tracker r2225 - in branches/indexer-split: . src/tracker-extract
- From: mr svn gnome org
- To: svn-commits-list gnome org
- Subject: tracker r2225 - in branches/indexer-split: . src/tracker-extract
- Date: Fri, 12 Sep 2008 09:04:28 +0000 (UTC)
Author: mr
Date: Fri Sep 12 09:04:28 2008
New Revision: 2225
URL: http://svn.gnome.org/viewvc/tracker?rev=2225&view=rev
Log:
* src/tracker-extract/tracker-extract-mp3.c: Clean up the code.
Modified:
branches/indexer-split/ChangeLog
branches/indexer-split/src/tracker-extract/tracker-extract-mp3.c
Modified: branches/indexer-split/src/tracker-extract/tracker-extract-mp3.c
==============================================================================
--- branches/indexer-split/src/tracker-extract/tracker-extract-mp3.c (original)
+++ branches/indexer-split/src/tracker-extract/tracker-extract-mp3.c Fri Sep 12 09:04:28 2008
@@ -1,6 +1,7 @@
/* -*- Mode: C; tab-width: 8; indent-tabs-mode: t; c-basic-offset: 8 -*- */
-/* Tracker Extract - extracts embedded metadata from files
+/*
* Copyright (C) 2006, Mr Jamie McCracken (jamiemcc gnome org)
+ * Copyright (C) 2008, Nokia
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public
@@ -18,12 +19,12 @@
* Boston, MA 02110-1301, USA.
*/
+#include "config.h"
+
#ifndef _GNU_SOURCE
#define _GNU_SOURCE
#endif
-#include "config.h"
-
#include <sys/types.h>
#include <sys/stat.h>
#ifndef OS_WIN32
@@ -34,68 +35,45 @@
#include <string.h>
#include <unistd.h>
#include <fcntl.h>
+
#include <glib.h>
#include <glib/gstdio.h>
-#include "tracker-extract.h"
-#define MAX_FILE_READ 1024 * 1024 * 10
+#include "tracker-extract.h"
+#define MAX_FILE_READ 1024 * 1024 * 10
+#define MAX_MP3_SCAN_DEEP 16768
typedef struct {
- char * text;
- char * type;
+ gchar *text;
+ gchar *type;
} Matches;
-#if 0 /* Unused */
+typedef struct {
+ gchar *title;
+ gchar *artist;
+ gchar *album;
+ gchar *year;
+ gchar *comment;
+ gchar *genre;
+} id3tag;
-static struct {
- char * name;
- char *meta_name;
- gboolean writable;
-} tags[] = {
- {"title", "Audio:Title", FALSE},
- {"artist", "Audio:Artist", FALSE},
- {"album", "Audio:Album", FALSE},
- {"albumartist", "Audio:AlbumArtist", FALSE},
- {"trackcount", "Audio:AlbumTrackCount", FALSE},
- {"tracknumber", "Audio:TrackNo", FALSE},
- {"DiscNo", "Audio:DiscNo", FALSE},
- {"Performer", "Audio:Performer", FALSE},
- {"TrackGain", "Audio:TrackGain", FALSE},
- {"TrackPeakGain", "Audio:TrackPeakGain", FALSE},
- {"AlbumGain", "Audio:AlbumGain", FALSE},
- {"AlbumPeakGain", "Audio:AlbumPeakGain", FALSE},
- {"date", "Audio:ReleaseDate", FALSE},
- {"comment", "Audio:Comment", FALSE},
- {"genre", "Audio:Genre", FALSE},
- {"Codec", "Audio:Codec", FALSE},
- {"CodecVersion", "Audio:CodecVersion", FALSE},
- {"Samplerate", "Audio:Samplerate", FALSE},
- {"Channels", "Audio:Channels", FALSE},
- {"MBAlbumID", "Audio:MBAlbumID", FALSE},
- {"MBArtistID", "Audio:MBArtistID", FALSE},
- {"MBAlbumArtistID", "Audio:MBAlbumArtistID", FALSE},
- {"MBTrackID", "Audio:MBTrackID", FALSE},
- {"Lyrics", "Audio:Lyrics", FALSE},
- {"Copyright", "File:Copyright", FALSE},
- {"License", "File:License", FALSE},
- {"Organization", "File:Organization", FALSE},
- {"Location", "DC:Location", FALSE},
- {"Publisher", "DC:Publishers", FALSE},
- {NULL, NULL, FALSE},
+enum {
+ MPEG_ERR,
+ MPEG_V1,
+ MPEG_V2,
+ MPEG_V25
};
-#endif /* Unused */
-
-typedef struct {
- char * title;
- char * artist;
- char * album;
- char * year;
- char * comment;
- char * genre;
-} id3tag;
+enum {
+ LAYER_ERR,
+ LAYER_1,
+ LAYER_2,
+ LAYER_3
+};
+static void extract_mp3 (const gchar *filename,
+ GHashTable *metadata);
static const char *const genre_names[] = {
"Blues",
@@ -248,56 +226,53 @@
"Synthpop"
};
+static const guint max_frames_scan = 1024;
