New options -charset, -in_charset, -out_charset

libisoburn/trunk/xorriso/convert_man_to_html.sh

@@ -65,6 +65,7 @@ then
   -e 's/<b>Settings for result writing:<\/b>/\&nbsp;<BR><b>Settings for result writing:<\/b><BR>\&nbsp;<BR>/' \
   -e 's/^706k = 706kB/\&nbsp;\&nbsp;706k = 706kB/' \
   -e 's/^5540k = 5540kB/\&nbsp;\&nbsp;5540k = 5540kB/' \
+  -e 's/<b>Character sets:<\/b>/\&nbsp;<BR><b>Character sets:<\/b><BR>\&nbsp;<BR>/' \
   -e 's/<b>Exception processing:<\/b>/\&nbsp;<BR><b>Exception processing:<\/b><BR>\&nbsp;<BR>/' \
   -e 's/<b>El Torito bootable ISO images:<\/b>/\&nbsp;<BR><b>El Torito bootable ISO images:<\/b><BR>\&nbsp;<BR>/' \
   -e 's/<b>Dialog mode control:<\/b>/\&nbsp;<BR><b>Dialog mode control:<\/b><BR>\&nbsp;<BR>/' \

libisoburn/trunk/xorriso/xorriso.1

@@ -555,6 +555,11 @@ until the next -dev or -indev. After the image has been loaded once, the
 setting is valid for -rollback until next -dev or -indev, where it
 will be reset to "auto".
 .TP
+\fB\-in_charset\fR character_set_name
+Set the character set from which to convert file names when loading an
+image. This has eventually to be done before specifying -dev , -indev or
+-rollback. See paragraph "Character sets" for more explanations.
+.TP
 \fB\-rom_toc_scan\fR "on"|"off"[:"emul_on"|"emul_off"]
 Read-only drives do not tell the actual media type but show any media as
 ROM (e.g. as DVD-ROM). The session history of MMC multi-session media might
@@ -1328,7 +1333,7 @@ If enabled by "on", generate Joliet info additional to Rock Ridge info.
 Specify the volume ID. xorriso accepts any text up to 32 characters,
 but according to rarely obeyed specs stricter rules apply:
 .br
-ECMA 119 demands character set [A-Z0-9_]. Like: "IMAGE_23"
+ECMA 119 demands ASCII characters out of [A-Z0-9_]. Like: "IMAGE_23"
 .br
 Joliet allows 16 UCS-2 characters. Like: "Windows name"
 .br
@@ -1348,6 +1353,10 @@ If you insist in -volid "ISOIMAGE", set it again after those commands.
 Set the publisher string to be written with the next -commit. Permissible
 are up to 128 characters.
 .TP
+\fB\-out_charset\fR character_set_name
+Set the character set to which file names get converted when writing an
+image. See paragraph "Character sets" for more explanations.
+.TP
 \fB\-uid\fR uid
 User id to be used for all files when the new ISO tree gets written to media.
 .TP
@@ -1454,7 +1463,7 @@ boot image which is present on the input media.
 Most safe is the default: -boot_image "any" "discard".
 .br
 
-A bootspec is a word of the form name=value and is used describe the
+A bootspec is a word of the form name=value and is used to describe the
 activation of a ISOLINUX boot image by an El Torito record and eventually
 a MBR. The names "dir" and "bin_path" lead to boot image activation.
 .br
@@ -1491,6 +1500,34 @@ write session if not the isohybrid signature is found in the bin_path file.
 Default is "isohybrid=auto" which silently omits the MBR if the signature is
 missing. 
 .TP
+.B Character sets:
+.PP
+File names are strings of non-zero bytes with 8 bit each. Unfortunately
+the same byte string may appear as different peculiar national characters
+on differently nationalized computers.
+The meanings of byte codes are defined in \fBcharacter sets\fR which have
+names. Shell command iconv -l lists them. 
+.br
+Character sets should not matter as long as only alphanumeric characters are
+used for file names or as long as the computer which runs xorriso uses the
+same character set as the computer which wrote the loaded image and those
+computers which shall read the newly emerging ISO image.
+Outside these constraints it may be necessary to let xorriso convert byte
+codes.
+.br
+There is an input conversion from input character set to the local character
+set which applies when an ISO image gets loaded. A conversion from local
+character set to the output character set is performed when a new
+image tree gets written. The sets can be defined independently by options
+-in_charset and -out_charset. Normally one will have both identical, if ever.
+.br
+xorriso expects and uses the local character set for command arguments
+and file name output. Other character sets will only be in effect on media.
+.TP
+\fB\-charset\fR character_set_name
+Set the character set from which to convert file names when loading an
+image and to which to convert when writing an image.
+.TP
 .B Exception processing:
 .PP
 Since the tasks of xorriso are manifold and prone to external influence, there

libisoburn/trunk/xorriso/xorriso.c

@@ -90,6 +90,10 @@ or
 #include <pwd.h>
 #include <grp.h>
 
+/* for -charset */
+#include <iconv.h>
+#include <langinfo.h>
+
 /* eventually, this is done in xorriso_private.h : #include <regex.h> */
 
