Rock Ridge CE entry points to a range that crosses a block boundary,
or of which the byte offset is larger than the block size of 2048.
Thanks to Joerg Meyer.
by miscalculating the number of root directory's CE blocks.
This lead to fatal offset of all further data by one block.
All sub directories and all data file content is affected.
Quite obvious to see. The problem existed since March 2009.
AAIP is used for recording of ACL and xattr. The problem gets enabled by call
iso_write_opts_set_aaip(opts, 1).
New IsoFileSourceIface version 2 with method clone_src().
New IsoStreamIface version 4 with method clone_stream().
New public function prototype iso_node_xinfo_cloner.
New API calls iso_node_xinfo_make_clonable(), iso_node_xinfo_get_cloner().
New public iso_node_xinfo_cloner instance aaip_xinfo_cloner().
New API calls iso_node_get_next_xinfo(), iso_node_remove_all_xinfo().
New API call iso_node_remove_tree().
an MBR with a first partiton table entry that bears non-zero start address.
A second set of volume descriptors and directory tree+tables gets created
which can be used to mount the image at the partition start.
Not yet implemented for second set: ISO 9660:1999, MD5 checksums.
a ZF entry. Storing ZF entries during image write if that filter is found
directly on a fsrc stream from the loaded image and content does not get
copied.
because ancient Apple ISO 9660 already used "AA".
Old AAIP-1.0 enhanced images can still be read and luckily their AAIP fields
can be distinguished from eventual Apple AA fields.
Bug fix: iso_write_opts_set_rrip_version_1_10() caused wrong size of
root record CE,
Stability enhancement: util.c:str2ascii() got a fallback for the case
that charset "WCHAR_T" is not available,
Debugging wrapper around iconv*() calls,
Experiments about inode numbers in fs_image,
Experiment to insert obsolete RR entries.
(Sorry for the obfuscation. Most is due to a hard ride on Solaris.
See macros at the end of libisofs/libisofs.h)