-#define GENRE_NAME_COUNT \
- ((unsigned int)(sizeof genre_names / sizeof (const char *const)))
-
-
-#define MAX_MP3_SCAN_DEEP 16768
-const unsigned int max_frames_scan=1024;
-enum{ MPEG_ERR=0,MPEG_V1=1,MPEG_V2=2,MPEG_V25=3};
-
-enum{ LAYER_ERR=0,LAYER_1=1,LAYER_2=2,LAYER_3=3};
-
-const unsigned int sync_mask=0xE0FF;
-const unsigned int mpeg_ver_mask=0x1800;
-const unsigned int mpeg_layer_mask=0x600;
-const unsigned int bitrate_mask=0xF00000;
-const unsigned int freq_mask=0xC0000;
-const unsigned int ch_mask=0xC0000000;
-const unsigned int pad_mask=0x20000;
-
-unsigned int bitrate_table[16][6]={
- {0,0,0,0,0,0},
- {32, 32, 32, 32, 32, 8},
- {64, 48, 40, 64, 48, 16},
- {96, 56, 48, 96, 56, 24},
- {128, 64 , 56 , 128, 64 , 32},
- {160, 80 , 64 , 160, 80 , 64},
- {192, 96 , 80 , 192, 96 , 80},
- {224, 112, 96 , 224, 112, 56},
- {256, 128, 112, 256, 128, 64},
- {288, 160, 128, 288, 160, 128},
- {320, 192, 160, 320, 192, 160},
- {352, 224, 192, 352, 224, 112},
- {384, 256, 224, 384, 256, 128},
- {416, 320, 256, 416, 320, 256},
- {448, 384, 320, 448, 384, 320},
- {-1,-1,-1,-1,-1,-1}
-};
-int freq_table[4][3]={
- {44100,22050,11025},
- {48000,24000,12000},
- {32000,16000,8000}
+static const guint sync_mask = 0xE0FF;
+static const guint mpeg_ver_mask = 0x1800;
+static const guint mpeg_layer_mask = 0x600;
+static const guint bitrate_mask = 0xF00000;
+static const guint freq_mask = 0xC0000;
+static const guint ch_mask = 0xC0000000;
+static const guint pad_mask = 0x20000;
+
+static guint bitrate_table[16][6] = {
+ {0 , 0 , 0 , 0 , 0 , 0},
+ {32 , 32 , 32 , 32 , 32 , 8},
+ {64 , 48 , 40 , 64 , 48 , 16},
+ {96 , 56 , 48 , 96 , 56 , 24},
+ {128, 64 , 56 , 128, 64 , 32},
+ {160, 80 , 64 , 160, 80 , 64},
+ {192, 96 , 80 , 192, 96 , 80},
+ {224, 112, 96 , 224, 112, 56},
+ {256, 128, 112, 256, 128, 64},
+ {288, 160, 128, 288, 160, 128},
+ {320, 192, 160, 320, 192, 160},
+ {352, 224, 192, 352, 224, 112},
+ {384, 256, 224, 384, 256, 128},
+ {416, 320, 256, 416, 320, 256},
+ {448, 384, 320, 448, 384, 320},
+ {-1, -1, -1, -1, -1, -1}
};
+static gint freq_table[4][3] = {
+ {44100, 22050, 11025},
+ {48000, 24000, 12000},
+ {32000, 16000, 8000}
+};
-#define OK 0
-#define SYSERR 1
-#define INVALID_ID3 2
+static TrackerExtractorData data[] = {
+ { "audio/mpeg", extract_mp3 },
+ { "audio/x-mp3", extract_mp3 },
+ { NULL, NULL }
+};
-static char *
-get_utf8 (const char *txt, int size, gpointer p1, gpointer p2, gpointer p3)
+static gchar *
+get_utf8 (const gchar *txt,
+ gint size,
+ gpointer p1,
+ gpointer p2,
+ gpointer p3)
{
if (!g_utf8_validate (txt, size, NULL)) {
return g_locale_to_utf8 (txt, size, NULL, NULL, NULL);
@@ -306,220 +281,233 @@
}
}
-static int get_id3 (const char * data,
- size_t size,
- id3tag * id3) {
-
- const char * pos;
-
+static gboolean
+get_id3 (const gchar *data,
+ size_t size,
+ id3tag *id3)
+{
+ const gchar *pos;
- if (size < 128) return INVALID_ID3;
+ if (size < 128) {
+ return FALSE;
+ }
pos = &data[size - 128];
- if (0 != strncmp ("TAG",
- pos,
- 3)) {
- return INVALID_ID3;
+ if (strncmp ("TAG", pos, 3) != 0) {
+ return FALSE;
}
pos += 3;
- id3->title = get_utf8 (pos,
- 30,
- NULL, NULL, NULL);
+ id3->title = get_utf8 (pos, 30, NULL, NULL, NULL);
pos += 30;
- id3->artist = get_utf8 (pos,
- 30,
- NULL, NULL, NULL);
+ id3->artist = get_utf8 (pos, 30, NULL, NULL, NULL);
pos += 30;
- id3->album = get_utf8 (pos,
- 30,
- NULL, NULL, NULL);
+ id3->album = get_utf8 (pos, 30, NULL, NULL, NULL);
pos += 30;
id3->year = get_utf8 (pos, 4, NULL, NULL, NULL);
pos += 4;
- id3->comment = get_utf8 (pos,
- 30,
- NULL, NULL, NULL);
+ id3->comment = get_utf8 (pos, 30, NULL, NULL, NULL);
pos += 30;
id3->genre = "";
- if ( (unsigned int)pos[0] < GENRE_NAME_COUNT) {