 #ifdef Xorriso_with_readlinE
@@ -3659,12 +3663,14 @@ int Xorriso_new(struct XorrisO ** xorriso,char *progname, int flag)
  m->indev[0]= 0;
  m->in_drive_handle= NULL;
  m->in_volset_handle= NULL;
+ m->in_charset= NULL;
  m->volset_change_pending= 0;
  m->no_volset_present= 0;
  m->in_sector_map= NULL;
  m->check_media_default= NULL;
  m->outdev[0]= 0;
  m->out_drive_handle= NULL;
+ m->out_charset= NULL;
  m->dev_fd_1= -1;
  m->grow_blindly_msc2= -1;
  m->ban_stdio_write= 0;
@@ -3827,6 +3833,10 @@ int Xorriso_destroy(struct XorrisO **xorriso, int flag)
  m= *xorriso;
  if(m==NULL)
    return(0);
+ if(m->in_charset!=NULL)
+   free(m->in_charset);
+ if(m->out_charset!=NULL)
+   free(m->out_charset);
  Xorriso_destroy_re(m,0);
  Exclusions_destroy(&(m->disk_exclusions), 0);
  Xorriso_detach_libraries(m, flag&1);
@@ -5342,6 +5352,7 @@ int Xorriso_status(struct XorrisO *xorriso, char *filter, FILE *fp, int flag)
 {
  int is_default, no_defaults, i, ret, adr_mode, bin_path_in_use= 0;
  char *line, sfe[5 * SfileadrL + 80], mode[80], *form, *treatment;
+ char *in_pt, *out_pt;
  static char channel_prefixes[4][4]= {".","R","I","M"};
  static char load_names[][20]= {"auto", "session", "track", "lba", "volid"};
  static int max_load_mode= 4;
@@ -5826,7 +5837,28 @@ int Xorriso_status(struct XorrisO *xorriso, char *filter, FILE *fp, int flag)
  if(!(is_default && no_defaults))
    Xorriso_status_result(xorriso,filter,fp,flag&2);
 
- if(strcmp(xorriso->indev,xorriso->outdev)==0) { 
+ is_default= (xorriso->out_charset == NULL && xorriso->in_charset == NULL);
+ in_pt= "";
+ if(xorriso->in_charset != NULL)
+   in_pt= xorriso->in_charset;
+ out_pt= "";
+ if(xorriso->out_charset != NULL)
+   out_pt= xorriso->out_charset;
+ if(strcmp(in_pt, out_pt) == 0 &&
+    strncmp(filter, "-in_charset", 11) != 0 &&
+    strncmp(filter, "-out_charset", 12) != 0) {
+   sprintf(line, "-charset %s\n", Text_shellsafe(in_pt, sfe, 0));
+   Xorriso_status_result(xorriso, filter, fp, flag & 2);
+ } else {
+   sprintf(line, "-in_charset %s\n", Text_shellsafe(in_pt, sfe, 0));
+   Xorriso_status_result(xorriso, filter, fp, flag & 2);
+   sprintf(line, "-out_charset %s\n", Text_shellsafe(out_pt, sfe, 0));
+   Xorriso_status_result(xorriso, filter, fp, flag & 2);
+ }
+ 
+ if(strcmp(xorriso->indev, xorriso->outdev) == 0 &&
+    strncmp(filter, "-indev", 6) != 0 &&
+    strncmp(filter, "-outdev", 7) != 0) {
    sprintf(line,"-dev %s\n", Text_shellsafe(xorriso->indev,sfe,0));
    Xorriso_status_result(xorriso,filter,fp,flag&2);
  } else {
@@ -10771,6 +10803,64 @@ int Xorriso_option_cdx(struct XorrisO *xorriso, char *disk_path, int flag)
 }
 
