id3v2

4 files changed, 305 insertions, 75 deletions

diff --git a/src/include/extractor.h b/src/include/extractor.h
index e826b8d..9c4ae60 100644
--- a/src/include/extractor.h
+++ b/src/include/extractor.h
@@ -136,7 +136,7 @@ enum EXTRACTOR_MetaType
     EXTRACTOR_METATYPE_PUBLICATION_DATE = 24,
     EXTRACTOR_METATYPE_BIBTEX_EPRINT = 25,
     EXTRACTOR_METATYPE_BIBTEX_ENTRY_TYPE = 26,
-    EXTRACTOR_METATYPE_DOCUMENT_LANGUAGE = 27,
+    EXTRACTOR_METATYPE_LANGUAGE = 27,
     EXTRACTOR_METATYPE_CREATION_TIME = 28,
     EXTRACTOR_METATYPE_URL = 29,
 
@@ -285,12 +285,21 @@ enum EXTRACTOR_MetaType
     EXTRACTOR_METATYPE_CHAPTER_NAME = 153,
     EXTRACTOR_METATYPE_SONG_COUNT = 154,
     EXTRACTOR_METATYPE_STARTING_SONG = 155,
+    EXTRACTOR_METATYPE_PLAY_COUNTER = 156,
+    EXTRACTOR_METATYPE_CONDUCTOR = 157,
+    EXTRACTOR_METATYPE_INTERPRETATION = 158,
+    EXTRACTOR_METATYPE_COMPOSER = 159,
+    EXTRACTOR_METATYPE_BEATS_PER_MINUTE = 160,
+    EXTRACTOR_METATYPE_ENCODED_BY = 161,
+    EXTRACTOR_METATYPE_ORIGINAL_TITLE = 162,
+    EXTRACTOR_METATYPE_ORIGINAL_ARTIST = 163,
+    EXTRACTOR_METATYPE_ORIGINAL_WRITER = 164,
+    EXTRACTOR_METATYPE_ORIGINAL_RELEASE_YEAR = 165,
+    EXTRACTOR_METATYPE_LYRICS = 166,
+    EXTRACTOR_METATYPE_POPULARITY_METER = 167,
 
     /* fixme: used up to here! */
 
-    EXTRACTOR_METATYPE_LYRICS = 67,
-    EXTRACTOR_METATYPE_CONDUCTOR = 64,
-    EXTRACTOR_METATYPE_INTERPRET = 65,
     EXTRACTOR_METATYPE_MUSIC_CD_IDENTIFIER = 117,
     EXTRACTOR_METATYPE_MUSICIAN_CREDITS_LIST = 123,
 
@@ -304,7 +313,6 @@ enum EXTRACTOR_MetaType
 
 
     /* numeric metrics */
-    EXTRACTOR_METATYPE_POPULARITY_METER = 119,
     EXTRACTOR_METATYPE_RATING = 145,
     EXTRACTOR_METATYPE_PRIORITY = 60,
 
@@ -316,7 +324,6 @@ enum EXTRACTOR_MetaType
     /* misc (see if these are still needed...) */
 
