 << 1) | (1 << 2) | (1 << 3) | (1 << 7);
iso_datetime_7(&TF[5], node->iso_self->attrib.st_mtime);
iso_datetime_7(&TF[12], node->iso_self->attrib.st_atime);
iso_datetime_7(&TF[19], node->iso_self->attrib.st_ctime);
susp_append(t, &node->susp, TF);
}
void
rrip_finalize(struct ecma119_write_target *t, struct ecma119_tree_node *dir)
{
int i;
assert(dir->type == ECMA119_DIR);
if (dir->parent != dir->dir.real_parent) {
uint8_t *PL = susp_find(&dir->dir.parent_susp, "PL");
assert(PL);
iso_bb(&PL[4], dir->dir.real_parent->block, 4);
}
for (i = 0; i < dir->dir.nchildren; i++) {
struct ecma119_tree_node *ch = dir->dir.children[i];
if (ch->type == ECMA119_FILE && ch->file.real_me) {
uint8_t *CL = susp_find(&ch->susp, "CL");
assert(CL);
iso_bb(&CL[4], ch->file.real_me->block, 4);
} else if (ch->type == ECMA119_DIR) {
rrip_finalize(t, ch);
}
}
}

26
libisofs/rockridge.h Executable file
View File

@ -0,0 +1,26 @@
/* vim: set noet ts=8 sts=8 sw=8 : */
/** Functions and structures used for Rock Ridge support. */
#ifndef ISO_ROCKRIDGE_H
#define ISO_ROCKRIDGE_H
struct ecma119_write_target;
struct ecma119_tree_node;
void rrip_add_PX(struct ecma119_write_target *, struct ecma119_tree_node *);
void rrip_add_PN(struct ecma119_write_target *, struct ecma119_tree_node *);
void rrip_add_SL(struct ecma119_write_target *, struct ecma119_tree_node *);
void rrip_add_NM(struct ecma119_write_target *, struct ecma119_tree_node *);
void rrip_add_CL(struct ecma119_write_target *, struct ecma119_tree_node *);
void rrip_add_RE(struct ecma119_write_target *, struct ecma119_tree_node *);
void rrip_add_TF(struct ecma119_write_target *, struct ecma119_tree_node *);
/* This is special because it doesn't modify the susp fields of the directory
* that gets passed to it; it modifies the susp fields of the ".." entry in
* that directory. */
void rrip_add_PL(struct ecma119_write_target *, struct ecma119_tree_node *);
void rrip_finalize(struct ecma119_write_target *, struct ecma119_tree_node *);
#endif /* ISO_ROCKRIDGE_H */

280
libisofs/susp.c Executable file
View File

@ -0,0 +1,280 @@
/* vim: set noet ts=8 sts=8 sw=8 : */
#include "susp.h"
#include "util.h"
#include "ecma119.h"
#include "ecma119_tree.h"
#include <stdlib.h>
#include <assert.h>
#include <string.h>
void susp_insert(struct ecma119_write_target *t,
struct susp_info *susp,
uint8_t *data,
int pos)
{
int i;
if (pos < 0) {
pos = susp->n_susp_fields;
}
assert(pos <= susp->n_susp_fields);
susp->n_susp_fields++;
susp->susp_fields = realloc(susp->susp_fields,
sizeof(void*) * susp->n_susp_fields);
for (i = susp->n_susp_fields-1; i > pos; i--) {
susp->susp_fields[i] = susp->susp_fields[i - 1];
}
susp->susp_fields[pos] = data;
}
void susp_append(struct ecma119_write_target *t,
struct susp_info *susp,
uint8_t *data)
{
susp_insert(t, susp, data, susp->n_susp_fields);
}
uint8_t *susp_find(struct susp_info *susp, const char *name)
{
int i;
for (i = 0; i < susp->n_susp_fields; i++) {
if (!strncmp((char *)susp->susp_fields[i], name, 2)) {
return susp->susp_fields[i];
}
}
return NULL;
}
/** Utility function for susp_add_CE because susp_add_CE needs to act 3 times
* on directories (for the "." and ".." entries.
*
* \param len The amount of space available for the System Use area.
*/
#define CE_LEN 28
static unsigned char *susp_add_single_CE(struct ecma119_write_target *t,
struct susp_info *susp,
int len)
{
int susp_length = 0, tmp_len;
int i;
unsigned char *CE;
for (i = 0; i < susp->n_susp_fields; i++) {
susp_length += susp->susp_fields[i][2];
}
if (susp_length <= len) {
/* no need for a CE field */
susp->non_CE_len = susp_length;
susp->n_fields_fit = susp->n_susp_fields;
return NULL;
}
tmp_len = susp_length;
for (i = susp->n_susp_fields - 1; i >= 0; i--) {
tmp_len -= susp->susp_fields[i][2];
if (tmp_len + CE_LEN <= len) {
susp->non_CE_len = tmp_len + CE_LEN;
susp->CE_len = susp_length - tmp_len;
/* i+1 because we have to count the CE field */
susp->n_fields_fit = i + 1;
CE = calloc(1, CE_LEN);
/* we don't fill in the BLOCK LOCATION or OFFSET
fields yet. */
CE[0] = 'C';
CE[1] = 'E';
CE[2] = (char)CE_LEN;
CE[3] = (char)1;
iso_bb(&CE[20], susp_length - tmp_len, 4);
return CE;
}
}
assert(0);
return NULL;
}
static void
try_add_CE(struct ecma119_write_target *t,
struct susp_info *susp,
size_t dirent_len)
{
uint8_t *CE = susp_add_single_CE(t, susp, 255 - dirent_len);
if (CE)
susp_insert(t, susp, CE, susp->n_fields_fit - 1);
}
/** See IEEE P1281 Draft Version 1.12/5.2. Because this function depends on the
* length of the other SUSP fields, it should always be calculated last. */
void
susp_add_CE(struct ecma119_write_target *t, struct ecma119_tree_node *node)
{
try_add_CE(t, &node->susp, node->dirent_len);
if (node->type == ECMA119_DIR) {
try_add_CE(t, &node->dir.self_susp, 34);
try_add_CE(t, &node->dir.parent_susp, 34);
}
}
/** See IEEE P1281 Draft Version 1.12/5.3 */
void
susp_add_SP(struct ecma119_write_target *t, struct ecma119_tree_node *dir)
{
unsigned char *SP = malloc(7);
assert(dir->type == ECMA119_DIR);
SP[0] = 'S';
SP[1] = 'P';
SP[2] = (char)7;
SP[3] = (char)1;
SP[4] = 0xbe;
SP[5] = 0xef;
SP[6] = 0;
susp_append(t, &dir->dir.self_susp, SP);
}
#if 0
/** See IEEE P1281 Draft Version 1.12/5.4 */
static void susp_add_ST(struct ecma119_write_target *t,
struct iso_tree_node *node)
{
unsigned char *ST = malloc(4);
ST[0] = 'S';
ST[1] = 'T';
ST[2] = (char)4;
ST[3] = (char)1;
susp_append(t, node, ST);
}
#endif
/** See IEEE P1281 Draft Version 1.12/5.5 FIXME: this is rockridge */
void
rrip_add_ER(struct ecma119_write_target *t, struct ecma119_tree_node *dir)
{
unsigned char *ER = malloc(182);
assert(dir->type == ECMA119_DIR);
ER[0] = 'E';
ER[1] = 'R';
ER[2] = 182;
ER[3] = 1;
ER[4] = 9;
ER[5] = 72;
ER[6] = 93;
ER[7] = 1;
memcpy(&ER[8], "IEEE_1282", 9);
memcpy(&ER[17], "THE IEEE 1282 PROTOCOL PROVIDES SUPPORT FOR POSIX "
"FILE SYSTEM SEMANTICS.", 72);
memcpy(&ER[89], "PLEASE CONTACT THE IEEE STANDARDS DEPARTMENT, "
"PISCATAWAY, NJ, USA FOR THE 1282 SPECIFICATION.", 93);
susp_append(t, &dir->dir.self_susp, ER);
}
/* calculate the location of the CE areas. Since CE areas don't need to be
* aligned to a block boundary, we contatenate all CE areas from a single
* directory and dump them immediately after all the directory records.
*
* Requires that the following be known:
* - position of the current directory (dir->block)
* - length of the current directory (dir->dir.len)
* - sum of the children's CE lengths (dir->dir.CE_len)
*/
static void
susp_fin_1_CE(struct ecma119_write_target *t,
struct susp_info *susp,
size_t block,
size_t *offset)
{
uint8_t *CE = susp->susp_fields[susp->n_fields_fit - 1];
if (!susp->CE_len) {
return;
}
iso_bb(&CE[4], block + (*offset) / t->block_size, 4);
iso_bb(&CE[12], (*offset) % t->block_size, 4);
*offset += susp->CE_len;
}
static void susp_fin_CE(struct ecma119_write_target *t,
struct ecma119_tree_node *dir)
{
int i;
size_t CE_offset = dir->dir.len;
assert(dir->type == ECMA119_DIR);
susp_fin_1_CE(t, &dir->dir.self_susp, dir->block, &CE_offset);
susp_fin_1_CE(t, &dir->dir.parent_susp, dir->block, &CE_offset);
for (i = 0; i < dir->dir.nchildren; i++) {
struct ecma119_tree_node *ch = dir->dir.children[i];
susp_fin_1_CE(t, &ch->susp, dir->block, &CE_offset);
}
assert(CE_offset == dir->dir.len + dir->dir.CE_len);
}
void
susp_finalize(struct ecma119_write_target *t, struct ecma119_tree_node *dir)
{
int i;
assert(dir->type = ECMA119_DIR);
if (dir->dir.depth != 1) {
susp_fin_CE(t, dir);
}
for (i = 0; i < dir->dir.nchildren; i++) {
if (dir->dir.children[i]->type == ECMA119_DIR)
susp_finalize(t, dir->dir.children[i]);
}
}
void susp_write(struct ecma119_write_target *t,
struct susp_info *susp,
unsigned char *buf)
{
int i;
int pos = 0;
for (i = 0; i < susp->n_fields_fit; i++) {
memcpy(&buf[pos], susp->susp_fields[i],
susp->susp_fields[i][2]);
pos += susp->susp_fields[i][2];
}
}
void susp_write_CE(struct ecma119_write_target *t, struct susp_info *susp,
unsigned char *buf)
{
int i;
int pos = 0;
for (i = susp->n_fields_fit; i < susp->n_susp_fields; i++) {
memcpy(&buf[pos], susp->susp_fields[i],
susp->susp_fields[i][2]);
pos += susp->susp_fields[i][2];
}
}
void susp_free_fields(struct susp_info *susp)
{
int i;
for (i=0; i<susp->n_susp_fields; i++) {
free(susp->susp_fields[i]);
}
if (susp->susp_fields) {
free(susp->susp_fields);
}
memset(susp, 0, sizeof(struct susp_info));
}

62
libisofs/susp.h Executable file
View File

@ -0,0 +1,62 @@
/* vim: set noet ts=8 sts=8 sw=8 : */
/** Functions and structures used for SUSP (IEEE 1281).
*/
#ifndef __ISO_SUSP
#define __ISO_SUSP
#include <stdint.h>
/* SUSP is only present in standard ecma119 */
struct ecma119_write_target;
struct ecma119_tree_node;
/** This contains the information that needs to go in the SUSP area of a file.
*/
struct susp_info
{
int n_susp_fields; /**< Number of SUSP fields */
uint8_t **susp_fields; /**< Data for each SUSP field */
/* the next 3 relate to CE and are filled out by susp_add_CE. */
int n_fields_fit; /**< How many of the above SUSP fields fit
* within this node's dirent. */
int non_CE_len; /**< Length of the part of the SUSP area that
* fits in the dirent. */
int CE_len; /**< Length of the part of the SUSP area that
* will go in a CE area. */
};
void susp_add_CE(struct ecma119_write_target *, struct ecma119_tree_node *);
/* these next 2 are special because they don't modify the susp fields of the
* directory; they modify the susp fields of the
* "." entry in the directory. */
void susp_add_SP(struct ecma119_write_target *, struct ecma119_tree_node *);
void rrip_add_ER(struct ecma119_write_target *, struct ecma119_tree_node *);
/** Once all the directories and files are laid out, recurse through the tree
* and finalize all SUSP CE entries. */
void susp_finalize(struct ecma119_write_target *, struct ecma119_tree_node *);
void susp_append(struct ecma119_write_target *,
struct susp_info *,
uint8_t *);
void susp_insert(struct ecma119_write_target *,
struct susp_info *,
uint8_t *,
int pos);
uint8_t *susp_find(struct susp_info *,
const char *);
void susp_write(struct ecma119_write_target *,
struct susp_info *,
uint8_t *);
void susp_write_CE(struct ecma119_write_target *,
struct susp_info *,
uint8_t *);
void susp_free_fields(struct susp_info *);
#endif /* __ISO_SUSP */