- id3->genre = g_strdup (genre_names[(unsigned) pos[0]]);
- }
-
- return OK;
-}
-
-#if 0 /* Unused */
-static gboolean
-tracker_metadata_mp3_is_writable (const char *meta)
-{
- int i;
-
- i = 0;
- while (tags[i].name != NULL) {
-
- if (strcmp (tags[i].meta_name, meta) == 0) {
- return tags[i].writable;
- }
-
- i++;
+ if ((guint) pos[0] < G_N_ELEMENTS (genre_names)) {
+ id3->genre = g_strdup (genre_names[(unsigned) pos[0]]);
}
- return FALSE;
-
-}
-
-static gboolean
-tracker_metadata_mp3_write (const char *meta_name, const char *value)
-{
- /* to do */
- return FALSE;
+ return TRUE;
}
-#endif /* Unused */
-
static void
-mp3_parse (const char *data, size_t size, GHashTable *metadata)
+mp3_parse (const gchar *data,
+ size_t size,
+ GHashTable *metadata)
{
-
- unsigned int header;
- int counter=0;
- char mpeg_ver=0;
- char layer_ver=0;
- int idx_num=0;
- unsigned int bitrate=0; /*used for each frame*/
- unsigned int avg_bps=0; /*average bitrate*/
- int vbr_flag=0;
- unsigned int length=0;
- unsigned int sample_rate=0;
- int ch=0;
- unsigned int frame_size;
- unsigned int frames=0;
+ guint header;
+ gint counter = 0;
+ gchar mpeg_ver = 0;
+ gchar layer_ver = 0;
+ gint idx_num = 0;
+ guint bitrate = 0;
+ guint avg_bps = 0;
+ gint vbr_flag = 0;
+ guint length = 0;
+ guint sample_rate = 0;
+ gint ch = 0;
+ guint frame_size;
+ guint frames = 0;
size_t pos = 0;
do {
- /* seek for frame start */
+ /* Seek for frame start */
if (pos + sizeof(header) > size) {
return;
- }/*unable to find header*/
+ }
memcpy (&header, &data[pos], sizeof (header));
- if ((header&sync_mask) == sync_mask) break;/*found header sync*/
+ if ((header&sync_mask) == sync_mask) {
+ /* Found header sync */
+ break;
+ }
pos++;
- counter++; /*next try*/
-
+ counter++;
} while (counter < MAX_MP3_SCAN_DEEP);
- if (counter>=MAX_MP3_SCAN_DEEP) {
+ if (counter >= MAX_MP3_SCAN_DEEP) {
+ /* Give up to find mp3 header */
return;
- };/*give up to find mp3 header*/
-
+ };
- do { /*ok, now we found a mp3 frame header*/
+ do {
frames++;
- switch (header & mpeg_ver_mask){
+ switch (header & mpeg_ver_mask) {
case 0x1000:
- mpeg_ver = MPEG_ERR; /*error*/
+ mpeg_ver = MPEG_ERR;
break;
case 0x800:
- g_hash_table_insert (metadata, g_strdup ("Audio:Codec"), g_strdup ("MPEG"));
- g_hash_table_insert (metadata, g_strdup ("Audio:CodecVersion"), g_strdup ("2"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Codec"),
+ g_strdup ("MPEG"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:CodecVersion"),
+ g_strdup ("2"));
mpeg_ver = MPEG_V2;
break;
case 0x1800:
- g_hash_table_insert (metadata, g_strdup ("Audio:Codec"), g_strdup ("MPEG"));
- g_hash_table_insert (metadata, g_strdup ("Audio:CodecVersion"), g_strdup ("1"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Codec"),
+ g_strdup ("MPEG"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:CodecVersion"),
+ g_strdup ("1"));
mpeg_ver = MPEG_V1;
break;
case 0:
- g_hash_table_insert (metadata, g_strdup ("Audio:Codec"), g_strdup ("MPEG"));
- g_hash_table_insert (metadata, g_strdup ("Audio:CodecVersion"), g_strdup ("2.5"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Codec"),
+ g_strdup ("MPEG"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:CodecVersion"),
+ g_strdup ("2.5"));
mpeg_ver = MPEG_V25;
break;
}
- switch(header&mpeg_layer_mask){
+
+ switch (header&mpeg_layer_mask) {
case 0x400:
- layer_ver=LAYER_2;
+ layer_ver = LAYER_2;
break;
case 0x200:
- layer_ver=LAYER_3;
+ layer_ver = LAYER_3;
break;
case 0x600:
- layer_ver=LAYER_1;
+ layer_ver = LAYER_1;
break;
case 0:
- layer_ver=LAYER_ERR;/*error*/
+ layer_ver = LAYER_ERR;
+ }
+
+ if (!layer_ver || !mpeg_ver) {
+ /* Unknown mpeg type */
+ return;
}
- if (!layer_ver||!mpeg_ver)
- return; /*unknown mpeg type*/
- if (mpeg_ver<3)