     EXTRACTOR_METATYPE_GENERATOR = 103,
-    EXTRACTOR_METATYPE_ENCODED_BY = 121,
 
     EXTRACTOR_METATYPE_FULL_DATA = 137,
 
diff --git a/src/main/extractor_metatypes.c b/src/main/extractor_metatypes.c
index 158d3a2..b98a395 100644
--- a/src/main/extractor_metatypes.c
+++ b/src/main/extractor_metatypes.c
@@ -380,14 +380,32 @@ static const struct MetaTypeDescription meta_type_descriptions[] = {
   /* 155 */
   { gettext_noop ("starting song"),
     gettext_noop ("number of the first song to play") }, 
-  { gettext_noop (""),
-    gettext_noop ("") }, 
-  { gettext_noop (""),
-    gettext_noop ("") }, 
-  { gettext_noop (""),
-    gettext_noop ("") }, 
-  { gettext_noop (""),
+  { gettext_noop ("play counter"),
+    gettext_noop ("number of times the media has been played") }, 
+  { gettext_noop ("conductor"),
+    gettext_noop ("name of the conductor") }, 
+  { gettext_noop ("interpretation"),
+    gettext_noop ("information about the people behind interpretations of an existing piece") }, 
+  { gettext_noop ("composer"),
+    gettext_noop ("name of the composer") }, 
+  /* 160 */
+  { gettext_noop ("beats per minute"),
     gettext_noop ("") }, 
+  { gettext_noop ("encoded by"),
+    gettext_noop ("name of person or organization that encoded the file") }, 
+  { gettext_noop ("original title"),
+    gettext_noop ("title of the original work") }, 
+  { gettext_noop ("original artist"),
+    gettext_noop ("name of the original artist") }, 
+  { gettext_noop ("original writer"),
+    gettext_noop ("name of the original lyricist or writer") }, 
+  /* 165 */
+  { gettext_noop ("original release year"),
+    gettext_noop ("year of the original release") }, 
+  { gettext_noop ("lyrics"),
+    gettext_noop ("lyrics of the song or text description of vocal activities") }, 
+  { gettext_noop ("popularity"),
+    gettext_noop ("information about the file's popularity") }, 
   { gettext_noop (""),
     gettext_noop ("") }, 
 #if 0
diff --git a/src/plugins/html_extractor.c b/src/plugins/html_extractor.c
index 435548a..313c4ac 100644
--- a/src/plugins/html_extractor.c
+++ b/src/plugins/html_extractor.c
@@ -44,7 +44,7 @@ static struct
   { "rights", EXTRACTOR_METATYPE_RIGHTS },
   { "dc.rights", EXTRACTOR_METATYPE_RIGHTS },
   { "copyright", EXTRACTOR_METATYPE_COPYRIGHT },
-  { "language", EXTRACTOR_METATYPE_DOCUMENT_LANGUAGE },  
+  { "language", EXTRACTOR_METATYPE_LANGUAGE },  
   { "keywords", EXTRACTOR_METATYPE_KEYWORDS },
   { "abstract", EXTRACTOR_METATYPE_ABSTRACT },
   { "formatter", EXTRACTOR_METATYPE_CREATED_BY_SOFTWARE },
diff --git a/src/plugins/id3v2_extractor.c b/src/plugins/id3v2_extractor.c
index fa5fea6..08ba124 100644
--- a/src/plugins/id3v2_extractor.c
+++ b/src/plugins/id3v2_extractor.c
@@ -28,46 +28,91 @@
 
 #define DEBUG_EXTRACT_ID3v2 0
 
+enum Id3v2Fmt
+  {
+    T, /* simple, 0-terminated string, prefixed by encoding */
+    U, /* 0-terminated ASCII string, no encoding */
+    UL, /* unsync'ed lyrics */
+    SL, /* sync'ed lyrics */
+    L, /* string with language prefix */
+    I /* image */
+  };
+
 typedef struct
 {
   const char *text;
   enum EXTRACTOR_MetaType type;
+  enum Id3v2Fmt fmt;
 } Matches;
 