223
libisofs/tree.c Executable file
View File

@ -0,0 +1,223 @@
/* vim: set noet ts=8 sts=8 sw=8 : */
/**
* \file tree.c
*
* Implement filesystem trees.
*/
#include <assert.h>
#include <stdlib.h>
#include <string.h>
#include <time.h>
#include <dirent.h>
#include <libgen.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <unistd.h>
#include <errno.h>
#include <err.h>
#include <stdio.h>
#include "tree.h"
#include "util.h"
#include "volume.h"
#include "exclude.h"
static void
set_default_stat(struct stat *s)
{
time_t now = time(NULL);
memset(s, 0, sizeof(struct stat));
s->st_mode = 0777 | S_IFREG;
s->st_atime = s->st_mtime = s->st_ctime = now;
}
static struct stat
get_attrib(const struct iso_tree_node *node)
{
struct stat st;
if (node) {
return node->attrib;
}
set_default_stat(&st);
return st;
}
static void
append_node(struct iso_tree_node *parent,
struct iso_tree_node *child)
{
assert((!parent || S_ISDIR(parent->attrib.st_mode)) && child);
if (!parent)
return;
parent->nchildren++;
parent->children =
realloc(parent->children, parent->nchildren * sizeof(void*));
parent->children[parent->nchildren-1] = child;
}
struct iso_tree_node*
iso_tree_new_root(struct iso_volume *vol)
{
assert(vol);
if (vol->root) {
iso_tree_free(vol->root);
}
vol->root = calloc(1, sizeof(struct iso_tree_node));
vol->root->volume = vol;
set_default_stat(&vol->root->attrib);
vol->root->attrib.st_mode = S_IFDIR | 0777;
vol->root->loc.type = LIBISO_NONE;
return vol->root;
}
struct iso_tree_node*
iso_tree_add_new_file(struct iso_tree_node *parent, const char *name)
{
struct iso_tree_node *f = calloc(1, sizeof(struct iso_tree_node));
assert((!parent || S_ISDIR(parent->attrib.st_mode)) && name);
f->volume = parent ? parent->volume : NULL;
f->parent = parent;
f->name = parent ? strdup(name) : NULL;
f->attrib = get_attrib(parent);
f->attrib.st_mode = 0777 | S_IFREG;
f->loc.type = LIBISO_NONE;
append_node(parent, f);
return f;
}
struct iso_tree_node*
iso_tree_add_new_dir(struct iso_tree_node *parent, const char *name)
{
struct iso_tree_node *d = iso_tree_add_new_file(parent, name);
assert((!parent || S_ISDIR(parent->attrib.st_mode)) && name);
d->attrib.st_mode = (d->attrib.st_mode & ~S_IFMT) | S_IFDIR;
return d;
}
struct iso_tree_node*
iso_tree_add_node(struct iso_tree_node *parent, const char *path)
{
char *p;
struct stat st;
struct iso_tree_node *ret;
assert((!parent || S_ISDIR(parent->attrib.st_mode)) && path);
if (lstat(path, &st) == -1)
return NULL;
p = strdup(path); /* because basename() might modify its arg */
/* it doesn't matter if we add a file or directory since we modify
* attrib anyway. */
ret = iso_tree_add_new_file(parent, basename(p));
ret->attrib = st;
ret->loc.type = LIBISO_FILESYS;
ret->loc.path = strdup(path);
free(p);
return ret;
}
struct iso_tree_node*
iso_tree_radd_dir (struct iso_tree_node *parent, const char *path)
{
struct iso_tree_node *new;
DIR *dir;
struct dirent *ent;
assert((!parent || S_ISDIR(parent->attrib.st_mode)) && path);
new = iso_tree_add_node(parent, path);
if (!new || !S_ISDIR(new->attrib.st_mode)) {
return new;
}
dir = opendir(path);
if (!dir) {
warn("couldn't open directory %s: %s\n", path, strerror(errno));
return new;
}
while ((ent = readdir(dir))) {
char child[strlen(ent->d_name) + strlen(path) + 2];
if (strcmp(ent->d_name, ".") == 0 ||
strcmp(ent->d_name, "..") == 0)
continue;
sprintf(child, "%s/%s", path, ent->d_name);
/* see if this child is excluded. */
if (iso_exclude_lookup(child))
continue;
iso_tree_radd_dir(new, child);
}
closedir(dir);
return new;
}
void
iso_tree_free(struct iso_tree_node *root)
{
size_t i;
for (i=0; i < root->nchildren; i++) {
iso_tree_free(root->children[i]);
}
free(root->name);
free(root->children);
free(root);
}
void
iso_tree_print(const struct iso_tree_node *root, int spaces)
{
size_t i;
char sp[spaces+1];
memset(sp, ' ', spaces);
sp[spaces] = '\0';
printf("%s%sn", sp, root->name);
for (i=0; i < root->nchildren; i++) {
iso_tree_print(root->children[i], spaces+2);
}
}
void
iso_tree_print_verbose(const struct iso_tree_node *root,
print_dir_callback dir,
print_file_callback file,
void *callback_data,
int spaces)
{
size_t i;
(S_ISDIR(root->attrib.st_mode) ? dir : file)
(root, callback_data, spaces);
for (i=0; i < root->nchildren; i++) {
iso_tree_print_verbose(root->children[i], dir,
file, callback_data, spaces+2);
}
}
void
iso_tree_node_set_name(struct iso_tree_node *file, const char *name)
{
free(file->name);
file->name = strdup(name);
}

159
libisofs/tree.h Executable file
View File

@ -0,0 +1,159 @@
/* vim: set noet ts=8 sts=8 sw=8 : */
/**
* \file tree.h
*
* Declare the structure of a libisofs filesystem tree. The files in this
* tree can come from either the local filesystem or from another .iso image
* (for multisession).
*
* This tree preserves as much information as it can about the files; names
* are stored in wchar_t and we preserve POSIX attributes. This tree does
* *not* include information that is necessary for writing out, for example,
* an ISO level 1 tree. That information will go in a different tree because
* the structure is sufficiently different.
*/
#ifndef LIBISO_TREE_H
#define LIBISO_TREE_H
#include <sys/types.h>
#include <sys/stat.h>
#include <stdint.h>
#include <wchar.h>
#include "libisofs.h"
enum file_location {
LIBISO_FILESYS,
LIBISO_PREVSESSION,
LIBISO_NONE /**< for files/dirs that were added with
* iso_tree_add_new_XXX. */
};
/**
* This tells us where to read the data from a file. Either we read from the
* local filesystem or we just point to the block on a previous session.
*/
struct iso_file_location
{
enum file_location type;
/* union {*/
char *path; /* in the current locale */
uint32_t block;
/* };*/
};
/**
* A node in the filesystem tree.
*/
struct iso_tree_node
{
struct iso_volume *volume;
struct iso_tree_node *parent;
char *name;
struct stat attrib; /**< The POSIX attributes of this node as
* documented in "man 2 stat". */
struct iso_file_location loc;
/**< Only used for regular files and symbolic
* links (ie. files for which we might have to
* copy data). */
size_t nchildren; /**< The number of children of this
* directory (if this is a directory). */
struct iso_tree_node **children;
size_t block; /**< The block at which this file will
* reside on disk. We store this here as
* well as in the various mangled trees
* because many different trees might point
* to the same file and they need to share the
* block location. */
};
/**
* Create a new root directory for a volume.
*
* \param vol The volume for which to create a new root directory.
*
* \pre \p vol is non-NULL.
* \post \p vol has a non-NULL, empty root directory with permissions 777.
* \return \p vol's new non-NULL, empty root directory.
*/
struct iso_tree_node *iso_tree_new_root(struct iso_volume *vol);
/**
* Create a new, empty, file.
*
* \param parent The parent directory of the new file. If this is null, create
* and return a new file node without adding it to any tree.
* \param name The name of the new file, encoded in the current locale.
* \pre \p name is non-NULL and it does not match any other file or directory
* name in \p parent.
* \post \p parent (if non-NULL) contains a file with the following properties:
* - the file's name is \p name (converted to wchar_t)
* - the file's POSIX permissions are the same as \p parent's
* - the file is a regular file
* - the file is empty
*
* \return \p parent's newly created file.
*/
struct iso_tree_node *iso_tree_add_new_file(struct iso_tree_node *parent,
const char *name);
/**
* Recursively free a directory.
*
* \param root The root of the directory heirarchy to free.
*
* \pre \p root is non-NULL.
*/
void iso_tree_free(struct iso_tree_node *root);
/**
* A function that prints verbose information about a directory.
*
* \param dir The directory about which to print information.
* \param data Unspecified function-dependent data.
* \param spaces The number of spaces to prepend to the output.
*
* \see iso_tree_print_verbose
*/
typedef void (*print_dir_callback) (const struct iso_tree_node *dir,
void *data,
int spaces);
/**
* A function that prints verbose information about a file.
*
* \param dir The file about which to print information.
* \param data Unspecified function-dependent data.
* \param spaces The number of spaces to prepend to the output.
*
* \see iso_tree_print_verbose
*/
typedef void (*print_file_callback) (const struct iso_tree_node *file,
void *data,
int spaces);
/**
* Recursively print a directory heirarchy. For each node in the directory
* heirarchy, call a callback function to print information more verbosely.
*
* \param root The root of the directory heirarchy to print.
* \param dir The callback function to call for each directory in the tree.
* \param file The callback function to call for each file in the tree.
* \param callback_data The data to pass to the callback functions.
* \param spaces The number of spaces to prepend to the output.
*
* \pre \p root is not NULL.
* \pre Neither of the callback functions modifies the directory heirarchy.
*/
void iso_tree_print_verbose(const struct iso_tree_node *root,
print_dir_callback dir,
print_file_callback file,
void *callback_data,
int spaces);
#define ISO_ISDIR(n) S_ISDIR(n->attrib.st_mode)
#endif /* LIBISO_TREE_H */