- idx_num=(mpeg_ver-1)*3+layer_ver-1;
- else
- idx_num=2+layer_ver;
- bitrate = 1000 * bitrate_table[(header&bitrate_mask)>>20][idx_num];
+ if (mpeg_ver<3) {
+ idx_num = (mpeg_ver - 1) * 3 + layer_ver - 1;
+ } else {
+ idx_num = 2 + layer_ver;
+ }
+
+ bitrate = 1000 * bitrate_table[(header & bitrate_mask) >> 20][idx_num];
- if (bitrate<0) {
+ if (bitrate < 0) {
frames--;
break;
}
- sample_rate = freq_table[(header&freq_mask)>>18][mpeg_ver-1];
- if (sample_rate<0) {
+ sample_rate = freq_table[(header & freq_mask) >> 18][mpeg_ver - 1];
+ if (sample_rate < 0) {
+ /* Error in header */
frames--;
break;
- } /*error in header*/
+ }
- if ((header&ch_mask)==ch_mask) {
- ch=1;
- g_hash_table_insert (metadata, g_strdup ("Audio:Channels"), g_strdup ("1"));
+ if ((header & ch_mask) == ch_mask) {
+ ch = 1;
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Channels"),
+ g_strdup ("1"));
} else {
ch=2; /*stereo non stereo select*/
- g_hash_table_insert (metadata, g_strdup ("Audio:Channels"), g_strdup ("2"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Channels"),
+ g_strdup ("2"));
}
- frame_size = 144*bitrate/(sample_rate?sample_rate:1)+((header&pad_mask)>>17);
- avg_bps += bitrate/1000;
+ frame_size = 144 * bitrate / (sample_rate ? sample_rate : 1) + ((header & pad_mask) >> 17);
+ avg_bps += bitrate / 1000;
+
+ pos += frame_size - 4;
- pos += frame_size-4;
+ if (frames > max_frames_scan) {
+ /* Optimization */
+ break;
+ }
- if (frames > max_frames_scan)
- break; /*optimization*/
- if (avg_bps/frames!=bitrate/1000)
- vbr_flag=1;
- if (pos + sizeof(header) > size)
- break; /* EOF */
- memcpy(&header,
- &data[pos],
- sizeof(header));
- } while ((header&sync_mask)==sync_mask);
+ if (avg_bps / frames != bitrate / 1000) {
+ vbr_flag = 1;
+ }
- if (!frames) return; /*no valid frames*/
+ if (pos + sizeof (header) > size) {
+ /* EOF */
+ break;
+ }
- avg_bps = avg_bps/frames;
+ memcpy(&header, &data[pos], sizeof (header));
+ } while ((header & sync_mask) == sync_mask);
- if (max_frames_scan){ /*if not all frames scaned*/
- length=size/(avg_bps?avg_bps:bitrate?bitrate:0xFFFFFFFF)/125;
- } else{
- length=1152*frames/(sample_rate?sample_rate:0xFFFFFFFF);
+ if (!frames) {
+ /* No valid frames */
+ return;
}
- g_hash_table_insert (metadata, g_strdup ("Audio:Duration"), g_strdup_printf ("%d", length));
- g_hash_table_insert (metadata, g_strdup ("Audio:Samplerate"), g_strdup_printf ("%d", sample_rate));
- g_hash_table_insert (metadata, g_strdup ("Audio:Bitrate"), g_strdup_printf ("%d", avg_bps));
+ avg_bps /= frames;
-}
+ if (max_frames_scan) {
+ /* If not all frames scaned */
+ length = size / (avg_bps ? avg_bps : bitrate ? bitrate : 0xFFFFFFFF) / 125;
+ } else{
+ length = 1152 * frames / (sample_rate ? sample_rate : 0xFFFFFFFF);
+ }
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Duration"),
+ g_strdup_printf ("%d", length));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Samplerate"),
+ g_strdup_printf ("%d", sample_rate));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Bitrate"),
+ g_strdup_printf ("%d", avg_bps));
+}
static void
-get_id3v24_tags (const char *data, size_t size, GHashTable *metadata)
+get_id3v24_tags (const gchar *data,
+ size_t size,
+ GHashTable *metadata)
{
+ gint unsync;
+ gint extendedHdr;
+ gint experimental;
+ gint footer;
+ guint tsize;
+ guint pos;
+ guint ehdrSize;
+ guint padding;
Matches tmap[] = {
{"COMM", "Audio:Comment"},
{"TCOP", "File:Copyright"},
@@ -542,25 +530,12 @@
{NULL, 0},
};
-
-
- int unsync;
- int extendedHdr;
- int experimental;
- int footer;
- unsigned int tsize;
- unsigned int pos;
- unsigned int ehdrSize;
- unsigned int padding;
-
- if ( (size < 16) ||
- (data[0] != 0x49) ||
- (data[1] != 0x44) ||
- (data[2] != 0x33) ||
