1 /* GStreamer AVI GAB2 subtitle parser
2 * Copyright (C) <2007> Thijs Vermeir <thijsvermeir@gmail.com>
3 * Copyright (C) <2007> Tim-Philipp Müller <tim centricular net>
4 *
5 * This library is free software; you can redistribute it and/or
6 * modify it under the terms of the GNU Library General Public
7 * License as published by the Free Software Foundation; either
8 * version 2 of the License, or (at your option) any later version.
9 *
10 * This library is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 * Library General Public License for more details.
14 *
15 * You should have received a copy of the GNU Library General Public
16 * License along with this library; if not, write to the
17 * Free Software Foundation, Inc., 51 Franklin St, Fifth Floor,
18 * Boston, MA 02110-1301, USA.
19 */
20
21 /**
22 * SECTION:element-avisubtitle
23 * @title: avisubtitle
24 *
25 * Parses the subtitle stream from an avi file.
26 *
27 * ## Example launch line
28 *
29 * |[
30 * gst-launch-1.0 filesrc location=subtitle.avi ! avidemux name=demux ! queue ! avisubtitle ! subparse ! textoverlay name=overlay ! videoconvert ! autovideosink demux. ! queue ! decodebin ! overlay.
31 * ]|
32 * This plays an avi file with a video and subtitle stream.
33 *
34 */
35
36 /* example of a subtitle chunk in an avi file
37 * 00000000: 47 41 42 32 00 02 00 10 00 00 00 45 00 6e 00 67 GAB2.......E.n.g
38 * 00000010: 00 6c 00 69 00 73 00 68 00 00 00 04 00 8e 00 00 .l.i.s.h........
39 * 00000020: 00 ef bb bf 31 0d 0a 30 30 3a 30 30 3a 30 30 2c ....1..00:00:00,
40 * 00000030: 31 30 30 20 2d 2d 3e 20 30 30 3a 30 30 3a 30 32 100 --> 00:00:02
41 * 00000040: 2c 30 30 30 0d 0a 3c 62 3e 41 6e 20 55 54 46 38 ,000..<b>An UTF8
42 * 00000050: 20 53 75 62 74 69 74 6c 65 20 77 69 74 68 20 42 Subtitle with B
43 * 00000060: 4f 4d 3c 2f 62 3e 0d 0a 0d 0a 32 0d 0a 30 30 3a OM</b>....2..00:
44 * 00000070: 30 30 3a 30 32 2c 31 30 30 20 2d 2d 3e 20 30 30 00:02,100 --> 00
45 * 00000080: 3a 30 30 3a 30 34 2c 30 30 30 0d 0a 53 6f 6d 65 :00:04,000..Some
46 * 00000090: 74 68 69 6e 67 20 6e 6f 6e 41 53 43 49 49 20 2d thing nonASCII -
47 * 000000a0: 20 c2 b5 c3 b6 c3 a4 c3 bc c3 9f 0d 0a 0d 0a ..............
48 */
49
50 #ifdef HAVE_CONFIG_H
51 #include "config.h"
52 #endif
53
54 #include <string.h>
55
56 #include "gstavielements.h"
57 #include "gstavisubtitle.h"
58
59 GST_DEBUG_CATEGORY_STATIC (avisubtitle_debug);
60 #define GST_CAT_DEFAULT avisubtitle_debug
61
62 static GstStaticPadTemplate sink_template = GST_STATIC_PAD_TEMPLATE ("sink",
63 GST_PAD_SINK,
64 GST_PAD_ALWAYS,
65 GST_STATIC_CAPS ("application/x-subtitle-avi")
66 );
67
68 static GstStaticPadTemplate src_template = GST_STATIC_PAD_TEMPLATE ("src",
69 GST_PAD_SRC,
70 GST_PAD_ALWAYS,
71 GST_STATIC_CAPS ("application/x-subtitle")
72 );
73
74 static void gst_avi_subtitle_title_tag (GstAviSubtitle * sub, gchar * title);
75 static GstFlowReturn gst_avi_subtitle_chain (GstPad * pad, GstObject * parent,
76 GstBuffer * buffer);
77 static GstStateChangeReturn gst_avi_subtitle_change_state (GstElement * element,
78 GstStateChange transition);
79 static gboolean gst_avi_subtitle_send_event (GstElement * element,
80 GstEvent * event);
81
82 #define gst_avi_subtitle_parent_class parent_class
83 G_DEFINE_TYPE (GstAviSubtitle, gst_avi_subtitle, GST_TYPE_ELEMENT);
84 GST_ELEMENT_REGISTER_DEFINE_WITH_CODE (avisubtitle, "avisubtitle",
85 GST_RANK_PRIMARY, GST_TYPE_AVI_SUBTITLE, avi_element_init (plugin));
86
87 #define IS_BOM_UTF8(data) ((GST_READ_UINT32_BE(data) >> 8) == 0xEFBBBF)