584
libisofs/util.c Executable file
View File

@ -0,0 +1,584 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set noet ts=8 sts=8 sw=8 : */
/**
* Utility functions for the Libisofs library.
*/
#ifdef HAVE_CONFIG_H
#include <config.h>
#endif
#include <wchar.h>
#include <iconv.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <ctype.h>
#include <assert.h>
#include <errno.h>
#include <locale.h>
#include "util.h"
/* avoids warning and names in iso, joliet and rockridge can't be > 255 bytes
* anyway. There are at most 31 characters in iso level 1, 255 for rockridge,
* 64 characters (* 2 since UCS) for joliet. */
#define NAME_BUFFER_SIZE 255
int div_up(int n, int div)
{
return (n + div - 1) / div;
}
int round_up(int n, int mul)
{
return div_up(n, mul) * mul;
}
/* this function must always return a name
* since the caller never checks if a NULL
* is returned. It also avoids some warnings. */
char *str2ascii(const char *src_arg)
{
wchar_t wsrc_[NAME_BUFFER_SIZE];
char *src = (char*)wsrc_;
char *ret_;
char *ret;
mbstate_t state;
iconv_t conv;
size_t numchars;
size_t outbytes;
size_t inbytes;
size_t n;
if (!src_arg)
return NULL;
/* convert the string to a wide character string. Note: outbytes
* is in fact the number of characters in the string and doesn't
* include the last NULL character. */
memset(&state, 0, sizeof(state));
numchars = mbsrtowcs(wsrc_, &src_arg, NAME_BUFFER_SIZE-1, &state);
if (numchars < 0)
return NULL;
inbytes = numchars * sizeof(wchar_t);
ret_ = malloc(numchars+1);
outbytes = numchars;
ret = ret_;
/* initialize iconv */
conv = iconv_open("ASCII", "WCHAR_T");
if (conv == (iconv_t)-1)
return NULL;
n = iconv(conv, &src, &inbytes, &ret, &outbytes);
while(n == -1) {
/* The destination buffer is too small. Stops here. */
if(errno == E2BIG)
break;
/* An incomplete multi bytes sequence was found. We
* can't do anything here. That's quite unlikely. */
if(errno == EINVAL)
break;
/* The last possible error is an invalid multi bytes
* sequence. Just replace the character with a "_".
* Probably the character doesn't exist in ascii like
* "é, è, à, ç, ..." in French. */
*ret++ = '_';
outbytes--;
if(!outbytes)
break;
/* There was an error with one character but some other remain
* to be converted. That's probably a multibyte character.
* See above comment. */
src += sizeof(wchar_t);
inbytes -= sizeof(wchar_t);
if(!inbytes)
break;
n = iconv(conv, &src, &inbytes, &ret, &outbytes);
}
iconv_close(conv);
*ret='\0';
return ret_;
}
/* FIXME: C&P */
uint16_t *str2ucs(const char *src_arg)
{
wchar_t wsrc_[NAME_BUFFER_SIZE];
char *src = (char*)wsrc_;
char *ret_;
char *ret;
mbstate_t state;
iconv_t conv;
size_t outbytes;
size_t numchars;
size_t inbytes;
size_t n;
if (!src_arg)
return calloc(2, 1); /* empty UCS string */
/* convert the string to a wide character string. Note: outbytes
* is in fact the number of characters in the string and doesn't
* include the last NULL character. */
memset(&state, 0, sizeof(state));
numchars = mbsrtowcs(wsrc_, &src_arg, NAME_BUFFER_SIZE-1, &state);
if (numchars < 0)
return calloc(2, 1); /* empty UCS string */
inbytes = numchars * sizeof(wchar_t);
outbytes = numchars * sizeof(uint16_t);
ret_ = malloc ((numchars+1) * sizeof(uint16_t));
ret = ret_;
/* initialize iconv */
conv = iconv_open("UCS-2BE", "WCHAR_T");
if (conv == (iconv_t)-1)
return calloc(2, 1); /* empty UCS string */
n = iconv(conv, &src, &inbytes, &ret, &outbytes);
while(n == -1) {
/* The destination buffer is too small. Stops here. */
if(errno == E2BIG)
break;
/* An incomplete multi bytes sequence was found. We
* can't do anything here. That's quite unlikely. */
if(errno == EINVAL)
break;
/* The last possible error is an invalid multi bytes
* sequence. Just replace the character with a "_".
* Probably the character doesn't exist in ascii like
* "é, è, à, ç, ..." in French. */
*((uint16_t*) ret) = '_';
ret += sizeof(uint16_t);
outbytes -= sizeof(uint16_t);
if(!outbytes)
break;
/* There was an error with one character but some other remain
* to be converted. That's probably a multibyte character.
* See above comment. */
src += sizeof(wchar_t);
inbytes -= sizeof(wchar_t);
if(!inbytes)
break;
n = iconv(conv, &src, &inbytes, &ret, &outbytes);
}
iconv_close(conv);
/* close the ucs string */
*((uint16_t*) ret) = 0;
return (uint16_t*)ret_;
}
static int valid_d_char(char c)
{
return (c >= '0' && c <= '9') || (c >= 'A' && c <= 'Z') || (c == '_');
}
static int valid_a_char(char c)
{
return (c >= ' ' && c <= '"') || (c >= '%' && c <= '?')
|| (c >= 'A' && c <= 'Z')
|| (c == '_');
}
static int valid_j_char(uint16_t c)
{
return !(c < (uint16_t)' ' || c == (uint16_t)'*' || c == (uint16_t)'/'
|| c == (uint16_t)':' || c == (uint16_t)';'
|| c == (uint16_t)'?' || c == (uint16_t)'\\');
}
/* FIXME: where are these documented? */
static int valid_p_char(char c)
{
return (c >= '0' && c <= '9') || (c >= 'A' && c <= 'Z')
|| (c >= 'a' && c <= 'z')
|| (c == '.') || (c == '_') || (c == '-');
}
static char *iso_dirid(const char *src, int size)
{
char *ret = str2ascii(src);
size_t len, i;
if (!ret)
return NULL;
len = strlen(ret);
if (len > size) {
ret[size] = '\0';
len = size;
}
for (i = 0; i < len; i++) {
char c = toupper(ret[i]);
ret[i] = valid_d_char(c) ? c : '_';
}
return ret;
}
char *iso_1_dirid(const char *src)
{
return iso_dirid(src, 8);
}
char *iso_2_dirid(const char *src)
{
return iso_dirid(src, 31);
}
char *iso_1_fileid(const char *src_arg)
{
char *src = str2ascii(src_arg);
char *dest;
char *dot; /* Position of the last dot in the
filename, will be used to calculate
lname and lext. */
int lname, lext, pos, i;
if (!src)
return NULL;
dest = malloc(15); /* 15 = 8 (name) + 1 (.) + 3 (ext) + 2
(;1) + 1 (\0) */
dot = strrchr(src, '.');
lext = dot ? strlen(dot + 1) : 0;
lname = strlen(src) - lext - (dot ? 1 : 0);
/* If we can't build a filename, return NULL. */
if (lname == 0 && lext == 0) {
free(src);
free(dest);
return NULL;
}
pos = 0;
/* Convert up to 8 characters of the filename. */
for (i = 0; i < lname && i < 8; i++) {
char c = toupper(src[i]);
dest[pos++] = valid_d_char(c) ? c : '_';
}
/* This dot is mandatory, even if there is no extension. */
dest[pos++] = '.';
/* Convert up to 3 characters of the extension, if any. */
for (i = 0; i < lext && i < 3; i++) {
char c = toupper(src[lname + 1 + i]);
dest[pos++] = valid_d_char(c) ? c : '_';
}
/* File versions are mandatory, even if they aren't used. */
dest[pos++] = ';';
dest[pos++] = '1';
dest[pos] = '\0';
dest = (char *)realloc(dest, pos + 1);
free(src);
return dest;
}
char *iso_2_fileid(const char *src_arg)
{
char *src = str2ascii(src_arg);
char *dest;
char *dot;
int lname, lext, lnname, lnext, pos, i;
if (!src)
return NULL;
dest = malloc(34); /* 34 = 30 (name + ext) + 1 (.) + 2
(;1) + 1 (\0) */
dot = strrchr(src, '.');
/* Since the maximum length can be divided freely over the name and
extension, we need to calculate their new lengths (lnname and
lnext). If the original filename is too long, we start by trimming
the extension, but keep a minimum extension length of 3. */
if (dot == NULL || dot == src || *(dot + 1) == '\0') {
lname = strlen(src);
lnname = (lname > 30) ? 30 : lname;
lext = lnext = 0;
} else {
lext = strlen(dot + 1);
lname = strlen(src) - lext - 1;
lnext = (strlen(src) > 31 && lext > 3)
? (lname < 27 ? 30 - lname : 3) : lext;
lnname = (strlen(src) > 31) ? 30 - lnext : lname;
}
if (lnname == 0 && lnext == 0) {
free(src);
free(dest);
return NULL;
}
pos = 0;
/* Convert up to lnname characters of the filename. */
for (i = 0; i < lnname; i++) {
char c = toupper(src[i]);
dest[pos++] = valid_d_char(c) ? c : '_';
}
dest[pos++] = '.';
/* Convert up to lnext characters of the extension, if any. */
for (i = 0; i < lnext; i++) {
char c = toupper(src[lname + 1 + i]);
dest[pos++] = valid_d_char(c) ? c : '_';
}
dest[pos++] = ';';
dest[pos++] = '1';
dest[pos] = '\0';
dest = (char *)realloc(dest, pos + 1);
free(src);
return dest;
}
char *
iso_p_fileid(const char *src)
{
char *ret = str2ascii(src);
size_t i, len;
if (!ret)
return NULL;
len = strlen(ret);
for (i = 0; i < len; i++) {
if (!valid_p_char(ret[i])) {
ret[i] = (uint16_t)'_';
}
}
return ret;
}
uint16_t *
iso_j_id(const char *src_arg)
{
uint16_t *j_str = str2ucs(src_arg);
size_t len = ucslen(j_str);
size_t n;
if (len > 128) {
j_str[128] = '\0';
len = 128;
}
for (n = 0; n < len; n++)
if (!valid_j_char(j_str[n]))
j_str[n] = '_';
return j_str;
}
void iso_lsb(uint8_t *buf, uint32_t num, int bytes)
{
int i;
assert(bytes <= 4);
for (i = 0; i < bytes; ++i)
buf[i] = (num >> (8 * i)) & 0xff;
}
void iso_msb(uint8_t *buf, uint32_t num, int bytes)
{
int i;
assert(bytes <= 4);
for (i = 0; i < bytes; ++i)
buf[bytes - 1 - i] = (num >> (8 * i)) & 0xff;
}
void iso_bb(uint8_t *buf, uint32_t num, int bytes)
{
iso_lsb(buf, num, bytes);
iso_msb(buf+bytes, num, bytes);
}
void iso_datetime_7(unsigned char *buf, time_t t)
{
static int tzsetup = 0;
int tzoffset;
struct tm tm;
if (!tzsetup) {
tzset();
tzsetup = 1;
}
localtime_r(&t, &tm);
buf[0] = tm.tm_year;
buf[1] = tm.tm_mon + 1;
buf[2] = tm.tm_mday;
buf[3] = tm.tm_hour;
buf[4] = tm.tm_min;
buf[5] = tm.tm_sec;
#ifdef HAVE_TM_GMTOFF
tzoffset = -tm.tm_gmtoff / 60 / 15;
#else
tzoffset = -timezone / 60 / 15;
#endif
if (tzoffset < -48)
tzoffset += 101;
buf[6] = tzoffset;
}
time_t iso_datetime_read_7(const uint8_t *buf)
{
struct tm tm;
tm.tm_year = buf[0];
tm.tm_mon = buf[1] + 1;
tm.tm_mday = buf[2];
tm.tm_hour = buf[3];
tm.tm_min = buf[4];
tm.tm_sec = buf[5];
return mktime(&tm) - buf[6] * 60 * 60;
}
void iso_datetime_17(unsigned char *buf, time_t t)
{
static int tzsetup = 0;
static int tzoffset;
struct tm tm;
if (t == (time_t) - 1) {
/* unspecified time */
memset(buf, '0', 16);
buf[16] = 0;
} else {
if (!tzsetup) {
tzset();
tzsetup = 1;
}
localtime_r(&t, &tm);
sprintf((char*)&buf[0], "%04d", tm.tm_year + 1900);
sprintf((char*)&buf[4], "%02d", tm.tm_mon + 1);
sprintf((char*)&buf[6], "%02d", tm.tm_mday);
sprintf((char*)&buf[8], "%02d", tm.tm_hour);
sprintf((char*)&buf[10], "%02d", tm.tm_min);
sprintf((char*)&buf[12], "%02d", MIN(59, tm.tm_sec));
memcpy(&buf[14], "00", 2);
#ifdef HAVE_TM_GMTOFF
tzoffset = -tm.tm_gmtoff / 60 / 15;
#else
tzoffset = -timezone / 60 / 15;
#endif
if (tzoffset < -48)
tzoffset += 101;
buf[16] = tzoffset;
}
}
time_t iso_datetime_read_17(const uint8_t *buf)
{
struct tm tm;
sscanf((char*)&buf[0], "%4d", &tm.tm_year);
sscanf((char*)&buf[4], "%2d", &tm.tm_mon);
sscanf((char*)&buf[6], "%2d", &tm.tm_mday);
sscanf((char*)&buf[8], "%2d", &tm.tm_hour);
sscanf((char*)&buf[10], "%2d", &tm.tm_min);
sscanf((char*)&buf[12], "%2d", &tm.tm_sec);
tm.tm_year -= 1900;
tm.tm_mon -= 1;
return mktime(&tm) - buf[16] * 60 * 60;
}
size_t ucslen(const uint16_t *str)
{
int i;
for (i=0; str[i]; i++)
;
return i;
}
/**
* Although each character is 2 bytes, we actually compare byte-by-byte
* (thats what the spec says).
*/
int ucscmp(const uint16_t *s1, const uint16_t *s2)
{
const char *s = (const char*)s1;
const char *t = (const char*)s2;
size_t len1 = ucslen(s1);
size_t len2 = ucslen(s2);
size_t i, len = MIN(len1, len2) * 2;
for (i=0; i < len; i++) {
if (s[i] < t[i]) {
return -1;
} else if (s[i] > t[i]) {
return 1;
}
}
if (len1 < len2)
return -1;
else if (len1 > len2)
return 1;
return 0;
}
uint32_t iso_read_lsb(const uint8_t *buf, int bytes)
{
int i;
uint32_t ret = 0;
for (i=0; i<bytes; i++) {
ret += ((uint32_t) buf[i]) << (i*8);
}
return ret;
}
uint32_t iso_read_msb(const uint8_t *buf, int bytes)
{
int i;
uint32_t ret = 0;
for (i=0; i<bytes; i++) {
ret += ((uint32_t) buf[bytes-i-1]) << (i*8);
}
return ret;
}
uint32_t iso_read_bb(const uint8_t *buf, int bytes)
{
uint32_t v1 = iso_read_lsb(buf, bytes);
uint32_t v2 = iso_read_msb(buf+bytes, bytes);
assert(v1 == v2);
return v1;
}

121
libisofs/util.h Executable file
View File

@ -0,0 +1,121 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set noet ts=8 sts=8 sw=8 : */
/**
* Utility functions for the Libisofs library.
*/
#ifndef LIBISO_UTIL_H
#define LIBISO_UTIL_H
#include <stdint.h>
#include <time.h>
#include <wchar.h>
#ifndef MAX
# define MAX(a, b) (((a) > (b)) ? (a) : (b))
#endif
#ifndef MIN
# define MIN(a, b) (((a) < (b)) ? (a) : (b))
#endif
extern inline int div_up(int n, int div)
{
return (n + div - 1) / div;
}
extern inline int round_up(int n, int mul)
{
return div_up(n, mul) * mul;
}
wchar_t *towcs(const char *);
char *str2ascii(const char*);
uint16_t *str2ucs(const char*);
/**
* Create a level 1 directory identifier.
*/
char *iso_1_dirid(const char *src);
/**
* Create a level 2 directory identifier.
*/
char *iso_2_dirid(const char *src);
/**
* Create a level 1 file identifier that consists of a name, extension and
* version number. The resulting string will have a file name of maximum
* length 8, followed by a separator (.), an optional extension of maximum
* length 3, followed by a separator (;) and a version number (digit 1).
* @return NULL if the original name and extension both are of length 0.
*/
char *iso_1_fileid(const char *src);
/**
* Create a level 2 file identifier that consists of a name, extension and
* version number. The combined file name and extension length will not exceed
* 30, the name and extension will be separated (.), and the extension will be
* followed by a separator (;) and a version number (digit 1).
* @return NULL if the original name and extension both are of length 0.
*/
char *iso_2_fileid(const char *src);
/**
* Create a Joliet file or directory identifier that consists of a name,
* extension and version number. The combined name and extension length will
* not exceed 128 bytes, the name and extension will be separated (.),
* and the extension will be followed by a separator (;) and a version number
* (digit 1). All characters consist of 2 bytes and the resulting string is
* NULL-terminated by a 2-byte NULL. Requires the locale to be set correctly.
*
* @param size will be set to the size (in bytes) of the identifier.
* @return NULL if the original name and extension both are of length 0 or the conversion from the current codeset to UCS-2BE is not available.
*/
uint16_t *iso_j_id(const char *src);
/**
* FIXME: what are the requirements for these next two? Is this for RR?
*
* Create a POSIX portable file name that consists of a name and extension.
* The resulting file name will not exceed 250 characters.
* @return NULL if the original name and extension both are of length 0.
*/
char *iso_p_fileid(const char *src);
/**
* Create a POSIX portable directory name.
* The resulting directory name will not exceed 250 characters.
* @return NULL if the original name is of length 0.
*/
char *iso_p_dirid(const char *src);
void iso_lsb(uint8_t *buf, uint32_t num, int bytes);
void iso_msb(uint8_t *buf, uint32_t num, int bytes);
void iso_bb(uint8_t *buf, uint32_t num, int bytes);
uint32_t iso_read_lsb(const uint8_t *buf, int bytes);
uint32_t iso_read_msb(const uint8_t *buf, int bytes);
uint32_t iso_read_bb(const uint8_t *buf, int bytes);
/** Records the date/time into a 7 byte buffer (9.1.5) */
void iso_datetime_7(uint8_t *buf, time_t t);
/** Records the date/time into a 17 byte buffer (8.4.26.1) */
void iso_datetime_17(uint8_t *buf, time_t t);
time_t iso_datetime_read_7(const uint8_t *buf);
time_t iso_datetime_read_17(const uint8_t *buf);
/**
* Like strlen, but for Joliet strings.
*/
size_t ucslen(const uint16_t *str);
/**
* Like strcmp, but for Joliet strings.
*/
int ucscmp(const uint16_t *s1, const uint16_t *s2);
#endif /* LIBISO_UTIL_H */