 static Matches tmap[] = {
-  {"TAL", EXTRACTOR_METATYPE_TITLE},
-  {"TT1", EXTRACTOR_METATYPE_GROUP},
-  {"TT2", EXTRACTOR_METATYPE_TITLE},
-  {"TT3", EXTRACTOR_METATYPE_TITLE},
-  {"TXT", EXTRACTOR_METATYPE_DESCRIPTION},
-  {"TPB", EXTRACTOR_METATYPE_PUBLISHER},
-  {"WAF", EXTRACTOR_METATYPE_LOCATION},
-  {"WAR", EXTRACTOR_METATYPE_LOCATION},
-  {"WAS", EXTRACTOR_METATYPE_LOCATION},
-  {"WCP", EXTRACTOR_METATYPE_COPYRIGHT},
-  {"WAF", EXTRACTOR_METATYPE_LOCATION},
-  {"WCM", EXTRACTOR_METATYPE_DISCLAIMER},
-  {"TSS", EXTRACTOR_METATYPE_FORMAT},
-  {"TYE", EXTRACTOR_METATYPE_DATE},
-  {"TLA", EXTRACTOR_METATYPE_LANGUAGE},
-  {"TP1", EXTRACTOR_METATYPE_ARTIST},
-  {"TP2", EXTRACTOR_METATYPE_ARTIST},
-  {"TP3", EXTRACTOR_METATYPE_CONDUCTOR},
-  {"TP4", EXTRACTOR_METATYPE_INTERPRET},
-  {"IPL", EXTRACTOR_METATYPE_CONTRIBUTOR},
-  {"TOF", EXTRACTOR_METATYPE_FILENAME},
-  {"TEN", EXTRACTOR_METATYPE_PRODUCER},
-  {"TCO", EXTRACTOR_METATYPE_SUBJECT},
-  {"TCR", EXTRACTOR_METATYPE_COPYRIGHT},
-  {"SLT", EXTRACTOR_METATYPE_LYRICS},
-  {"TOA", EXTRACTOR_METATYPE_ARTIST},
-  {"TRC", EXTRACTOR_METATYPE_ISRC},
-  {"TRK", EXTRACTOR_METATYPE_TRACK_NUMBER},
-  {"TCM", EXTRACTOR_METATYPE_CREATOR},
-  {"TOT", EXTRACTOR_METATYPE_ALBUM},
-  {"TOL", EXTRACTOR_METATYPE_AUTHOR},
-  {"COM", EXTRACTOR_METATYPE_COMMENT},
-  {"", EXTRACTOR_METATYPE_KEYWORDS},
+  /* skipping UFI */
+  {"TT1", EXTRACTOR_METATYPE_SECTION, T},
+  {"TT2", EXTRACTOR_METATYPE_TITLE, T},
+  {"TT3", EXTRACTOR_METATYPE_SONG_VERSION, T},
+  {"TP1", EXTRACTOR_METATYPE_ARTIST, T},
+  {"TP2", EXTRACTOR_METATYPE_PERFORMER, T},
+  {"TP3", EXTRACTOR_METATYPE_CONDUCTOR, T},
+  {"TP4", EXTRACTOR_METATYPE_INTERPRETATION, T},
+  {"TCM", EXTRACTOR_METATYPE_COMPOSER, T},
+  {"TXT", EXTRACTOR_METATYPE_WRITER, T},
+  {"TLA", EXTRACTOR_METATYPE_LANGUAGE, T},
+  {"TCO", EXTRACTOR_METATYPE_GENRE, T},
+  {"TAL", EXTRACTOR_METATYPE_ALBUM, T},
+  {"TPA", EXTRACTOR_METATYPE_DISC_NUMBER, T},
+  {"TRK", EXTRACTOR_METATYPE_TRACK_NUMBER, T},
+  {"TRC", EXTRACTOR_METATYPE_ISRC, T},