- (data[3] != 0x04) ||
- (data[4] != 0x00) ) {
-
- ////g_print ("no id3v24 tags info (stage 1)\n");
+ if ((size < 16) ||
+ (data[0] != 0x49) ||
+ (data[1] != 0x44) ||
+ (data[2] != 0x33) ||
+ (data[3] != 0x04) ||
+ (data[4] != 0x00) ) {
return;
}
@@ -568,130 +543,144 @@
extendedHdr = (data[5] & 0x40) > 0;
experimental = (data[5] & 0x20) > 0;
footer = (data[5] & 0x10) > 0;
- tsize = ( ( (data[6] & 0x7F) << 21 ) |
- ( (data[7] & 0x7F) << 14 ) |
- ( (data[8] & 0x7F) << 7 ) |
- ( (data[9] & 0x7F) << 0 ) );
-
- if ( (tsize + 10 > size) || (experimental) ) {
+ tsize = (((data[6] & 0x7F) << 21) |
+ ((data[7] & 0x7F) << 14) |
+ ((data[8] & 0x7F) << 7) |
+ ((data[9] & 0x7F) << 0));
- ////g_print ("no id3v24 tags info (stage 2)\n");
+ if ((tsize + 10 > size) || (experimental)) {
return;
}
-
pos = 10;
padding = 0;
+
if (extendedHdr) {
- ehdrSize = ( ( (data[10] & 0x7F) << 21 ) |
- ( (data[11] & 0x7F) << 14 ) |
- ( (data[12] & 0x7F) << 7 ) |
- ( (data[13] & 0x7F) << 0 ) );
+ ehdrSize = (((data[10] & 0x7F) << 21) |
+ ((data[11] & 0x7F) << 14) |
+ ((data[12] & 0x7F) << 7) |
+ ((data[13] & 0x7F) << 0));
pos += ehdrSize;
}
-
while (pos < tsize) {
size_t csize;
- int i;
+ gint i;
unsigned short flags;
if (pos + 10 > tsize) {
-
- ////g_print ("no id3v24 tags info (stage 3)\n");
return;
}
- csize = ( ( (data[pos+4] & 0x7F) << 21 ) |
- ( (data[pos+5] & 0x7F) << 14 ) |
- ( (data[pos+6] & 0x7F) << 7 ) |
- ( (data[pos+7] & 0x7F) << 0 ) );
-
- if ( (pos + 10 + csize > tsize) ||
- (csize > tsize) ||
- (csize == 0) )
+ csize = (((data[pos+4] & 0x7F) << 21) |
+ ((data[pos+5] & 0x7F) << 14) |
+ ((data[pos+6] & 0x7F) << 7) |
+ ((data[pos+7] & 0x7F) << 0));
+
+ if ((pos + 10 + csize > tsize) ||
+ (csize > tsize) ||
+ (csize == 0)) {
break;
+ }
- flags = (data[pos+8]<<8) + data[pos+9];
- if ( ( (flags & 0x80) > 0) /* compressed, not yet supported */ ||
- ( (flags & 0x40) > 0) /* encrypted, not supported */ ) {
+ flags = (data[pos + 8] << 8) + data[pos + 9];
+ if (((flags & 0x80) > 0) ||
+ ((flags & 0x40) > 0)) {
pos += 10 + csize;
continue;
}
i = 0;
while (tmap[i].text != NULL) {
- if (0 == strncmp(tmap[i].text,
- (const char*) &data[pos],
- 4)) {
- char * word;
- if ( (flags & 0x20) > 0) {
- /* "group" identifier, skip a byte */
+ if (strncmp (tmap[i].text, (const char*) &data[pos], 4) == 0) {
+ gchar * word;
+
+ if ((flags & 0x20) > 0) {
+ /* The "group" identifier, skip a byte */
pos++;
csize--;
}
- /* this byte describes the encoding
- try to convert strings to UTF-8
- if it fails, then forget it */
- switch (data[pos+10]) {
- case 0x00 :
- word = get_utf8 ((const char*) &data[pos+11],
- csize,
- NULL, NULL, NULL);
- break;
- case 0x01 :
- word = get_utf8((const char*) &data[pos+11],
- csize,
- NULL, NULL, NULL);
- break;
- case 0x02 :
- word = get_utf8((const char*) &data[pos+11],
- csize,
- NULL, NULL, NULL);
- break;
- case 0x03 :
- word = malloc(csize+1);
- memcpy(word,
- &data[pos+11],
- csize);
- word[csize] = '\0';
- break;
-
- default:
- /* bad encoding byte,
- try to convert from iso-8859-1 */
- word = get_utf8((const char*) &data[pos+11],
+ /* This byte describes the encoding
+ * try to convert strings to UTF-8
+ * if it fails, then forget it
+ */
+ switch (data[pos + 10]) {
+ case 0x00:
+ word = get_utf8 ((const char*) &data[pos + 11],
+ csize,
+ NULL, NULL, NULL);
+ break;
+ case 0x01 :
+ word = get_utf8 ((const char*) &data[pos + 11],
+ csize,
+ NULL, NULL, NULL);
+ break;
+ case 0x02 :
+ word = get_utf8 ((const char*) &data[pos + 11],
+ csize,
+ NULL, NULL, NULL);
+ break;
+ case 0x03 :
+ word = malloc (csize + 1);
+ memcpy (word,
+ &data[pos + 11],
+ csize);
+ word[csize] = '\0';
+ break;
+
+ default:
+ /* Bad encoding byte,
+ * try to convert from
+ * iso-8859-1
+ */