 
+/* Option -charset */
+/* @param flag bit0= set in_charset
+               bit1= set out_charset
+*/
+int Xorriso_option_charset(struct XorrisO *xorriso, char *name, int flag)
+{
+ char *name_pt= NULL, sfe[5 * SfileadrL];
+ iconv_t iconv_ret= (iconv_t) -1;
+
+ if(name != NULL)
+   if(name[0] != 0)
+     name_pt= name;
+ if(flag & 1) {
+   if(name_pt != NULL) {
+     iconv_ret= iconv_open(nl_langinfo(CODESET), name_pt);
+     if(iconv_ret == (iconv_t) -1) {
+       sprintf(xorriso->info_text,
+               "-%scharset: Cannot convert from charset %s",
+               flag & 2 ? "" : "in_", Text_shellsafe(name_pt, sfe, 0));
+       Xorriso_msgs_submit(xorriso, 0, xorriso->info_text, errno, "FAILURE",
+                           0);
+       return(0);
+     } else
+       iconv_close(iconv_ret);
+   }
+   if(Sregex_string(&(xorriso->in_charset), name_pt, 0) <= 0) {
+     Xorriso_no_malloc_memory(xorriso, NULL, 0);
+     return(-1);
+   }
+ }
+ if(flag & 2) {
+   if(name_pt != NULL) {
+     iconv_ret= iconv_open(nl_langinfo(CODESET), name_pt);
+     if(iconv_ret == (iconv_t) -1) {
+       sprintf(xorriso->info_text, "-%scharset: Cannot convert to charset %s",
+               flag & 1 ? "" : "out_", Text_shellsafe(name_pt, sfe, 0));
+       Xorriso_msgs_submit(xorriso, 0, xorriso->info_text, errno, "FAILURE",
+                           0);
+       return(0);
+     } else
+       iconv_close(iconv_ret);
+   }
+   if(Sregex_string(&(xorriso->out_charset), name_pt, 0) <= 0) {
+     Xorriso_no_malloc_memory(xorriso, NULL, 0);
+     return(-1);
+   }
+ }
+
+ if(name_pt == NULL)
+   name_pt= nl_langinfo(CODESET);
+ sprintf(xorriso->info_text, "Character set for %sconversion is now: %s",
+         (flag & 3) == 1 ? "input " : (flag & 3) == 2 ? "output " : "",
+         Text_shellsafe(name_pt, sfe, 0));
+ Xorriso_msgs_submit(xorriso, 0, xorriso->info_text, errno, "NOTE", 0);
+ return(1);
+}
+
+
 /* Option -check_media */
 int Xorriso_option_check_media(struct XorrisO *xorriso,
                                int argc, char **argv, int *idx, int flag)
@@ -14670,11 +14760,12 @@ int Xorriso_count_args(struct XorrisO *xorriso, int argc, char **argv,
  };
  static char arg1_commands[][40]= {
     "abort_on","add_plainly","backslash_codes","blank",
-    "cd","cdi","cdx","close","commit_eject",
+    "cd","cdi","cdx","charset","close","commit_eject",
     "dev", "dummy","dialog","disk_pattern","eject","iso_rr_pattern","follow",
     "format","fs","gid","grow_blindly",
-    "history","indev","joliet","list_delimiter","mark","not_leaf",
+    "history","indev","in_charset","joliet","list_delimiter",
-    "not_list","not_mgt","options_from_file","osirrox","outdev","overwrite",
+    "mark","not_leaf","not_list","not_mgt",
+    "options_from_file","osirrox","outdev","out_charset","overwrite",
     "pacifier","padding","path_list","pathspecs","pkt_output","print","prompt",
     "prog","prog_help","publisher","quoted_not_list","quoted_path_list",
     "reassure","report_about","rom_toc_scan",
@@ -14866,6 +14957,10 @@ next_command:;
    (*idx)++;
    ret= Xorriso_option_cdx(xorriso, arg1, 0);
 
+ } else if(strcmp(cmd,"charset")==0) {
+   (*idx)++;
+   ret= Xorriso_option_charset(xorriso, arg1, 3);
+
  } else if(strcmp(cmd,"check_media")==0) {
    ret= Xorriso_option_check_media(xorriso, argc, argv, idx, 0);
 
@@ -15049,6 +15144,10 @@ next_command:;
    (*idx)++;
    ret= Xorriso_option_dev(xorriso, arg1, 1);
 
+ } else if(strcmp(cmd,"in_charset")==0) {
+   (*idx)++;
+   ret= Xorriso_option_charset(xorriso, arg1, 1);
+
  } else if(strcmp(cmd,"iso_rr_pattern")==0) {
    (*idx)++;
    ret= Xorriso_option_iso_rr_pattern(xorriso, arg1, 0);
@@ -15131,13 +15230,17 @@ next_command:;
    if(ret==3)
      goto ex;
 