189
libisofs/volume.c Executable file
View File

@ -0,0 +1,189 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set ts=8 sts=8 sw=8 noet : */
#include <stdlib.h>
#include <string.h>
#include <libgen.h>
#include "libisofs.h"
#include "tree.h"
#include "util.h"
#include "volume.h"
struct iso_volset*
iso_volset_new(struct iso_volume *vol, const char *id)
{
struct iso_volset *volset = calloc(1, sizeof(struct iso_volset));
volset->volset_size = 1;
volset->refcount = 1;
volset->volume = malloc(sizeof(void *));
volset->volume[0] = vol;
volset->volset_id = strdup(id);
vol->refcount++;
return volset;
}
void
iso_volset_free(struct iso_volset *volset)
{
if (--volset->refcount < 1) {
int i;
for (i = 0; i < volset->volset_size; i++) {
iso_volume_free(volset->volume[i]);
}
free(volset->volume);
free(volset->volset_id);
}
}
struct iso_volume*
iso_volume_new(const char *volume_id,
const char *publisher_id,
const char *data_preparer_id)
{
return iso_volume_new_with_root(volume_id,
publisher_id,
data_preparer_id,
NULL);
}
struct iso_volume*
iso_volume_new_with_root(const char *volume_id,
const char *publisher_id,
const char *data_preparer_id,
struct iso_tree_node *root)
{
struct iso_volume *volume;
volume = calloc(1, sizeof(struct iso_volume));
volume->refcount = 1;
volume->root = root ? root : iso_tree_new_root(volume);
if (volume_id != NULL)
volume->volume_id = strdup(volume_id);
if (publisher_id != NULL)
volume->publisher_id = strdup(publisher_id);
if (data_preparer_id != NULL)
volume->data_preparer_id = strdup(data_preparer_id);
return volume;
}
void
iso_volume_free(struct iso_volume *volume)
{
/* Only free if no references are in use. */
if (--volume->refcount < 1) {
iso_tree_free(volume->root);
free(volume->volume_id);
free(volume->publisher_id);
free(volume->data_preparer_id);
free(volume);
}
}
struct iso_tree_node *
iso_volume_get_root(const struct iso_volume *volume)
{
return volume->root;
}
struct iso_tree_node *
iso_tree_volume_path_to_node(struct iso_volume *volume, const char *path)
{
struct iso_tree_node *node;
char *ptr, *brk_info, *component;
/* get the first child at the root of the volume
* that is "/" */
node=iso_volume_get_root(volume);
if (!strcmp (path, "/"))
return node;
if (!node->nchildren)
return NULL;
/* the name of the nodes is in wide characters so first convert path
* into wide characters. */
ptr = strdup(path);
/* get the first component of the path */
component=strtok_r(ptr, "/", &brk_info);
while (component) {
size_t max;
size_t i;
/* search among all the children of this directory if this path component exists */
max=node->nchildren;
for (i=0; i < max; i++) {
if (!strcmp(component, node->children[i]->name)) {
node=node->children[i];
break;
}
}
/* see if a node could be found */
if (i==max) {
node=NULL;
break;
}
component=strtok_r(NULL, "/", &brk_info);
}
free(ptr);
return node;
}
struct iso_tree_node *
iso_tree_volume_add_path(struct iso_volume *volume,
const char *disc_path,
const char *path)
{
char *tmp;
struct iso_tree_node *node;
struct iso_tree_node *parent_node;
tmp=strdup(disc_path);
parent_node = iso_tree_volume_path_to_node(volume, dirname(tmp));
free(tmp);
if (!parent_node)
return NULL;
node = iso_tree_radd_dir(parent_node, path);
if (!node)
return NULL;
tmp=strdup(disc_path);
iso_tree_node_set_name(node, basename(tmp));
free(tmp);
return node;
}
struct iso_tree_node *
iso_tree_volume_add_new_dir(struct iso_volume *volume,
const char *disc_path)
{
char *tmp;
struct iso_tree_node *node;
struct iso_tree_node *parent_node;
tmp=strdup(disc_path);
parent_node = iso_tree_volume_path_to_node(volume, dirname(tmp));
free(tmp);
if (!parent_node)
return NULL;
tmp=strdup(disc_path);
node = iso_tree_add_new_dir(parent_node, basename(tmp));
free(tmp);
return node;
}

45
libisofs/volume.h Executable file
View File

@ -0,0 +1,45 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set noet sts=8 ts=8 sw=8 : */
/**
* Extra declarations for use with the iso_volume structure.
*/
#ifndef LIBISO_VOLUME_H
#define LIBISO_VOLUME_H
#include "libisofs.h"
/**
* Data volume.
*/
struct iso_volume
{
int refcount; /**< Number of used references to this
volume. */
struct iso_tree_node *root; /**< Root of the directory tree for the
volume. */
char *volume_id; /**< Volume identifier. */
char *publisher_id; /**< Volume publisher. */
char *data_preparer_id; /**< Volume data preparer. */
};
/**
* A set of data volumes.
*/
struct iso_volset
{
int refcount;
struct iso_volume **volume; /**< The volumes belonging to this
volume set. */
int volset_size; /**< The number of volumes in this
volume set. */
char *volset_id; /**< The id of this volume set, encoded
in the current locale. */
};
#endif /* __ISO_VOLUME */

4
test/burn.c Normal file
View File

@ -0,0 +1,4 @@
int main()
{
return 0;
}

107
test/iso.c Normal file
View File

@ -0,0 +1,107 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
/* vim: set ts=8 sts=8 sw=8 noet : */
#define _GNU_SOURCE
#include "libisofs.h"
#include "libburn/libburn.h"
#include <getopt.h>
#include <stdlib.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <assert.h>
#include <sys/types.h>
#include <dirent.h>
#include <string.h>
#include <err.h>
#define SECSIZE 2048
const char * const optstring = "JRL:h";
extern char *optarg;
extern int optind;
void usage()
{
printf("test [OPTIONS] DIRECTORY OUTPUT\n");
}
void help()
{
printf(
"Options:\n"
" -J Add Joliet support\n"
" -R Add Rock Ridge support\n"
" -L <num> Set the ISO level (1 or 2)\n"
" -h Print this message\n"
);
}
int main(int argc, char **argv)
{
struct iso_volset *volset;
struct iso_volume *volume;
struct iso_tree_node *root;
struct burn_source *src;
unsigned char buf[2048];
FILE *fd;
int c;
int level=1, flags=0;
while ((c = getopt(argc, argv, optstring)) != -1) {
switch(c) {
case 'h':
usage();
help();
exit(0);
break;
case 'J':
flags |= ECMA119_JOLIET;
break;
case 'R':
flags |= ECMA119_ROCKRIDGE;
break;
case 'L':
level = atoi(optarg);
break;
case '?':
usage();
exit(1);
break;
}
}
if (argc < 2) {
printf ("must pass directory to build iso from\n");
usage();
return 1;
}
if (argc < 3) {
printf ("must supply output file\n");
usage();
return 1;
}
fd = fopen(argv[optind+1], "w");
if (!fd) {
err(1, "error opening output file");
}
root = iso_tree_radd_dir(NULL, argv[optind]);
if (!root) {
err(1, "error opening input directory");
}
volume = iso_volume_new_with_root( "VOLID", "PUBID", "PREPID", root );
volset = iso_volset_new( volume, "VOLSETID" );
src = iso_source_new_ecma119(volset, 0, level, flags);
while (src->read(src, buf, 2048) == 2048) {
fwrite(buf, 1, 2048, fd);
}
fclose(fd);
return 0;
}

297
test/iso.py Normal file
View File

@ -0,0 +1,297 @@
import struct
import tree
import sys
voldesc_fmt = "B" "5s" "B" "2041x"
# all these fields are common between the pri and sec voldescs
privoldesc_fmt = "B" "5s" "B" "x" "32s" "32s" "8x" "8s" "32x" "4s" "4s" "4s" "8s" "4s4s" "4s4s" "34s" "128s" \
"128s" "128s" "128s" "37s" "37s" "37s" "17s" "17s" "17s" "17s" "B" "x" "512s" "653x"
# the fields unique to the sec_vol_desc
secvoldesc_fmt = "x" "5x" "x" "B" "32x" "32x" "8x" "8x" "32s" "4x" "4x" "4x" "8x" "4x4x" "4x4x" "34x" "128x" \
"128x" "128x" "128x" "37x" "37x" "37x" "17x" "17x" "17x" "17x" "x" "x" "512x" "653x"
dirrecord_fmt = "B" "B" "8s" "8s" "7s" "B" "B" "B" "4s" "B" # + file identifier, padding field and SU area
pathrecord_fmt = "B" "B" "4s" "2s" # + directory identifier and padding field
def read_bb(str, le, be):
val1, = struct.unpack(le, str)
val2, = struct.unpack(be, str)
if val1 != val2:
print "val1=%d, val2=%d" % (val1, val2)
raise AssertionError, "values are not equal in dual byte-order field"
return val1
def read_bb4(str):
return read_bb(str, "<I4x", ">4xI")
def read_bb2(str):
return read_bb(str, "<H2x", ">2xH")
def read_lsb4(str):
return struct.unpack("<I", str)[0]
def read_lsb2(str):
return struct.unpack("<H", str)[0]
def read_msb4(str):
return struct.unpack(">I", str)[0]
def read_msb2(str):
return struct.unpack(">H", str)[0]
class VolDesc(object):
def __init__(self, data):
print "fmt len=%d, data len=%d" % ( struct.calcsize(voldesc_fmt), len(data) )
self.vol_desc_type, self.standard_id, self.vol_desc_version = struct.unpack(voldesc_fmt, data)
class PriVolDesc(VolDesc):
def __init__(self, data):
self.vol_desc_type, \
self.standard_id, \
self.vol_desc_version, \
self.system_id, \
self.volume_id, \
self.vol_space_size, \
self.vol_set_size, \
self.vol_seq_num, \
self.block_size, \
self.path_table_size, \
self.l_table_pos, \
self.l_table2_pos, \
self.m_table_pos, \
self.m_table2_pos, \
self.root_record, \
self.volset_id, \
self.publisher_id, \
self.preparer_id, \
self.application_id, \
self.copyright_file, \
self.abstract_file, \
self.bibliographic_file, \
self.creation_timestamp, \
self.modification_timestamp, \
self.expiration_timestamp, \
self.effective_timestamp, \
self.file_struct_version, \
self.application_use = struct.unpack(privoldesc_fmt, data)
# take care of reading the integer types
self.vol_space_size = read_bb4(self.vol_space_size)
self.vol_set_size = read_bb2(self.vol_set_size)
self.vol_seq_num = read_bb2(self.vol_seq_num)
self.block_size = read_bb2(self.block_size)
self.path_table_size = read_bb4(self.path_table_size)
self.l_table_pos = read_lsb4(self.l_table_pos)
self.l_table2_pos = read_lsb4(self.l_table2_pos)
self.m_table_pos = read_msb4(self.m_table_pos)
self.m_table2_pos = read_msb4(self.m_table2_pos)
# parse the root directory record
self.root_record = DirRecord(self.root_record)
def readPathTables(self, file):
file.seek( self.block_size * self.l_table_pos )
self.l_table = PathTable( file.read(self.path_table_size), 0 )
file.seek( self.block_size * self.m_table_pos )
self.m_table = PathTable( file.read(self.path_table_size), 1 )
if self.l_table2_pos:
file.seek( self.block_size * self.l_table2_pos )
self.l_table2 = PathTable( file.read(self.path_table_size), 0 )
else:
self.l_table2 = None
if self.m_table2_pos:
file.seek( self.block_size * self.m_table2_pos )
self.m_table2 = PathTable( file.read(self.path_table_size), 1 )
else:
self.m_table2 = None
def toTree(self, isofile):
ret = tree.Tree(isofile=isofile.name)
ret.root = self.root_record.toTreeNode(parent=None, isofile=isofile)
return ret
class SecVolDesc(PriVolDesc):
def __init__(self, data):
super(SecVolDesc,self).__init__(data)
self.flags, self.escape_sequences = struct.unpack(secvoldesc_fmt, data)
# return a single volume descriptor of the appropriate type
def readVolDesc(data):
desc = VolDesc(data)
if desc.standard_id != "CD001":
print "Unexpected standard_id " +desc.standard_id
return None
if desc.vol_desc_type == 1:
return PriVolDesc(data)
elif desc.vol_desc_type == 2:
return SecVolDesc(data)
elif desc.vol_desc_type == 3:
print "I don't know about partitions yet!"
return None
elif desc.vol_desc_type == 255:
return desc
else:
print "Unknown volume descriptor type %d" % (desc.vol_desc_type,)
return None
def readVolDescSet(file):
ret = [ readVolDesc(file.read(2048)) ]
while ret[-1].vol_desc_type != 255:
ret.append( readVolDesc(file.read(2048)) )
for vol in ret:
if vol.vol_desc_type == 1 or vol.vol_desc_type == 2:
vol.readPathTables(file)
return ret
class DirRecord:
def __init__(self, data):
self.len_dr, \
self.len_xa, \
self.block, \
self.len_data, \
self.timestamp, \
self.flags, \
self.unit_size, \
self.gap_size, \
self.vol_seq_number, \
self.len_fi = struct.unpack(dirrecord_fmt, data[:33])
self.children = []
if self.len_dr > len(data):
raise AssertionError, "Error: not enough data to read in DirRecord()"
elif self.len_dr < 34:
raise AssertionError, "Error: directory record too short"
fmt = str(self.len_fi) + "s"
if self.len_fi % 2 == 0:
fmt += "1x"
len_su = self.len_dr - (33 + self.len_fi + 1 - (self.len_fi % 2))
fmt += str(len_su) + "s"
if len(data) >= self.len_dr:
self.file_id, self.su = struct.unpack(fmt, data[33 : self.len_dr])
else:
print "Error: couldn't read file_id: not enough data"
self.file_id = "BLANK"
self.su = ""
# convert to integers
self.block = read_bb4(self.block)
self.len_data = read_bb4(self.len_data)
self.vol_seq_number = read_bb2(self.vol_seq_number)
def toTreeNode(self, parent, isofile, path=""):
ret = tree.TreeNode(parent=parent, isofile=isofile.name)
if len(path) > 0:
path += "/"
path += self.file_id
ret.path = path
if self.flags & 2: # we are a directory, recurse
isofile.seek( 2048 * self.block )
data = isofile.read( self.len_data )
pos = 0
while pos < self.len_data:
try:
child = DirRecord( data[pos:] )
pos += child.len_dr
if child.len_fi == 1 and (child.file_id == "\x00" or child.file_id == "\x01"):
continue
print "read child named " +child.file_id
self.children.append( child )
ret.children.append( child.toTreeNode(ret, isofile, path) )
except AssertionError:
print "Couldn't read child of directory %s, position is %d, len is %d" % \
(path, pos, self.len_data)
raise
return ret
class PathTableRecord:
def __init__(self, data, readint2, readint4):
self.len_di, self.len_xa, self.block, self.parent_number = struct.unpack(pathrecord_fmt, data[:8])
if len(data) < self.len_di + 8:
raise AssertionError, "Error: not enough data to read path table record"
fmt = str(self.len_di) + "s"
self.dir_id, = struct.unpack(fmt, data[8:8+self.len_di])
self.block = readint4(self.block)
self.parent_number = readint2(self.parent_number)
class PathTable:
def __init__(self, data, m_type):
if m_type:
readint2 = read_msb2
readint4 = read_msb4
else:
readint2 = read_lsb2
readint4 = read_lsb4
pos = 0
self.records = []
while pos < len(data):
try:
self.records.append( PathTableRecord(data[pos:], readint2, readint4) )
print "Read path record %d: dir_id %s, block %d, parent_number %d" %\
(len(self.records), self.records[-1].dir_id, self.records[-1].block, self.records[-1].parent_number)
pos += self.records[-1].len_di + 8
pos += pos % 2
except AssertionError:
print "Last successfully read path table record had dir_id %s, block %d, parent_number %d" % \
(self.records[-1].dir_id, self.records[-1].block, self.records[-1].parent_number)
print "Error was near offset %x" % (pos,)
raise
def findRecord(self, dir_id, block, parent_number):
number=1
for record in self.records:
if record.dir_id == dir_id and record.block == block and record.parent_number == parent_number:
return number, record
number += 1
return None, None
# check this path table for consistency against the actual directory heirarchy
def crossCheckDirRecords(self, root, parent_number=1):
number, rec = self.findRecord(root.file_id, root.block, parent_number)
if not rec:
print "Error: directory record parent_number %d, dir_id %s, block %d doesn't match a path table record" % \
(parent_number, root.file_id, root.block)
parent = self.records[parent_number]
print "Parent has parent_number %d, dir_id %s, block %d" % (parent.parent_number, parent.dir_id, parent.block)
return 0
for child in root.children:
if child.flags & 2:
self.crossCheckDirRecords(child, number)
if len(sys.argv) != 2:
print "Please enter the name of the .iso file to open"
sys.exit(1)
f = file(sys.argv[1])
f.seek(2048 * 16) # system area
volumes = readVolDescSet(f)
vol = volumes[0]
t = vol.toTree(f)
vol.l_table.crossCheckDirRecords(vol.root_record)
vol.m_table.crossCheckDirRecords(vol.root_record)
vol = volumes[1]
try:
t = vol.toTree(f)
vol.l_table.crossCheckDirRecords(vol.root_record)
vol.m_table.crossCheckDirRecords(vol.root_record)
except AttributeError:
pass