+  {"TYE", EXTRACTOR_METATYPE_PUBLICATION_YEAR, T},
+  /*
+    FIXME: these two and TYE should be combined into
+    the actual publication date (if TRD is missing)
+  {"TDA", EXTRACTOR_METATYPE_PUBLICATION_DATE},
+  {"TIM", EXTRACTOR_METATYPE_PUBLICATION_DATE},
+  */
+  {"TRD", EXTRACTOR_METATYPE_CREATION_TIME, T},
+  {"TMT", EXTRACTOR_METATYPE_SOURCE, T},
+  {"TFT", EXTRACTOR_METATYPE_FORMAT_VERSION, T},
+  {"TBP", EXTRACTOR_METATYPE_BEATS_PER_MINUTE, T},
+  {"TCR", EXTRACTOR_METATYPE_COPYRIGHT, T},
+  {"TPB", EXTRACTOR_METATYPE_PUBLISHER, T},
+  {"TEN", EXTRACTOR_METATYPE_ENCODED_BY, T},
+  {"TSS", EXTRACTOR_METATYPE_PRODUCED_BY_SOFTWARE, T},
+  {"TOF", EXTRACTOR_METATYPE_FILENAME, T},
+  {"TLE", EXTRACTOR_METATYPE_DURATION, T}, /* FIXME: should append 'ms' as unit */
+  {"TSI", EXTRACTOR_METATYPE_EMBEDDED_FILE_SIZE, T},
+  /* skipping TDY, TKE */
+  {"TOT", EXTRACTOR_METATYPE_ORIGINAL_TITLE, T},
+  {"TOA", EXTRACTOR_METATYPE_ORIGINAL_ARTIST, T},
+  {"TOL", EXTRACTOR_METATYPE_ORIGINAL_WRITER, T},
+  {"TOR", EXTRACTOR_METATYPE_ORIGINAL_RELEASE_YEAR, T},
+  /* skipping TXX */
+
+  {"WAF", EXTRACTOR_METATYPE_URL, U},
+  {"WAR", EXTRACTOR_METATYPE_URL, U},
+  {"WAS", EXTRACTOR_METATYPE_URL, U},
+  {"WCM", EXTRACTOR_METATYPE_URL, U},
+  {"WCP", EXTRACTOR_METATYPE_RIGHTS, U},
+  {"WCB", EXTRACTOR_METATYPE_URL, U},
+  /* skipping WXX */
+  {"IPL", EXTRACTOR_METATYPE_CONTRIBUTOR_NAME, T},
+  /* skipping MCI */
+  /* skipping ETC */
+  /* skipping MLL */
+  /* skipping STC */
+  {"ULT", EXTRACTOR_METATYPE_LYRICS, UL},
+  {"SLT", EXTRACTOR_METATYPE_LYRICS, SL},
+  {"COM", EXTRACTOR_METATYPE_COMMENT, L},
+  /* skipping RVA */
+  /* skipping EQU */
+  /* skipping REV */
+  {"PIC", EXTRACTOR_METATYPE_PICTURE, I},
+  /* skipping GEN */
+  /* {"CNT", EXTRACTOR_METATYPE_PLAY_COUNTER, XXX}, */
+  /*  {"POP", EXTRACTOR_METATYPE_POPULARITY_METER, XXX}, */
+  /* skipping BUF */
+  /* skipping CRM */
+  /* skipping CRA */
+  /* {"LNK", EXTRACTOR_METATYPE_URL, XXX}, */
   {NULL, 0},
 };
 