+ word = get_utf8 ((const char*) &data[pos + 11],
csize,
NULL, NULL, NULL);
- break;
+ break;
}
+
pos++;
csize--;
- if ((word != NULL) && (strlen(word) > 0)) {
+ if (word != NULL && strlen (word) > 0) {
if (strcmp (tmap[i].text, "COMM") == 0) {
- char *s = g_strdup (word+strlen(word)+1);
+ gchar *s;
+
+ s = g_strdup (word + strlen (word) + 1);
g_free (word);
word = s;
}
- g_hash_table_insert (metadata, g_strdup (tmap[i].type), g_strdup (word));
+ g_hash_table_insert (metadata,
+ g_strdup (tmap[i].type),
+ g_strdup (word));
} else {
g_free (word);
}
+
break;
}
+
i++;
}
+
pos += 10 + csize;
}
}
-
static void
-get_id3v23_tags (const char *data, size_t size, GHashTable *metadata)
+get_id3v23_tags (const gchar *data,
+ size_t size,
+ GHashTable *metadata)
{
+ gint unsync;
+ gint extendedHdr;
+ gint experimental;
+ guint tsize;
+ guint pos;
+ guint ehdrSize;
+ guint padding;
Matches tmap[] = {
{"COMM", "Audio:Comment"},
{"TCOP", "File:Copyright"},
@@ -714,142 +703,152 @@
{NULL, 0},
};
- int unsync;
- int extendedHdr;
- int experimental;
- unsigned int tsize;
- unsigned int pos;
- unsigned int ehdrSize;
- unsigned int padding;
-
- if ( (size < 16) ||
- (data[0] != 0x49) ||
- (data[1] != 0x44) ||
- (data[2] != 0x33) ||
- (data[3] != 0x03) ||
- (data[4] != 0x00) ) {
-
- //g_print ("no id3v23 tags info (stage 1)\n");
+ if ((size < 16) ||
+ (data[0] != 0x49) ||
+ (data[1] != 0x44) ||
+ (data[2] != 0x33) ||
+ (data[3] != 0x03) ||
+ (data[4] != 0x00)) {
return;
}
+
unsync = (data[5] & 0x80) > 0;
extendedHdr = (data[5] & 0x40) > 0;
experimental = (data[5] & 0x20) > 0;
- tsize = ( ( (data[6] & 0x7F) << 21 ) |
- ( (data[7] & 0x7F) << 14 ) |
- ( (data[8] & 0x7F) << 7 ) |
- ( (data[9] & 0x7F) << 0 ) );
- if ( (tsize + 10 > size) || (experimental) ) {
+ tsize = (((data[6] & 0x7F) << 21) |
+ ((data[7] & 0x7F) << 14) |
+ ((data[8] & 0x7F) << 7) |
+ ((data[9] & 0x7F) << 0));
- //g_print ("no id3v23 tags info (stage 2)\n");
+ if ((tsize + 10 > size) || (experimental)) {
return;
}
+
pos = 10;
padding = 0;
+
if (extendedHdr) {
- ehdrSize = ( ( (data[10]) << 24 ) |
- ( (data[11]) << 16 ) |
- ( (data[12]) << 8 ) |
- ( (data[12]) << 0 ) );
-
- padding = ( ( (data[15]) << 24 ) |
- ( (data[16]) << 16 ) |
- ( (data[17]) << 8 ) |
- ( (data[18]) << 0 ) );
+ ehdrSize = (((data[10]) << 24) |
+ ((data[11]) << 16) |
+ ((data[12]) << 8) |
+ ((data[12]) << 0));
+
+ padding = (((data[15]) << 24) |
+ ((data[16]) << 16) |
+ ((data[17]) << 8) |
+ ((data[18]) << 0));
+
pos += 4 + ehdrSize;
+
if (padding < tsize)
tsize -= padding;
else {
- //g_print ("no id3v23 tags info (stage 3)\n");
return;
}
}
-
while (pos < tsize) {
size_t csize;
- int i;
+ gint i;
unsigned short flags;
if (pos + 10 > tsize) {
-
- //g_print ("no id3v23 tags info (stage 4)\n");
return;
}
- csize = (data[pos+4] << 24) + (data[pos+5] << 16) + (data[pos+6] << 8) + data[pos+7];
- if ( (pos + 10 + csize > tsize) ||
- (csize > tsize) ||
- (csize == 0) )
+
+ csize = (data[pos + 4] << 24) +
+ (data[pos + 5] << 16) +
+ (data[pos + 6] << 8) +
+ data[pos + 7];
+
+ if ((pos + 10 + csize > tsize) ||
+ (csize > tsize) ||
+ (csize == 0)) {
break;
- flags = (data[pos+8]<<8) + data[pos+9];
- if ( ( (flags & 0x80) > 0) /* compressed, not yet supported */ ||
- ( (flags & 0x40) > 0) /* encrypted, not supported */ ) {
+ }
+
+ flags = (data[pos + 8] << 8) + data[pos + 9];
+
+ if (((flags & 0x80) > 0) || ((flags & 0x40) > 0)) {
pos += 10 + csize;
continue;
}
+
i = 0;
while (tmap[i].text != NULL) {
- if (0 == strncmp(tmap[i].text,
- (const char*) &data[pos],
- 4)) {
- char * word;
- if ( (flags & 0x20) > 0) {
- /* "group" identifier, skip a byte */
+ if (strncmp (tmap[i].text, (const gchar*) &data[pos], 4) == 0) {
+ gchar * word;
+
+ if ((flags & 0x20) > 0) {