View File

@ -1,17 +1,16 @@
/* test/libburner.c , API illustration of burning data or audio tracks to CD */
/* Copyright (C) 2005 - 2006 Thomas Schmitt <scdbackup@gmx.net> */
/* Provided under GPL, see also "License and copyright aspects" at file end */
/* test/libburner.c , API illustration of burning a single data track to CD */
/* Copyright (C) 2005 - 2006 Thomas Schmitt <scdbackup@gmx.net> */
/* Provided under GPL, see also "License and copyright aspects" at file end */
/** Overview
libburner is a minimal demo application for the library libburn as provided
on http://libburnia.pykix.org . It can list the available devices, can
blank a CD-RW, can format a DVD-RW, and can burn to CD-R, CD-RW, DVD+RW,
DVD-RAM or DVD-RW.
on http://libburn.pykix.org . It can list the available devices, can
blank a CD-RW and can burn to CD-R or CD-RW.
It's main purpose, nevertheless, is to show you how to use libburn and also
to serve the libburnia team as reference application. libburner.c does indeed
to serve the libburn team as reference application. libburner.c does indeed
define the standard way how above three gestures can be implemented and
stay upward compatible for a good while.
@ -26,9 +25,9 @@
libburner_aquire_by_driveno() demonstrates a scan-and-choose approach
With that aquired drive you can blank a CD-RW
libburner_blank_disc()
or you can format a DVD-RW to profile "Restricted Overwrite" (needed once)
libburner_format_row()
With the aquired drive you can burn to CD-R, CD-RW, DVD+RW, DVD-RAM, DVD-RW
Between blanking and burning one eventually has to reload the drive status
libburner_regrab()
With the aquired drive you can burn to CD-R or blank CD-RW
libburner_payload()
When everything is done, main() releases the drive and shuts down libburn:
burn_drive_release();
@ -50,8 +49,6 @@
#include <stdlib.h>
#include <time.h>
#include <errno.h>
#include <sys/stat.h>
#include <fcntl.h>
/** For simplicity i use global variables to represent the drives.
@ -71,10 +68,6 @@ static unsigned int drive_count;
finally released */
static int drive_is_grabbed = 0;
/** A number and a text describing the type of media in aquired drive */
static int current_profile= -1;
static char current_profile_name[80]= {""};
/* Some in-advance definitions to allow a more comprehensive ordering
of the functions and their explanations in here */
@ -103,13 +96,7 @@ int libburner_aquire_drive(char *drive_adr, int *driveno)
ret = libburner_aquire_by_adr(drive_adr);
else
ret = libburner_aquire_by_driveno(driveno);
if (ret <= 0)
return ret;
burn_disc_get_profile(drive_list[0].drive, &current_profile,
current_profile_name);
if (current_profile_name[0])
printf("Detected media type: %s\n", current_profile_name);
return 1;
return ret;
}
@ -120,21 +107,15 @@ int libburner_aquire_drive(char *drive_adr, int *driveno)
int libburner_aquire_by_adr(char *drive_adr)
{
int ret;
char libburn_drive_adr[BURN_DRIVE_ADR_LEN];
/* This tries to resolve links or alternative device files */
ret = burn_drive_convert_fs_adr(drive_adr, libburn_drive_adr);
if (ret<=0) {
fprintf(stderr,"Address does not lead to a CD burner: '%s'\n",
drive_adr);
return ret;
}
printf("Aquiring drive '%s' ...\n",libburn_drive_adr);
ret = burn_drive_scan_and_grab(&drive_list,libburn_drive_adr,1);
printf("Aquiring drive '%s' ...\n",drive_adr);
ret = burn_drive_scan_and_grab(&drive_list,drive_adr,1);
if (ret <= 0) {
fprintf(stderr,"FAILURE with persistent drive address '%s'\n",
libburn_drive_adr);
drive_adr);
if (strncmp(drive_adr,"/dev/sg",7) != 0 &&
strncmp(drive_adr,"/dev/hd",7) != 0)
fprintf(stderr,"\nHINT: Consider addresses like '/dev/hdc' or '/dev/sg0'\n");
} else {
printf("Done\n");
drive_is_grabbed = 1;
@ -194,6 +175,7 @@ int libburner_aquire_by_driveno(int *driveno)
}
printf("-----------------------------------------------------------------------------\n\n");
/*
On multi-drive systems save yourself from sysadmins' revenge.
@ -260,16 +242,19 @@ int libburner_aquire_by_driveno(int *driveno)
int libburner_blank_disc(struct burn_drive *drive, int blank_fast)
{
enum burn_disc_status disc_state;
struct burn_progress p;
int percent = 1;
struct burn_progress progress;
disc_state = burn_disc_get_status(drive);
while (burn_drive_get_status(drive, NULL) != BURN_DRIVE_IDLE)
usleep(1001);
while ((disc_state = burn_disc_get_status(drive)) == BURN_DISC_UNREADY)
usleep(1001);
printf(
"Drive media status: %d (see libburn/libburn.h BURN_DISC_*)\n",
disc_state);
if (disc_state == BURN_DISC_BLANK) {
fprintf(stderr,
"IDLE: Blank media detected. Will leave it untouched\n");
"IDLE: Blank CD media detected. Will leave it untouched\n");
return 2;
} else if (disc_state == BURN_DISC_FULL ||
disc_state == BURN_DISC_APPENDABLE) {
@ -279,7 +264,7 @@ int libburner_blank_disc(struct burn_drive *drive, int blank_fast)
return 0;
} else {
fprintf(stderr,
"FATAL: Unsuitable drive and media state\n");
"FATAL: Cannot recognize drive and media state\n");
return 0;
}
if(!burn_disc_erasable(drive)) {
@ -288,14 +273,12 @@ int libburner_blank_disc(struct burn_drive *drive, int blank_fast)
return 0;
}
printf(
"Beginning to %s-blank media.\n", (blank_fast?"fast":"full"));
"Beginning to %s-blank CD media.\n", (blank_fast?"fast":"full"));
printf(
"Expect some garbage sector numbers and some zeros at first.\n");
burn_disc_erase(drive, blank_fast);
sleep(1);
while (burn_drive_get_status(drive, &p) != BURN_DRIVE_IDLE) {
if(p.sectors>0 && p.sector>=0) /* display 1 to 99 percent */
percent = 1.0 + ((double) p.sector+1.0)
/ ((double) p.sectors) * 98.0;
printf("Blanking ( %d%% done )\n", percent);
while (burn_drive_get_status(drive, &progress) != BURN_DRIVE_IDLE) {
printf("Blanking sector %d\n", progress.sector);
sleep(1);
}
printf("Done\n");
@ -303,44 +286,24 @@ int libburner_blank_disc(struct burn_drive *drive, int blank_fast)
}
/** Persistently changes DVD-RW profile 0014h "Sequential Recording"
to profile 0013h "Restricted Overwrite" which is usable with libburner.
Expect a behavior similar to blanking with unusual noises from the drive.
/** This gesture is necessary to get the drive info after blanking.
It opens a small gap for losing the drive to another libburn instance.
We will work on closing this gap.
*/
int libburner_format_row(struct burn_drive *drive)
{
struct burn_progress p;
int percent = 1;
int libburner_regrab(struct burn_drive *drive) {
int ret;
if (current_profile == 0x13) {
fprintf(stderr, "IDLE: DVD-RW media is already formatted\n");
return 2;
} else if (current_profile != 0x14) {
fprintf(stderr, "FATAL: Can only format DVD-RW\n");
return 0;
}
printf("Beginning to format media.\n");
burn_disc_format(drive, (off_t) 0, 0);
sleep(1);
while (burn_drive_get_status(drive, &p) != BURN_DRIVE_IDLE) {
if(p.sectors>0 && p.sector>=0) /* display 1 to 99 percent */
percent = 1.0 + ((double) p.sector+1.0)
/ ((double) p.sectors) * 98.0;
printf("Formatting ( %d%% done )\n", percent);
sleep(1);
}
burn_disc_get_profile(drive_list[0].drive, &current_profile,
current_profile_name);
printf("Media type now: %4.4xh \"%s\"\n",
current_profile, current_profile_name);
if (current_profile != 0x13) {
fprintf(stderr,
"FATAL: Failed to change media profile to desired value\n");
return 0;
}
return 1;
printf("Releasing and regrabbing drive ...\n");
if (drive_is_grabbed)
burn_drive_release(drive, 0);
drive_is_grabbed = 0;
ret = burn_drive_grab(drive, 0);
if (ret != 0) {
drive_is_grabbed = 1;
printf("Done\n");
} else
printf("FAILED\n");
return !!ret;
}
@ -354,7 +317,7 @@ int libburner_format_row(struct burn_drive *drive)
*/
int libburner_payload(struct burn_drive *drive,
char source_adr[][4096], int source_adr_count,
int multi, int simulate_burn, int all_tracks_type)
off_t stdin_size, int simulate_burn, int all_tracks_type)
{
struct burn_source *data_src;
struct burn_disc *target_disc;
@ -364,10 +327,8 @@ int libburner_payload(struct burn_drive *drive,
struct burn_track *track, *tracklist[99];
struct burn_progress progress;
time_t start_time;
int last_sector = 0, padding = 0, trackno, write_mode_tao = 0, fd;
off_t fixed_size;
int last_sector = 0, padding = 0, trackno;
char *adr;
struct stat stbuf;
if (all_tracks_type != BURN_AUDIO) {
all_tracks_type = BURN_MODE1;
@ -384,21 +345,12 @@ int libburner_payload(struct burn_drive *drive,
burn_track_define_data(track, 0, padding, 1, all_tracks_type);
adr = source_adr[trackno];
fixed_size = 0;
if (adr[0] == '-' && adr[1] == 0) {
fd = 0;
} else {
fd = open(adr, O_RDONLY);
if (fd>=0)
if (fstat(fd,&stbuf)!=-1)
if((stbuf.st_mode&S_IFMT)==S_IFREG)
fixed_size = stbuf.st_size;
}
if (fixed_size==0)
write_mode_tao = 1;
data_src = NULL;
if (fd>=0)
data_src = burn_fd_source_new(fd, -1, fixed_size);
data_src = burn_fd_source_new(0, -1, stdin_size);
printf("Note: using standard input as source with %.f bytes\n",
(double) stdin_size);
} else
data_src = burn_file_source_new(adr, NULL);
if (data_src == NULL) {
fprintf(stderr,
"FATAL: Could not open data source '%s'.\n",adr);
@ -413,36 +365,49 @@ int libburner_payload(struct burn_drive *drive,
}
burn_session_add_track(session, track, BURN_POS_END);
printf("Track %d : source is '%s'\n", trackno+1, adr);
printf("Track %d : source is '%s'\n", trackno, adr);
burn_source_free(data_src);
} /* trackno loop end */
while (burn_drive_get_status(drive, NULL) != BURN_DRIVE_IDLE)
usleep(100001);
/* Evaluate drive and media */
disc_state = burn_disc_get_status(drive);
if (disc_state == BURN_DISC_APPENDABLE) {
write_mode_tao = 1;
} else if (disc_state != BURN_DISC_BLANK) {
if (disc_state == BURN_DISC_FULL) {
fprintf(stderr, "FATAL: Closed media with data detected. Need blank or appendable media.\n");
while ((disc_state = burn_disc_get_status(drive)) == BURN_DISC_UNREADY)
usleep(100001);
if (disc_state != BURN_DISC_BLANK) {
if (disc_state == BURN_DISC_FULL ||
disc_state == BURN_DISC_APPENDABLE) {
fprintf(stderr,
"FATAL: Media with data detected. Need blank media.\n");
if (burn_disc_erasable(drive))
fprintf(stderr, "HINT: Try --blank_fast\n\n");
} else if (disc_state == BURN_DISC_EMPTY)
fprintf(stderr,"FATAL: No media detected in drive\n");
else
fprintf(stderr,
"FATAL: Cannot recognize state of drive and media\n");
"FATAL: Cannot recognize drive and media state\n");
return 0;
}
burn_options = burn_write_opts_new(drive);
burn_write_opts_set_perform_opc(burn_options, 0);
burn_write_opts_set_multi(burn_options, !!multi);
if (write_mode_tao)
burn_write_opts_set_write_type(burn_options,
BURN_WRITE_TAO, BURN_BLOCK_MODE1);
else
burn_write_opts_set_write_type(burn_options,
BURN_WRITE_SAO, BURN_BLOCK_SAO);
#ifdef Libburner_raw_mode_which_i_do_not_likE
/* This yields higher CD capacity but hampers my IDE controller
with burning on one drive and reading on another simultaneously.
My burner does not obey the order --try_to_simulate in this mode.
*/
burn_write_opts_set_write_type(burn_options,
BURN_WRITE_RAW, BURN_BLOCK_RAW96R);
#else
/* This is by what cdrskin competes with cdrecord -sao which
i understand is the mode preferrably advised by Joerg Schilling */
burn_write_opts_set_write_type(burn_options,
BURN_WRITE_SAO, BURN_BLOCK_SAO);
#endif
if(simulate_burn)
printf("\n*** Will TRY to SIMULATE burning ***\n\n");
burn_write_opts_set_simulate(burn_options, simulate_burn);
@ -462,11 +427,8 @@ int libburner_payload(struct burn_drive *drive,
printf(
"Thank you for being patient since %d seconds.\n",
(int) (time(0) - start_time));
else if(write_mode_tao)
printf("Track %d : sector %d\n", progress.track+1,
progress.sector);
else
printf("Track %d : sector %d of %d\n",progress.track+1,
printf("Track %d : sector %d of %d\n", progress.track,
progress.sector, progress.sectors);
last_sector = progress.sector;
sleep(1);
@ -477,22 +439,19 @@ int libburner_payload(struct burn_drive *drive,
burn_track_free(tracklist[trackno]);
burn_session_free(session);
burn_disc_free(target_disc);
if (multi && current_profile != 0x1a && current_profile != 0x13 &&
current_profile != 0x12) /* not with DVD+RW, DVD-RW, DVD-RAM */
printf("NOTE: Media left appendable.\n");
if (simulate_burn)
if(simulate_burn)
printf("\n*** Did TRY to SIMULATE burning ***\n\n");
return 1;
return 0;
}
/** The setup parameters of libburner */
static char drive_adr[BURN_DRIVE_ADR_LEN] = {""};
static int driveno = 0;
static int do_blank = 0;
/** The setup parameters of libburn */
static char drive_adr[BURN_DRIVE_ADR_LEN]= {""};
static int driveno= 0;
static int do_blank= 0;
static char source_adr[99][4096];
static int source_adr_count = 0;
static int do_multi = 0;
static int source_adr_count= 0;
static off_t stdin_size= 650*1024*1024;
static int simulate_burn = 0;
static int all_tracks_type = BURN_MODE1;
@ -537,15 +496,15 @@ int libburner_setup(int argc, char **argv)
}
strcpy(drive_adr, argv[i]);
}
} else if (!strcmp(argv[i], "--format_overwrite")) {
do_blank = 101;
} else if (!strcmp(argv[i], "--multi")) {
do_multi = 1;
} else if (!strcmp(argv[i], "--stdin_size")) { /* obsoleted */
i++;
} else if (!strcmp(argv[i], "--stdin_size")) {
++i;
if (i >= argc) {
fprintf(stderr,"--stdin_size requires an argument\n");
return 3;
} else
stdin_size = atoi(argv[i]);
if (stdin_size < 600*1024) /* minimum readable track size */
stdin_size = 600*1024;
} else if (!strcmp(argv[i], "--try_to_simulate")) {
simulate_burn = 1;
@ -577,29 +536,29 @@ int libburner_setup(int argc, char **argv)
insuffient_parameters = 0;
if (print_help || insuffient_parameters ) {
printf("Usage: %s\n", argv[0]);
printf(" [--drive <address>|<driveno>|\"-\"] [--audio]\n");
printf(" [--blank_fast|--blank_full|--format_overwrite]\n");
printf(" [--try_to_simulate]\n");
printf(" [--multi] [<one or more imagefiles>|\"-\"]\n");
printf(" [--drive <address>|<driveno>|\"-\"]\n");
printf(" [--blank_fast|--blank_full] [--audio]\n");
printf(" [--try_to_simulate] [--stdin_size <bytes>]\n");
printf(" [<one or more imagefiles>|\"-\"]\n");
printf("Examples\n");
printf("A bus scan (needs rw-permissions to see a drive):\n");
printf(" %s --drive -\n",argv[0]);
printf("Burn a file to drive chosen by number, leave appendable:\n");
printf(" %s --drive 0 --multi my_image_file\n", argv[0]);
printf("Burn a file to drive chosen by persistent address, close:\n");
printf("Burn a file to drive chosen by number:\n");
printf(" %s --drive 0 my_image_file\n",
argv[0]);
printf("Burn a file to drive chosen by persistent address:\n");
printf(" %s --drive /dev/hdc my_image_file\n", argv[0]);
printf("Blank a used CD-RW (is combinable with burning in one run):\n");
printf(" %s --drive /dev/hdc --blank_fast\n",argv[0]);
printf("Format a DVD-RW once before first use with libburner:\n");
printf(" %s --drive /dev/hdc --format_overwrite\n", argv[0]);
printf("Burn two audio tracks:\n");
printf("Burn two audio tracks\n");
printf(" lame --decode -t /path/to/track1.mp3 track1.cd\n");
printf(" test/dewav /path/to/track2.wav -o track2.cd\n");
printf(" %s --drive /dev/hdc --audio track1.cd track2.cd\n", argv[0]);
printf("Burn a compressed afio archive on-the-fly:\n");
printf("Burn a compressed afio archive on-the-fly, pad up to 700 MB:\n");
printf(" ( cd my_directory ; find . -print | afio -oZ - ) | \\\n");
printf(" %s --drive /dev/hdc -\n", argv[0]);
printf("To be read from *not mounted* media via: afio -tvZ /dev/hdc\n");
printf(" %s --drive /dev/hdc --stdin_size 734003200 -\n", argv[0]);
printf("To be read from *not mounted* CD via: afio -tvZ /dev/hdc\n");
printf("Program tar would need a clean EOF which our padded CD cannot deliver.\n");
if (insuffient_parameters)
return 6;
}
@ -615,7 +574,7 @@ int main(int argc, char **argv)
if (ret)
exit(ret);
printf("Initializing libburnia.pykix.org ...\n");
printf("Initializing libburn.pykix.org ...\n");
if (burn_initialize())
printf("Done\n");
else {
@ -640,18 +599,22 @@ int main(int argc, char **argv)
if (ret == 2)
{ ret = 0; goto release_drive; }
if (do_blank) {
if (do_blank > 100)
ret = libburner_format_row(drive_list[driveno].drive);
else
ret = libburner_blank_disc(drive_list[driveno].drive,
do_blank == 1);
ret = libburner_blank_disc(drive_list[driveno].drive,
do_blank == 1);
if (ret<=0)
{ ret = 36; goto release_drive; }
if (ret != 2 && source_adr_count > 0)
ret = libburner_regrab(drive_list[driveno].drive);
if (ret<=0) {
fprintf(stderr,
"FATAL: Cannot release and grab again drive after blanking\n");
{ ret = 37; goto finish_libburn; }
}
}
if (source_adr_count > 0) {
ret = libburner_payload(drive_list[driveno].drive,
source_adr, source_adr_count,
do_multi, simulate_burn, all_tracks_type);
source_adr, source_adr_count, stdin_size,
simulate_burn, all_tracks_type);
if (ret<=0)
{ ret = 38; goto release_drive; }
}
@ -679,7 +642,7 @@ Be also invited to study the code of cdrskin/cdrskin.c et al.
Clarification in my name and in the name of Mario Danic, copyright holder
on toplevel of libburnia. To be fully in effect after the remaining other
on toplevel of libburn. To be fully in effect after the remaining other
copyrighted code has been replaced by ours and by copyright-free contributions
of our friends:

126
test/master.c Normal file
View File

@ -0,0 +1,126 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
#include <libburn/libburn.h>
#include <stdio.h>
#include <assert.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>
#include <string.h>
#include <stdlib.h>
static struct burn_drive_info *drives;
static unsigned int n_drives;
void burn_files(struct burn_drive *drive, struct burn_disc *disc)
{
struct burn_write_opts *o;
enum burn_disc_status s;
if (!burn_drive_grab(drive, 1)) {
printf("Unable to open the drive!\n");
return;
}
while (burn_drive_get_status(drive, NULL))
usleep(1000);
while ((s = burn_disc_get_status(drive)) == BURN_DISC_UNREADY)
usleep(1000);
if (s != BURN_DISC_BLANK) {
burn_drive_release(drive, 0);
printf("Please insert blank media in the drive\n");
return;
}
o = burn_write_opts_new(drive);
burn_drive_set_speed(drive, 0, 2816);
burn_write_opts_set_perform_opc(o, 0);
burn_write_opts_set_write_type(o, BURN_WRITE_TAO, BURN_BLOCK_MODE1);
burn_write_opts_set_simulate(o, 1);
/* want failure on seggy while debugging :) */
burn_write_opts_set_underrun_proof(o, 0);
burn_structure_print_disc(disc);
burn_disc_write(o, disc);
burn_write_opts_free(o);
while (burn_drive_get_status(drive, NULL)) {
sleep(1);
}
printf("\n");
burn_drive_release(drive, 0);
burn_disc_free(disc);
}
void parse_args(int argc, char **argv, int *drive, struct burn_disc **disc)
{
int i, tmode = BURN_AUDIO;
int help = 1;
struct burn_session *session;
struct burn_track *tr;
struct burn_source *src;
*disc = burn_disc_create();
session = burn_session_create();
burn_disc_add_session(*disc, session, BURN_POS_END);
for (i = 1; i < argc; ++i) {
if (!strcmp(argv[i], "--drive")) {
++i;
if (i >= argc)
printf("--drive requires an argument\n");
else
*drive = atoi(argv[i]);
} else if (!strcmp(argv[i], "--audio")) {
tmode = BURN_AUDIO;
} else if (!strcmp(argv[i], "--data")) {
tmode = BURN_MODE1;
} else if (!strcmp(argv[i], "--verbose")) {
++i;
if (i >= argc)
printf("--verbose requires an argument\n");
else
burn_set_verbosity(atoi(argv[i]));
} else if (!strcmp(argv[i], "--help")) {
help = 1; /* who cares */
} else {
help = 0;
printf("%s is a track\n", argv[i]);
tr = burn_track_create();
src = burn_file_source_new(argv[i], NULL);
burn_track_set_source(tr, src);
burn_source_free(src);
burn_track_define_data(tr, 0, 0, 1, tmode);
burn_session_add_track(session, tr, BURN_POS_END);
}
}
if (help) {
printf("Usage: %s [--drive <num>] [--verbose <level>] files\n",
argv[0]);
exit(EXIT_FAILURE);
}
}
int main(int argc, char **argv)
{
int drive = 0;
struct burn_disc *disc;
parse_args(argc, argv, &drive, &disc);
printf("Initializing library...");
if (burn_initialize())
printf("Success\n");
else {
printf("Failed\n");
return 1;
}
printf("Scanning for devices...");
while (!burn_drive_scan(&drives, &n_drives)) ;
printf("Done\n");
burn_files(drives[drive].drive, disc);
burn_drive_info_free(drives);
burn_finish();
return 0;
}

54
test/rip.c Normal file
View File

@ -0,0 +1,54 @@
/* THIS IS NOT A PROPER EXAMPLE */
#include <stdlib.h>
#include <stdio.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>
#include <libburn/libburn.h>
static struct burn_drive_info *drives;
static unsigned int n_drives;
#warning this example is totally fried
int main()
{
#if 0
struct burn_drive *drive;
struct burn_read_opts o;
burn_initialize();
o.datafd =
open("/xp/burn/blah.data", O_CREAT | O_WRONLY,
S_IRUSR | S_IWUSR);
o.subfd =
open("/xp/burn/blah.sub", O_CREAT | O_WRONLY,
S_IRUSR | S_IWUSR);
o.raw = 1;
o.c2errors = 0;
o.subcodes_audio = 1;
o.subcodes_data = 1;
o.hardware_error_recovery = 1;
o.report_recovered_errors = 0;
o.transfer_damaged_blocks = 1;
o.hardware_error_retries = 1;
printf("Scanning for devices...");
while (!burn_drive_scan(&drives, &n_drives)) ;
printf("Done\n");
drive = drives[0].drive;
burn_drive_set_speed(drive, 0, 0);
if (!burn_drive_grab(drive, 1)) {
printf("Unable to open the drive!\n");
return EXIT_FAILURE;
}
while (burn_drive_get_status(drive, NULL))
usleep(1000);
burn_disc_read(drive, &o);
#endif
return EXIT_SUCCESS;
}