@@ -83,6 +128,9 @@ EXTRACTOR_id3v2_extract (const unsigned char *data,
   int unsync;
   unsigned int tsize;
   unsigned int pos;
+  unsigned int off;
+  enum EXTRACTOR_MetaType type;
+  const char *mime;
 
   if ((size < 16) ||
       (data[0] != 0x49) ||
@@ -102,10 +150,10 @@ EXTRACTOR_id3v2_extract (const unsigned char *data,
       size_t csize;
       int i;
 
-      if (pos + 6 > tsize)
+      if (pos + 7 > tsize)
         return 0;
       csize = (data[pos + 3] << 16) + (data[pos + 4] << 8) + data[pos + 5];
-      if ((pos + 6 + csize > tsize) || (csize > tsize) || (csize == 0))
+      if ((pos + 7 + csize > tsize) || (csize > tsize) || (csize == 0))
         break;
       i = 0;
       while (tmap[i].text != NULL)
@@ -116,33 +164,190 @@ EXTRACTOR_id3v2_extract (const unsigned char *data,
               /* this byte describes the encoding
                  try to convert strings to UTF-8
                  if it fails, then forget it */
-              switch (data[pos + 6])
-                {
-                case 0x00:
-                  word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 7],
-                                        csize, "ISO-8859-1");
-                  break;
-                case 0x01:
-                  word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 7],
-                                        csize, "UCS-2");
-                  break;
-                default:
-                  /* bad encoding byte,
-                     try to convert from iso-8859-1 */
-                  word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 7],
-                                        csize, "ISO-8859-1");
-                  break;
-                }
-              pos++;
-              csize--;
-              if ((word != NULL) && (strlen (word) > 0))
-                {
-                  prev = addKeyword (prev, word, tmap[i].type);
-                }
-              else
+              switch (tmap[i].fmt)
+                {
+                case T:           
+                  switch (data[pos + 6])
+                    {
+                    case 0x00:
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 7],
+                                                               csize - 1, "ISO-8859-1");
+                      break;
+                    case 0x01:
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 7],
+                                                               csize - 1, "UCS-2");
+                      break;
+                    default:
+                      /* bad encoding byte,
+                         try to convert from iso-8859-1 */
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 7],
+                                                               csize - 1, "ISO-8859-1");
+                      break;
+                    }
+                  break;
+                case U:
+                  word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 6],
+                                                           csize, "ISO-8859-1");
+                  break;
+                case UL:
+                  if (csize < 6)
+                    return 0; /* malformed */
+                  /* find end of description */
+                  off = 10;
+                  while ( (off < size) &&
+                          (off - pos < csize) &&
+                          (data[pos + off] == '\0') )
+                    off++;
+                  if ( (off >= csize) ||
+                       (data[pos+off] != '\0') )
+                    return 0; /* malformed */
+                  off++;
+                  switch (data[pos + 6])
+                    {
+                    case 0x00:
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + off],
+                                                               csize - off, "ISO-8859-1");
+                      break;
+                    case 0x01:
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + off],
+                                                               csize - off, "UCS-2");
+                      break;
+                    default:
+                      /* bad encoding byte,
+                         try to convert from iso-8859-1 */
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + off],
+                                                               csize - off, "ISO-8859-1");
+                      break;
+                    }
+                  break;
+                case SL:
+                  if (csize < 7)
+                    return 0; /* malformed */
+                  /* find end of description */
+                  switch (data[pos + 6])
+                    {
+                    case 0x00:
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 12],
+                                                               csize - 6, "ISO-8859-1");
+                      break;
+                    case 0x01:
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 12],
+                                                               csize - 6, "UCS-2");
+                      break;
+                    default:
+                      /* bad encoding byte,
+                         try to convert from iso-8859-1 */
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 12],
+                                                               csize - 6, "ISO-8859-1");
+                      break;
+                    }
+                  break;
+                case L:
+                  if (csize < 5)
+                    return 0; /* malformed */
+                  /* find end of description */
+                  switch (data[pos + 6])
+                    {
+                    case 0x00:
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 10],
+                                                               csize - 4, "ISO-8859-1");
+                      break;
+                    case 0x01:
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 10],
+                                                               csize - 4, "UCS-2");
+                      break;
+                    default:
+                      /* bad encoding byte,
+                         try to convert from iso-8859-1 */
+                      word = EXTRACTOR_common_convert_to_utf8 ((const char *) &data[pos + 10],
+                                                               csize - 4, "ISO-8859-1");
+                      break;
+                    }
+                  break;
+                case I:
+                  if (csize < 6)
+                    return 0; /* malformed */
+                  /* find end of description */
+                  off = 12;
+                  while ( (off < size) &&
+                          (off - pos < csize) &&
+                          (data[pos + off] == '\0') )
+                    off++;
+                  if ( (off >= csize) ||
+                       (data[pos+off] != '\0') )
+                    return 0; /* malformed */
+                  off++;
+                  switch (data[pos+11])
+                    {
+                    case 0x03:
+                    case 0x04:
+                      type = EXTRACTOR_METATYPE_COVER_PICTURE;
+                      break;
+                    case 0x07:
+                    case 0x08:
+                    case 0x09:
+                    case 0x0A:
+                    case 0x0B:
+                    case 0x0C:
+                      type = EXTRACTOR_METATYPE_CONTRIBUTOR_PICTURE;
+                      break;
+                    case 0x0D:
+                    case 0x0E:
+                    case 0x0F:
+                      type = EXTRACTOR_METATYPE_EVENT_PICTURE;
+                      break;
+                    case 0x14:
+                      type = EXTRACTOR_METATYPE_LOGO;
+                      type = EXTRACTOR_METATYPE_LOGO;
+                      break;
+                    default:
+                      type = EXTRACTOR_METATYPE_PICTURE;
+                      break;
+                    }
+                  if (0 == strncasecmp ("PNG",
+                                        (const char*) &data[pos + 7], 3))
+                    mime = "image/png";
+                  else if (0 == strncasecmp ("JPG",
+                                             (const char*) &data[pos + 7], 3))
+                    mime = "image/jpeg";
+                  else
+                    mime = NULL;
+                  if (0 == strncasecmp ("-->",
+                                        (const char*) &data[pos + 7], 3))
+                    {
+                      /* not supported */
+                    }
+                  else
+                    {
+                      if (0 != proc (proc_cls,
+                                     "id3v2",
+                                     type,
+                                     EXTRACTOR_METAFORMAT_BINARY,
+                                     mime,
+                                     (const char*) &data[pos + off],
+                                     csize + 6 - off))                  
+                        return 1;
+                    }
+                  word = NULL;
+                  break;
+                default:
+                  return 0;
+                }
+              if ((word != NULL) && (strlen (word) > 0))
                 {
-                  free (word);
-                }
+                  if (0 != proc (proc_cls,
+                                 "id3v2",
+                                 type,
+                                 EXTRACTOR_METAFORMAT_UTF8,
+                                 "text/plain",
+                                 word,
+                                 strlen(word)+1))
+                    {
+                      free (word);
+                      return 1;
+                    }
+                }
+              free (word);
               break;
             }
           i++;