+ /* The "group" identifier, skip a byte */
pos++;
csize--;
}
+
csize--;
- /* this byte describes the encoding
- try to convert strings to UTF-8
- if it fails, then forget it */
+
+ /* This byte describes the encoding
+ * try to convert strings to UTF-8 if
+ * it fails, then forget it./
+ */
- switch (data[pos+10]) {
- case 0x00 :
- word = get_utf8((const char*) &data[pos+11],
- csize,
- NULL, NULL, NULL);
- break;
- case 0x01 :
- word = get_utf8((const char*) &data[pos+11],
- csize,
- NULL, NULL, NULL);
- break;
- default:
- /* bad encoding byte,
- try to convert from iso-8859-1 */
- word = get_utf8((const char*) &data[pos+11],
- csize,
- NULL, NULL, NULL);
- break;
+ switch (data[pos + 10]) {
+ case 0x00:
+ word = get_utf8 ((const gchar*) &data[pos + 11],
+ csize,
+ NULL, NULL, NULL);
+ break;
+ case 0x01 :
+ word = get_utf8 ((const gchar*) &data[pos + 11],
+ csize,
+ NULL, NULL, NULL);
+ break;
+ default:
+ /* Bad encoding byte, try to
+ * convert from iso-8859-1
+ */
+ word = get_utf8 ((const gchar*) &data[pos + 11],
+ csize,
+ NULL, NULL, NULL);
+ break;
}
pos++;
- if ((word != NULL) && (strlen(word) > 0)) {
-
+
+ if (word != NULL && strlen(word) > 0) {
if (strcmp (tmap[i].text, "COMM") == 0) {
- char *s = g_strdup (word+strlen(word)+1);
+ gchar *s;
+
+ s = g_strdup (word + strlen (word) + 1);
g_free (word);
word = s;
}
- g_hash_table_insert (metadata, g_strdup (tmap[i].type), g_strdup (word));
+ g_hash_table_insert (metadata,
+ g_strdup (tmap[i].type),
+ g_strdup (word));
} else {
g_free (word);
}
+
break;
}
+
i++;
}
+
pos += 10 + csize;
}
-
}
-
static void
-get_id3v2_tags (const char *data, size_t size, GHashTable *metadata)
+get_id3v2_tags (const gchar *data,
+ size_t size,
+ GHashTable *metadata)
{
-
+ gint unsync;
+ guint tsize;
+ guint pos;
Matches tmap[] = {
{"TAL", "Audio:Title"},
{"TT1", "Audio:Artist"},
@@ -879,113 +878,108 @@
{ NULL, 0},
};
- int unsync;
- unsigned int tsize;
- unsigned int pos;
-
if ((size < 16) ||
- (data[0] != 0x49) ||
- (data[1] != 0x44) ||
- (data[2] != 0x33) ||
- (data[3] != 0x02) ||
- (data[4] != 0x00)) {
-
- //g_print ("no id3v2 tags info (stage 1)\n");
+ (data[0] != 0x49) ||
+ (data[1] != 0x44) ||
+ (data[2] != 0x33) ||
+ (data[3] != 0x02) ||
+ (data[4] != 0x00)) {
return;
}
unsync = (data[5] & 0x80) > 0;
tsize = (((data[6] & 0x7F) << 21) |
- ((data[7] & 0x7F) << 14) |
- ((data[8] & 0x7F) << 07) |
- ((data[9] & 0x7F) << 00));
+ ((data[7] & 0x7F) << 14) |
+ ((data[8] & 0x7F) << 07) |
+ ((data[9] & 0x7F) << 00));
if (tsize + 10 > size) {
-
- //g_print ("no id3v2 tags info (stage 2)\n");
return;
}
pos = 10;
+
while (pos < tsize) {
size_t csize;
- int i;
+ gint i;
if (pos + 6 > tsize) {
-
- //g_print ("no id3v2 tags info (stage 3)\n");
return;
}
- csize = (data[pos+3] << 16) + (data[pos+4] << 8) + data[pos+5];
+ csize = (data[pos+3] << 16) + (data[pos + 4] << 8) + data[pos + 5];
if ((pos + 6 + csize > tsize) ||
- (csize > tsize) ||
- (csize == 0))
+ (csize > tsize) ||
+ (csize == 0)) {
break;
-
+ }
i = 0;
- while (tmap[i].text != NULL) {
-
- if (0 == strncmp(tmap[i].text,
- (const char*) &data[pos],
- 3)) {
- char * word;
-
- /* 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 = get_utf8((const char*) &data[pos+7],
- csize,
- NULL, NULL, NULL);
+ while (tmap[i].text != NULL) {
+ if (strncmp(tmap[i].text, (const char*) &data[pos], 3) == 0) {
+ gchar * word;
+
+ /* 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 = get_utf8 ((const gchar*) &data[pos + 7],
+ csize,
+ NULL, NULL, NULL);
break;
- default:
- /* bad encoding byte,
- try to convert from iso-8859-1 */
- word = get_utf8((const char*) &data[pos+7],
- csize,
- NULL, NULL, NULL);
- break;
+ default:
+ /* Bad encoding byte, try to
+ * convert from iso-8859-1.