View File

@ -1,684 +0,0 @@
/* test/telltoc.c , API illustration of obtaining media status info */
/* Copyright (C) 2006 Thomas Schmitt <scdbackup@gmx.net> Provided under GPL */
/** Overview
telltoc is a minimal demo application for the library libburn as provided
on http://libburnia.pykix.org . It can list the available devices, can
display some drive properties, the type of media, eventual table of content
and multisession info for mkisofs option -C .
It's main purpose, nevertheless, is to show you how to use libburn and also
to serve the libburn team as reference application. telltoc.c does indeed
define the standard way how above gestures can be implemented and stay upward
compatible for a good while.
Before you can do anything, you have to initialize libburn by
burn_initialize()
as it is done in main() at the end of this file. Then you aquire a
drive in an appropriate way conforming to the API. The two main
approaches are shown here in application functions:
telltoc_aquire_by_adr() demonstrates usage as of cdrecord traditions
telltoc_aquire_by_driveno() demonstrates a scan-and-choose approach
With that aquired drive you can call
telltoc_media() prints some information about the media in a drive
telltoc_toc() prints a table of content (if there is content)
telltoc_msinfo() prints parameters for mkisofs option -C
When everything is done, main() releases the drive and shuts down libburn:
burn_drive_release();
burn_finish()
*/
/** See this for the decisive API specs . libburn.h is The Original */
/* For using the installed header file : #include <libburn/libburn.h> */
/* This program insists in the own headerfile. */
#include "../libburn/libburn.h"
/* libburn is intended for Linux systems with kernel 2.4 or 2.6 for now */
#include <stdio.h>
#include <ctype.h>
#include <sys/types.h>
#include <unistd.h>
#include <string.h>
#include <stdlib.h>
#include <time.h>
#include <errno.h>
#include <sys/stat.h>
#include <fcntl.h>
/** For simplicity i use global variables to represent the drives.
Drives are systemwide global, so we do not give away much of good style.
*/
/** This list will hold the drives known to libburn. This might be all CD
drives of the system and thus might impose severe impact on the system.
*/
static struct burn_drive_info *drive_list;
/** If you start a long lasting operation with drive_count > 1 then you are
not friendly to the users of other drives on those systems. Beware. */
static unsigned int drive_count;
/** This variable indicates wether the drive is grabbed and must be
finally released */
static int drive_is_grabbed = 0;
/* Some in-advance definitions to allow a more comprehensive ordering
of the functions and their explanations in here */
int telltoc_aquire_by_adr(char *drive_adr);
int telltoc_aquire_by_driveno(int *drive_no, int silent);
/* ------------------------------- API gestures ---------------------------- */
/** You need to aquire a drive before burning. The API offers this as one
compact call and alternatively as application controllable gestures of
whitelisting, scanning for drives and finally grabbing one of them.
If you have a persistent address of the drive, then the compact call is
to prefer because it only touches one drive. On modern Linux kernels,
there should be no fatal disturbance of ongoing burns of other libburn
instances with any of our approaches. We use open(O_EXCL) by default.
On /dev/hdX it should cooperate with growisofs and some cdrecord variants.
On /dev/sgN versus /dev/scdM expect it not to respect other programs.
*/
int telltoc_aquire_drive(char *drive_adr, int *driveno, int silent_drive)
{
int ret;
if(drive_adr != NULL && drive_adr[0] != 0)
ret = telltoc_aquire_by_adr(drive_adr);
else
ret = telltoc_aquire_by_driveno(driveno, silent_drive);
return ret;
}
/** If the persistent drive address is known, then this approach is much
more un-obtrusive to the systemwide livestock of drives. Only the
given drive device will be opened during this procedure.
*/
int telltoc_aquire_by_adr(char *drive_adr)
{
int ret;
char libburn_drive_adr[BURN_DRIVE_ADR_LEN];
/* This tries to resolve links or alternative device files */
ret = burn_drive_convert_fs_adr(drive_adr, libburn_drive_adr);
if (ret<=0) {
fprintf(stderr,"Address does not lead to a CD burner: '%s'\n",
drive_adr);
return ret;
}
fprintf(stderr,"Aquiring drive '%s' ...\n",libburn_drive_adr);
ret = burn_drive_scan_and_grab(&drive_list,libburn_drive_adr,1);
if (ret <= 0) {
fprintf(stderr,"FAILURE with persistent drive address '%s'\n",
libburn_drive_adr);
} else {
fprintf(stderr,"Done\n");
drive_is_grabbed = 1;
}
return ret;
}
/** This method demonstrates how to use libburn without knowing a persistent
drive address in advance. It has to make sure that after assessing the list
of available drives, all unwanted drives get closed again. As long as they
are open, no other libburn instance can see them. This is an intended
locking feature. The application is responsible for giving up the locks
by either burn_drive_release() (only after burn_drive_grab() !),
burn_drive_info_forget(), burn_drive_info_free(), or burn_finish().
@param driveno the index number in libburn's drive list. This will get
set to 0 on success and will then be the drive index to
use in the further dourse of processing.
@param silent_drive 1=do not print "Drive found :" line with *driveno >= 0
@return 1 success , <= 0 failure
*/
int telltoc_aquire_by_driveno(int *driveno, int silent_drive)
{
char adr[BURN_DRIVE_ADR_LEN];
int ret, i;
fprintf(stderr, "Beginning to scan for devices ...\n");
while (!burn_drive_scan(&drive_list, &drive_count))
usleep(1002);
if (drive_count <= 0 && *driveno >= 0) {
fprintf(stderr, "FAILED (no drives found)\n");
return 0;
}
fprintf(stderr, "Done\n");
for (i = 0; i < drive_count; i++) {
if (*driveno >= 0 && (silent_drive || *driveno != i))
continue;
if (burn_drive_get_adr(&(drive_list[i]), adr) <=0)
strcpy(adr, "-get_adr_failed-");
printf("Drive found : %d --drive '%s' : ", i,adr);
printf("%-8s %-16s (%4s)\n",
drive_list[i].vendor,drive_list[i].product,
drive_list[i].revision);
}
if (*driveno < 0) {
fprintf(stderr,
"Pseudo-drive \"-\" given : bus scanning done.\n");
return 2; /* the program will end after this */
}
/* We already made our choice via command line. (default is 0)
So we just have to keep our desired drive and drop all others.
*/
if (drive_count <= *driveno) {
fprintf(stderr,
"Found only %d drives. Number %d not available.\n",
drive_count, *driveno);
return 0; /* the program will end after this */
}
/* Drop all drives which we do not want to use */
for (i = 0; i < drive_count; i++) {
if (i == *driveno) /* the one drive we want to keep */
continue;
ret = burn_drive_info_forget(&(drive_list[i]),0);
if (ret != 1)
fprintf(stderr, "Cannot drop drive %d. Please report \"ret=%d\" to libburn-hackers@pykix.org\n",
i, ret);
else
fprintf(stderr, "Dropped unwanted drive %d\n",i);
}
/* Make the one we want ready for inquiry */
ret= burn_drive_grab(drive_list[*driveno].drive, 1);
if (ret != 1)
return 0;
drive_is_grabbed = 1;
return 1;
}
/** This gesture is necessary to get my NEC DVD_RW ND-4570A out of a state
of noisy overexcitement after it was inquired for Next Writeable Address.
The noise then still lasts 20 seconds. Same with cdrecord -toc, btw.
It opens a small gap for losing the drive to another libburn instance.
Not a problem in telltoc. This is done as very last drive operation.
Eventually the other libburn instance will have the same sanitizing effect.
*/
int telltoc_regrab(struct burn_drive *drive) {
int ret;
if (drive_is_grabbed)
burn_drive_release(drive, 0);
drive_is_grabbed = 0;
ret = burn_drive_grab(drive, 0);
if (ret != 0) {
drive_is_grabbed = 1;
}
return !!ret;
}
int telltoc_media(struct burn_drive *drive)
{
int ret, media_found = 0, profile_no = -1;
double max_speed = 0.0, min_speed = 0.0, speed_conv;
enum burn_disc_status s;
char profile_name[80], speed_unit[40];
printf("Media current: ");
ret = burn_disc_get_profile(drive, &profile_no, profile_name);
if (profile_no > 0 && ret >0) {
if (profile_name[0])
printf("%s\n", profile_name);
else
printf("%4.4Xh\n", profile_no);
} else
printf("is not recognizable\n");
speed_conv = 176.4;
strcpy(speed_unit,"176.4 kB/s (CD, data speed 150 KiB/s)");
if (strstr(profile_name, "DVD") == profile_name) {
speed_conv = 1385.0;
strcpy(speed_unit,"1385.0 kB/s (DVD)");
}
/* >>> libburn does not obtain full profile list yet */
printf("Media status : ");
s = burn_disc_get_status(drive);
if (s==BURN_DISC_FULL) {
printf("is written , is closed\n");
media_found = 1;
} else if (s==BURN_DISC_APPENDABLE) {
printf("is written , is appendable\n");
media_found = 1;
} else if (s==BURN_DISC_BLANK) {
printf("is blank\n");
media_found = 1;
} else if (s==BURN_DISC_EMPTY)
printf("is not present\n");
else
printf("is not recognizable\n");
printf("Media reuse : ");
if (media_found) {
if (burn_disc_erasable(drive))
printf("is erasable\n");
else
printf("is not erasable\n");
} else
printf("is not recognizable\n");
ret= burn_drive_get_write_speed(drive);
max_speed = ((double ) ret) / speed_conv;
ret= burn_drive_get_min_write_speed(drive);
min_speed = ((double ) ret) / speed_conv;
if (!media_found)
printf("Drive speed : max=%.1f , min=%.1f\n",
max_speed, min_speed);
else
printf("Avail. speed : max=%.1f , min=%.1f\n",
max_speed, min_speed);
ret = 0;
if (media_found)
ret= burn_disc_read_atip(drive);
if(ret>0) {
ret= burn_drive_get_min_write_speed(drive);
min_speed = ((double ) ret) / speed_conv;
ret= burn_drive_get_write_speed(drive);
max_speed = ((double ) ret) / speed_conv;
printf("Media speed : max=%.1f , min=%.1f\n",
max_speed, min_speed);
}
printf("Speed unit 1x: %s\n", speed_unit);
return 1;
}
int telltoc_speedlist(struct burn_drive *drive)
{
int ret, has_modern_entries = 0;
struct burn_speed_descriptor *speed_list, *sd;
ret = burn_drive_get_speedlist(drive, &speed_list);
if (ret <= 0) {
fprintf(stderr, "SORRY: Cannot obtain speed list info\n");
return 2;
}
for (sd = speed_list; sd != NULL; sd = sd->next)
if (sd->source == 2)
has_modern_entries = 1;
for (sd = speed_list; sd != NULL; sd = sd->next) {
if (has_modern_entries && sd->source < 2)
continue;
if (sd->write_speed <= 0)
continue;
printf("Speed descr. : %d kB/s", sd->write_speed);
if (sd->end_lba >= 0)
printf(", %.1f MiB", ((double) sd->end_lba) / 512.0);
if (sd->profile_name[0])
printf(", %s", sd->profile_name);
printf("\n");
}
burn_drive_free_speedlist(&speed_list);
return 1;
}
int telltoc_formatlist(struct burn_drive *drive)
{
int ret, i, status, num_formats, profile_no, type;
off_t size;
unsigned dummy;
char status_text[80], profile_name[90];
ret = burn_disc_get_formats(drive, &status, &size, &dummy,
&num_formats);
if (ret <= 0) {
fprintf(stderr, "SORRY: Cannot obtain format list info\n");
return 2;
}
if (status == BURN_FORMAT_IS_UNFORMATTED)
sprintf(status_text, "unformatted, up to %.1f MiB",
((double) size) / 1024.0 / 1024.0);
else if(status == BURN_FORMAT_IS_FORMATTED)
sprintf(status_text, "formatted, with %.1f MiB",
((double) size) / 1024.0 / 1024.0);
else if(status == BURN_FORMAT_IS_UNKNOWN) {
burn_disc_get_profile(drive, &profile_no, profile_name);
if (profile_no > 0)
sprintf(status_text, "intermediate or unknown");
else
sprintf(status_text, "no media or unknown media");
} else
sprintf(status_text, "illegal status according to MMC-5");
printf("Format status: %s\n", status_text);
for (i = 0; i < num_formats; i++) {
ret = burn_disc_get_format_descr(drive, i,
&type, &size, &dummy);
if (ret <= 0)
continue;
printf("Format descr.: %2.2Xh , %.1f MiB (%.fs)\n",
type, ((double) size) / 1024.0 / 1024.0,
((double) size) / 2048.0);
}
return 1;
}
int telltoc_toc(struct burn_drive *drive)
{
int num_sessions = 0 , num_tracks = 0 , lba = 0;
int track_count = 0;
int session_no, track_no;
struct burn_disc *disc= NULL;
struct burn_session **sessions;
struct burn_track **tracks;
struct burn_toc_entry toc_entry;
disc = burn_drive_get_disc(drive);
if (disc==NULL) {
fprintf(stderr, "SORRY: Cannot obtain Table Of Content\n");
return 2;
}
sessions = burn_disc_get_sessions(disc, &num_sessions);
for (session_no = 0; session_no<num_sessions; session_no++) {
tracks = burn_session_get_tracks(sessions[session_no],
&num_tracks);
if (tracks==NULL)
continue;
for(track_no= 0; track_no<num_tracks; track_no++) {
track_count++;
burn_track_get_entry(tracks[track_no], &toc_entry);
lba= burn_msf_to_lba(toc_entry.pmin, toc_entry.psec,
toc_entry.pframe);
printf("Media content: session %2d ", session_no+1);
printf("track %2d %s lba: %9d %2.2u:%2.2u:%2.2u\n",
track_count,
((toc_entry.control&7)<4?"audio":"data "),
lba,
toc_entry.pmin,
toc_entry.psec,
toc_entry.pframe);
}
burn_session_get_leadout_entry(sessions[session_no],
&toc_entry);
lba = burn_msf_to_lba(toc_entry.pmin,
toc_entry.psec, toc_entry.pframe);
printf("Media content: session %2d ", session_no+1);
printf("leadout lba: %9d %2.2u:%2.2u:%2.2u\n",
lba,
toc_entry.pmin,
toc_entry.psec,
toc_entry.pframe);
}
if (disc!=NULL)
burn_disc_free(disc);
return 1;
}
int telltoc_msinfo(struct burn_drive *drive,
int msinfo_explicit, int msinfo_alone)
{
int num_sessions, session_no, ret, num_tracks;
int nwa = -123456789, lba = -123456789, aux_lba, lout_lba;
enum burn_disc_status s;
struct burn_disc *disc= NULL;
struct burn_session **sessions;
struct burn_track **tracks;
struct burn_toc_entry toc_entry;
struct burn_write_opts *o= NULL;
s = burn_disc_get_status(drive);
if (s!=BURN_DISC_APPENDABLE) {
if (!msinfo_explicit)
return 2;
fprintf(stderr,
"SORRY: --msinfo can only operate on appendable media.\n");
return 0;
}
/* man mkisofs , option -C :
The first number is the sector number of the first sector in
the last session of the disk that should be appended to.
*/
disc = burn_drive_get_disc(drive);
if (disc==NULL) {
fprintf(stderr,"SORRY: Cannot obtain info about CD content\n");
return 2;
}
sessions = burn_disc_get_sessions(disc, &num_sessions);
for (session_no = 0; session_no<num_sessions; session_no++) {
tracks = burn_session_get_tracks(sessions[session_no],
&num_tracks);
if (tracks==NULL || num_tracks<=0)
continue;
burn_track_get_entry(tracks[0], &toc_entry);
lba= burn_msf_to_lba(toc_entry.pmin, toc_entry.psec,
toc_entry.pframe);
}
if(lba==-123456789) {
fprintf(stderr,"SORRY: Cannot find any track on media\n");
{ ret = 0; goto ex; }
}
/* Prepare a qualified guess as fallback for nwa inquiry */
burn_session_get_leadout_entry(sessions[num_sessions-1], &toc_entry);
lout_lba= burn_msf_to_lba(toc_entry.pmin,toc_entry.psec,
toc_entry.pframe);
/* man mkisofs , option -C :
The second number is the starting sector number of the new session.
*/
/* Set some write opts to be sent to drive. LG GSA-4082B needs it. */
o= burn_write_opts_new(drive);
if(o!=NULL) {
burn_write_opts_set_perform_opc(o, 0);
burn_write_opts_set_write_type(o,
BURN_WRITE_TAO, BURN_BLOCK_MODE1);
}
/* Now try to inquire nwa from drive */
ret= burn_disc_track_lba_nwa(drive,o,0,&aux_lba,&nwa);
telltoc_regrab(drive); /* necessary to calm down my NEC drive */
if(ret<=0) {
fprintf(stderr,
"NOTE: Guessing next writeable address from leadout\n");
if(num_sessions>0)
nwa= lout_lba+6900;
else
nwa= lout_lba+11400;
}
if (!msinfo_alone)
printf("Media msinfo : mkisofs ... -C ");
printf("%d,%d\n",lba,nwa);
ret = 1;
ex:;
if (disc!=NULL)
burn_disc_free(disc);
if (o!=NULL)
burn_write_opts_free(o);
return ret;
}
/** The setup parameters of telltoc */
static char drive_adr[BURN_DRIVE_ADR_LEN] = {""};
static int driveno = 0;
static int do_media = 0;
static int do_toc = 0;
static int do_msinfo = 0;
static int print_help = 0;
static int do_capacities = 0;
/** Converts command line arguments into above setup parameters.
drive_adr[] must provide at least BURN_DRIVE_ADR_LEN bytes.
source_adr[] must provide at least 4096 bytes.
*/
int telltoc_setup(int argc, char **argv)
{
int i;
for (i = 1; i < argc; ++i) {
if (!strcmp(argv[i], "--drive")) {
++i;
if (i >= argc) {
fprintf(stderr,"--drive requires an argument\n");
return 1;
} else if (strcmp(argv[i], "-") == 0) {
drive_adr[0] = 0;
driveno = -1;
} else if (isdigit(argv[i][0])) {
drive_adr[0] = 0;
driveno = atoi(argv[i]);
} else {
if(strlen(argv[i]) >= BURN_DRIVE_ADR_LEN) {
fprintf(stderr,"--drive address too long (max. %d)\n",
BURN_DRIVE_ADR_LEN-1);
return 2;
}
strcpy(drive_adr, argv[i]);
}
} else if (strcmp(argv[i],"--media")==0) {
do_media = 1;
} else if (!strcmp(argv[i], "--msinfo")) {
do_msinfo = 1;
} else if (!strcmp(argv[i], "--capacities")) {
do_capacities = 1;
} else if (!strcmp(argv[i], "--toc")) {
do_toc = 1;
} else if (!strcmp(argv[i], "--help")) {
print_help = 1;
} else {
fprintf(stderr, "Unidentified option: %s\n", argv[i]);
return 7;
}
}
if (argc==1)
print_help = 1;
if (print_help) {
printf("Usage: %s\n", argv[0]);
printf(" [--drive <address>|<driveno>|\"-\"]\n");
printf(" [--media] [--capacities] [--toc] [--msinfo]\n");
printf("Examples\n");
printf("A bus scan (needs rw-permissions to see a drive):\n");
printf(" %s --drive -\n",argv[0]);
printf("Obtain info about the type of loaded media:\n");
printf(" %s --drive /dev/hdc --media\n",argv[0]);
printf("Obtain table of content:\n");
printf(" %s --drive /dev/hdc --toc\n",argv[0]);
printf("Obtain parameters for option -C of program mkisofs:\n");
printf(" msinfo=$(%s --drive /dev/hdc --msinfo 2>/dev/null)\n",
argv[0]);
printf(" mkisofs ... -C \"$msinfo\" ...\n");
printf("Obtain what is available about drive 0 and its media\n");
printf(" %s --drive 0\n",argv[0]);
}
return 0;
}
int main(int argc, char **argv)
{
int ret, toc_failed = 0, msinfo_alone = 0, msinfo_explicit = 0;
int full_default = 0;
ret = telltoc_setup(argc, argv);
if (ret)
exit(ret);
/* Behavior shall be different if --msinfo is only option */
if (do_msinfo) {
msinfo_explicit = 1;
if (!(do_media || do_toc))
msinfo_alone = 1;
}
/* Default option is to do everything if possible */
if (do_media==0 && do_msinfo==0 && do_capacities==0 && do_toc==0
&& driveno!=-1) {
if(print_help)
exit(0);
full_default = do_media = do_msinfo = do_capacities= do_toc = 1;
}
fprintf(stderr, "Initializing libburnia.pykix.org ...\n");
if (burn_initialize())
fprintf(stderr, "Done\n");
else {
fprintf(stderr,"\nFATAL: Failed to initialize.\n");
exit(33);
}
/* Print messages of severity SORRY or more directly to stderr */
burn_msgs_set_severities("NEVER", "SORRY", "telltoc : ");
/** Note: driveno might change its value in this call */
ret = telltoc_aquire_drive(drive_adr, &driveno, !full_default);
if (ret<=0) {
fprintf(stderr,"\nFATAL: Failed to aquire drive.\n");
{ ret = 34; goto finish_libburn; }
}
if (ret == 2)
{ ret = 0; goto release_drive; }
if (do_media) {
ret = telltoc_media(drive_list[driveno].drive);
if (ret<=0)
{ret = 36; goto release_drive; }
}
if (do_capacities) {
ret = telltoc_speedlist(drive_list[driveno].drive);
if (ret<=0)
{ret = 39; goto release_drive; }
ret = telltoc_formatlist(drive_list[driveno].drive);
if (ret<=0)
{ret = 39; goto release_drive; }
}
if (do_toc) {
ret = telltoc_toc(drive_list[driveno].drive);
if (ret<=0)
{ret = 37; goto release_drive; }
if (ret==2)
toc_failed = 1;
}
if (do_msinfo) {
ret = telltoc_msinfo(drive_list[driveno].drive,
msinfo_explicit, msinfo_alone);
if (ret<=0)
{ret = 38; goto release_drive; }
}
ret = 0;
if (toc_failed)
ret = 37;
release_drive:;
if (drive_is_grabbed)
burn_drive_release(drive_list[driveno].drive, 0);
finish_libburn:;
/* This app does not bother to know about exact scan state.
Better to accept a memory leak here. We are done anyway. */
/* burn_drive_info_free(drive_list); */
burn_finish();
exit(ret);
}
/* License and copyright aspects:
See libburner.c
*/