+ } else if(strcmp(cmd,"osirrox")==0) {
+   (*idx)++;
+   ret= Xorriso_option_osirrox(xorriso,arg1,0);
+
  } else if(strcmp(cmd,"outdev")==0) {
    (*idx)++;
    ret= Xorriso_option_dev(xorriso, arg1, 2);
 
- } else if(strcmp(cmd,"osirrox")==0) {
+ } else if(strcmp(cmd,"out_charset")==0) {
    (*idx)++;
-   ret= Xorriso_option_osirrox(xorriso,arg1,0);
+   ret= Xorriso_option_charset(xorriso, arg1, 2);
 
  } else if(strcmp(cmd,"overwrite")==0) {
    (*idx)++;

libisoburn/trunk/xorriso/xorriso_private.h

@@ -158,6 +158,7 @@ struct XorrisO { /* the global context of xorriso */
  char indev[SfileadrL];
  void *in_drive_handle;  /* interpreted only by xorrisoburn.c */
  void *in_volset_handle; /* interpreted only by xorrisoburn.c */
+ char *in_charset;       /* The charset to interpret the filename bytes */
 
  int volset_change_pending; /* whether -commit would make sense */
  int no_volset_present;     /* set to 1 on first failure */
@@ -168,6 +169,7 @@ struct XorrisO { /* the global context of xorriso */
 
  char outdev[SfileadrL];
  void *out_drive_handle; /* interpreted only by xorrisoburn.c */
+ char *out_charset;      /* The charset to produce the filename bytes for */
  int dev_fd_1; /* The fd which substitutes for /dev/fd/1 and is
                      connected to externaly perveived stdout.
                   */

libisoburn/trunk/xorriso/xorriso_timestamp.h

@@ -1 +1 @@
-#define Xorriso_timestamP "2008.10.30.114313"
+#define Xorriso_timestamP "2008.11.03.115721"

libisoburn/trunk/xorriso/xorrisoburn.c

@@ -402,7 +402,7 @@ int Xorriso_create_empty_iso(struct XorrisO *xorriso, int flag)
    return(ret);
  /* Note: no return before isoburn_ropt_destroy() */
  isoburn_ropt_set_extensions(ropts, isoburn_ropt_pretend_blank);
- isoburn_ropt_set_input_charset(ropts, NULL);
+ isoburn_ropt_set_input_charset(ropts, xorriso->in_charset);
  isoburn_set_read_pacifier(drive, NULL, NULL);
  ret= isoburn_read_image(drive, ropts, &volset);
  Xorriso_process_msg_queues(xorriso,0);
@@ -613,7 +613,7 @@ int Xorriso_aquire_drive(struct XorrisO *xorriso, char *adr, int flag)
    goto ex;
  isoburn_ropt_set_extensions(ropts, isoburn_ropt_noiso1999);
  isoburn_ropt_set_default_perms(ropts, (uid_t) 0, (gid_t) 0, (mode_t) 0555);
- isoburn_ropt_set_input_charset(ropts, NULL);
+ isoburn_ropt_set_input_charset(ropts, xorriso->in_charset);
  
  Xorriso_set_image_severities(xorriso, 1); /* No DEBUG messages */
  Xorriso_pacifier_reset(xorriso, 0);
@@ -1130,6 +1130,15 @@ int Xorriso_write_session(struct XorrisO *xorriso, int flag)
        Xorriso_msgs_submit(xorriso, 0, xorriso->info_text, 0, "FAILURE", 0);
        {ret= 0; goto ex;}
      }
+
+#ifdef Xorriso_with_make_isohybrid_mbR
+
+     /* ??? must the boot image have a low LBA to boot from MBR ? */
+     isoburn_igopt_set_sort_files(sopts, isoburn_igopt_sort_files_by_weight);
+     iso_node_set_sort_weight(node, 0x7fffffff);
+
+#endif
+
      ret= Xorriso_node_from_path(xorriso, image, xorriso->boot_image_cat_path,
                                  &node, 1);
      if(ret > 0) {
@@ -1230,7 +1239,7 @@ int Xorriso_write_session(struct XorrisO *xorriso, int flag)
  isoburn_igopt_set_sort_files(sopts, 1);
  isoburn_igopt_set_over_mode(sopts, 0, 0, (mode_t) 0, (mode_t) 0);
  isoburn_igopt_set_over_ugid(sopts, 0, 0, (uid_t) 0, (gid_t) 0);
- isoburn_igopt_set_out_charset(sopts, NULL);
+ isoburn_igopt_set_out_charset(sopts, xorriso->out_charset);
  isoburn_igopt_set_fifo_size(sopts, xorriso->fs * 2048);
 
  if(image!=NULL &&