+ */
+ word = get_utf8 ((const gchar*) &data[pos + 7],
+ csize,
+ NULL, NULL, NULL);
+ break;
}
+
pos++;
csize--;
- if ((word != NULL) && (strlen(word) > 0)) {
-
+ if (word != NULL && strlen(word) > 0) {
if (strcmp (tmap[i].text, "COM") == 0) {
- char *s = g_strdup (word+strlen(word)+1);
+ gchar *s;
+
+ s = g_strdup (word + strlen(word) + 1);
g_free (word);
word = s;
}
- g_hash_table_insert (metadata, g_strdup (tmap[i].type), g_strdup (word));
+ g_hash_table_insert (metadata,
+ g_strdup (tmap[i].type),
+ g_strdup (word));
} else {
g_free (word);
}
break;
}
+
i++;
}
+
pos += 6 + csize;
}
-
-
}
static void
-tracker_extract_mp3 (const char *filename, GHashTable *metadata)
+extract_mp3 (const gchar *filename,
+ GHashTable *metadata)
{
- int file;
- void * buffer;
- struct stat fstatbuf;
- size_t size;
- id3tag info;
+ gint file;
+ void *buffer;
+ struct stat fstatbuf;
+ size_t size;
+ id3tag info;
info.title = NULL;
info.artist = NULL;
@@ -1000,113 +994,115 @@
file = g_open (filename, O_RDONLY, 0);
#endif
- if ((file == -1) || (stat (filename, &fstatbuf) == -1)) {
- close(file);
- //g_print ("could not open file %s\n", filename);
+ if (file == -1 || stat (filename, &fstatbuf) == -1) {
+ close (file);
return;
}
size = fstatbuf.st_size;
if (size == 0) {
- //g_print ("could not stat file %s\n", filename);
- close(file);
+ close (file);
return;
}
-
- if (size > MAX_FILE_READ) {
+ if (size > MAX_FILE_READ) {
size = MAX_FILE_READ;
}
- //g_print ("file size is %d\n", size);
#ifndef OS_WIN32
buffer = mmap (NULL, size, PROT_READ, MAP_PRIVATE, file, 0);
#endif
- if ((buffer == NULL) || (buffer == (void *) -1)) {
- //g_print ("mmap failure\n");
+ if (buffer == NULL || buffer == (void*) -1) {
close(file);
return;
}
-
-
- if (get_id3 (buffer, size, &info) != 0) {
- //g_print ("no id3 info detected\n");
+ if (!get_id3 (buffer, size, &info)) {
+ /* Do nothing? */
}
-
if (info.title && strlen (info.title) > 0) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Title"), g_strdup (info.title));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Title"),
+ g_strdup (info.title));
}
if (info.artist && strlen (info.artist) > 0) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Artist"), g_strdup (info.artist));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Artist"),
+ g_strdup (info.artist));
}
if (info.album && strlen (info.album) > 0) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Album"), g_strdup (info.album));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Album"),
+ g_strdup (info.album));
}
if (info.year && strlen (info.year) > 0) {
- g_hash_table_insert (metadata, g_strdup ("Audio:ReleaseDate"), g_strdup (info.year));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:ReleaseDate"),
+ g_strdup (info.year));
}
if (info.genre && strlen (info.genre) > 0) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Genre"), g_strdup (info.genre));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Genre"),
+ g_strdup (info.genre));
}
if (info.comment && strlen (info.comment) > 0) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Comment"), g_strdup (info.comment));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Comment"),
+ g_strdup (info.comment));
}
- free(info.title);
- free(info.year);
- free(info.album);
- free(info.artist);
- free(info.comment);
-
- /* get other embedded tags */
+ free (info.title);
+ free (info.year);
+ free (info.album);
+ free (info.artist);
+ free (info.comment);
+ /* Get other embedded tags */
get_id3v2_tags (buffer, size, metadata);
get_id3v23_tags (buffer, size, metadata);
get_id3v24_tags (buffer, size, metadata);
- /* get mp3 stream info */
+ /* Get mp3 stream info */
mp3_parse (buffer, size, metadata);
/* Check that we have the minimum data. FIXME We should not need to do this */
if (!g_hash_table_lookup (metadata, "Audio:Title")) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Title"), g_strdup ("tracker:unknown"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Title"),
+ g_strdup ("tracker:unknown"));
}
if (!g_hash_table_lookup (metadata, "Audio:Album")) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Album"), g_strdup ("tracker:unknown"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Album"),
+ g_strdup ("tracker:unknown"));
}
if (!g_hash_table_lookup (metadata, "Audio:Artist")) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Artist"), g_strdup ("tracker:unknown"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Artist"),
+ g_strdup ("tracker:unknown"));
}
if (!g_hash_table_lookup (metadata, "Audio:Genre")) {
- g_hash_table_insert (metadata, g_strdup ("Audio:Genre"), g_strdup ("tracker:unknown"));
+ g_hash_table_insert (metadata,
+ g_strdup ("Audio:Genre"),
+ g_strdup ("tracker:unknown"));
}
#ifndef OS_WIN32
munmap (buffer, size);
#endif
close(file);
-
}
-
-TrackerExtractorData data[] = {
- { "audio/mpeg", tracker_extract_mp3 },
- { "audio/x-mp3", tracker_extract_mp3 },
- { NULL, NULL }
-};
-
-
TrackerExtractorData *
tracker_get_extractor_data (void)
{
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]