103
test/toc.c Normal file
View File

@ -0,0 +1,103 @@
/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */
#include <libburn/libburn.h>
#include <unistd.h>
#include <stdlib.h>
#include <stdio.h>
#include <assert.h>
static struct burn_drive_info *drives;
static unsigned int n_drives;
static void show_tocs()
{
struct burn_session **sessions;
struct burn_track **tracks;
struct burn_disc *disc;
int nses, ntracks, hidefirst;
unsigned int i, j, k;
struct burn_toc_entry e;
enum burn_disc_status s;
for (i = 0; i < n_drives; ++i) {
fprintf(stderr, "TOC for disc in %s - %s:\n",
drives[i].vendor, drives[i].product);
if (!burn_drive_grab(drives[i].drive, 1)) {
fprintf(stderr, "Unable to open the drive!\n");
continue;
}
while (burn_drive_get_status(drives[i].drive, NULL))
usleep(1000);
while ((s = burn_disc_get_status(drives[i].drive))
== BURN_DISC_UNREADY)
usleep(1000);
if (s != BURN_DISC_FULL) {
burn_drive_release(drives[i].drive, 0);
fprintf(stderr, "No disc found!\n");
continue;
}
disc = burn_drive_get_disc(drives[i].drive);
sessions = burn_disc_get_sessions(disc, &nses);
for (k = 0; k < nses; ++k) {
tracks = burn_session_get_tracks(sessions[k],
&ntracks);
hidefirst = burn_session_get_hidefirst(sessions[k]);
if (hidefirst)
fprintf(stderr,
"track: GAP (%2d) lba: %9d (%9d) %02d:%02d:%02d adr: X control: X mode: %d\n",
k + 1, 0, 0, 0, 2, 0,
burn_track_get_mode(tracks[0]));
for (j = !!hidefirst; j < ntracks; ++j) {
burn_track_get_entry(tracks[j], &e);
fprintf(stderr,
"track: %3d (%2d) lba: %9d (%9d) %02d:%02d:%02d "
"adr: %d control: %d mode: %d\n",
e.point, e.session,
burn_msf_to_lba(e.pmin, e.psec,
e.pframe),
burn_msf_to_lba(e.pmin, e.psec,
e.pframe) * 4,
e.pmin, e.psec, e.pframe, e.adr,
e.control,
burn_track_get_mode(tracks[j]));
}
burn_session_get_leadout_entry(sessions[k], &e);
fprintf(stderr,
"track:lout (%2d) lba: %9d (%9d) %02d:%02d:%02d "
"adr: %d control: %d mode: %d\n",
k + 1, burn_msf_to_lba(e.pmin, e.psec,
e.pframe),
burn_msf_to_lba(e.pmin, e.psec,
e.pframe) * 4, e.pmin,
e.psec, e.pframe, e.adr, e.control, -1);
}
burn_disc_free(disc);
burn_drive_release(drives[i].drive, 0);
}
}
int main()
{
fprintf(stderr, "Initializing library...");
if (burn_initialize())
fprintf(stderr, "Success\n");
else {
printf("Failed\n");
return 1;
}
fprintf(stderr, "Scanning for devices...");
while (!burn_drive_scan(&drives, &n_drives)) ;
fprintf(stderr, "Done\n");
show_tocs();
burn_drive_info_free(drives);
burn_finish();
return 0;
}

77
test/tree.py Normal file
View File

@ -0,0 +1,77 @@
# a module to help with handling of filenames, directory trees, etc.
import os
import os.path
import stat
def pathsubtract(a, b):
index = a.find(b)
if index == -1:
return None
res = a[ (index + len(b)): ]
if res.find("/") == 0:
res = res[1:]
return res
# same as C strcmp()
def strcmp(a, b):
if a < b:
return -1
if a > b:
return 1
return 0
class TreeNode:
# path is the location of the file/directory. It is either a full path or
# a path relative to $PWD
def __init__(self, parent, path=".", root=".", isofile=None):
if isofile:
self.root = os.path.abspath(isofile)
self.path = ""
else:
fullpath = os.path.abspath( path )
fullroot = os.path.abspath( root )
self.root = fullroot
self.path = pathsubtract( fullpath, fullroot )
self.parent = parent
self.children = []
if self.path == None:
raise NameError, "Invalid paths %s and %s" % (fullpath, fullroot)
# if this is a directory, add its children recursively
def addchildren(self):
if not stat.S_ISDIR( os.lstat(self.root + "/" + self.path).st_mode ):
return
children = os.listdir( self.root + "/" + self.path )
for child in children:
if self.path:
child = self.path + "/" + child
self.children.append( TreeNode(self, child, self.root) )
for child in self.children:
child.addchildren()
def printAll(self, spaces=0):
print " "*spaces + self.root + "/" + self.path
for child in self.children:
child.printAll(spaces + 2)
def isValidISO1(self):
pass
class Tree:
def __init__(self, root=None, isofile=None):
if isofile:
self.root = TreeNode(parent=None, isofile=isofile)
else:
self.root = TreeNode(parent=None, path=root, root=root)
self.root.addchildren()
def isValidISO1(self):
return root.isValidISO1();
#t = Tree(root=".")
#t.root.printAll()

BIN
test/tree.pyc Normal file

Binary file not shown.