88 #define IS_BOM_UTF16_BE(data) (GST_READ_UINT16_BE(data) == 0xFEFF)
89 #define IS_BOM_UTF16_LE(data) (GST_READ_UINT16_LE(data) == 0xFEFF)
90 #define IS_BOM_UTF32_BE(data) (GST_READ_UINT32_BE(data) == 0xFEFF)
91 #define IS_BOM_UTF32_LE(data) (GST_READ_UINT32_LE(data) == 0xFEFF)
92
93 static GstBuffer *
gst_avi_subtitle_extract_file(GstAviSubtitle * sub,GstBuffer * buffer,guint offset,guint len)94 gst_avi_subtitle_extract_file (GstAviSubtitle * sub, GstBuffer * buffer,
95 guint offset, guint len)
96 {
97 const gchar *input_enc = NULL;
98 GstBuffer *ret = NULL;
99 gchar *data;
100 GstMapInfo map;
101
102 gst_buffer_map (buffer, &map, GST_MAP_READ);
103 data = (gchar *) (map.data + offset);
104
105 if (len >= (3 + 1) && IS_BOM_UTF8 (data) &&
106 g_utf8_validate (data + 3, len - 3, NULL)) {
107 ret =
108 gst_buffer_copy_region (buffer, GST_BUFFER_COPY_ALL, offset + 3,
109 len - 3);
110 } else if (len >= 2 && IS_BOM_UTF16_BE (data)) {
111 input_enc = "UTF-16BE";
112 data += 2;
113 len -= 2;
114 } else if (len >= 2 && IS_BOM_UTF16_LE (data)) {
115 input_enc = "UTF-16LE";
116 data += 2;
117 len -= 2;
118 } else if (len >= 4 && IS_BOM_UTF32_BE (data)) {
119 input_enc = "UTF-32BE";
120 data += 4;
121 len -= 4;
122 } else if (len >= 4 && IS_BOM_UTF32_LE (data)) {
123 input_enc = "UTF-32LE";
124 data += 4;
125 len -= 4;
126 } else if (g_utf8_validate (data, len, NULL)) {
127 /* not specified, check if it's UTF-8 */
128 ret = gst_buffer_copy_region (buffer, GST_BUFFER_COPY_ALL, offset, len);
129 } else {
130 /* we could fall back to gst_tag_freeform_to_utf8() here */
131 GST_WARNING_OBJECT (sub, "unspecified encoding, and not UTF-8");
132 ret = NULL;
133 goto done;
134 }
135
136 g_return_val_if_fail (ret != NULL || input_enc != NULL, NULL);
137
138 if (input_enc) {
139 GError *err = NULL;
140 gchar *utf8;
141 gsize slen;
142
143 GST_DEBUG_OBJECT (sub, "converting subtitles from %s to UTF-8", input_enc);
144 utf8 = g_convert (data, len, "UTF-8", input_enc, NULL, NULL, &err);
145
146 if (err != NULL) {
147 GST_WARNING_OBJECT (sub, "conversion to UTF-8 failed : %s", err->message);
148 g_error_free (err);
149 ret = NULL;
150 goto done;
151 }
152
153 ret = gst_buffer_new ();
154 slen = strlen (utf8);
155 gst_buffer_append_memory (ret,
156 gst_memory_new_wrapped (0, utf8, slen, 0, slen, utf8, g_free));
157
158 GST_BUFFER_OFFSET (ret) = 0;
159 }
160
161 done:
162 gst_buffer_unmap (buffer, &map);
163
164 return ret;
165 }
166
167 /**
168 * gst_avi_subtitle_title_tag:
169 * @sub: subtitle element
170 * @title: the title of this subtitle stream
171 *
172 * Send an event to the srcpad of the @sub element with the title
173 * of the subtitle stream as a GST_TAG_TITLE
174 */
175 static void
gst_avi_subtitle_title_tag(GstAviSubtitle * sub,gchar * title)176 gst_avi_subtitle_title_tag (GstAviSubtitle * sub, gchar * title)
177 {
178 gst_pad_push_event (sub->src,
179 gst_event_new_tag (gst_tag_list_new (GST_TAG_TITLE, title, NULL)));
180 }
181
182 static GstFlowReturn
gst_avi_subtitle_parse_gab2_chunk(GstAviSubtitle * sub,GstBuffer * buf)183 gst_avi_subtitle_parse_gab2_chunk (GstAviSubtitle * sub, GstBuffer * buf)
184 {
185 gchar *name_utf8;
186 guint name_length;
187 guint file_length;
188 GstMapInfo map;
189
190 gst_buffer_map (buf, &map, GST_MAP_READ);
191
192 /* check the magic word "GAB2\0", and the next word must be 2 */
193 if (map.size < 12 || memcmp (map.data, "GAB2\0\2\0", 5 + 2) != 0)
194 goto wrong_magic_word;
195
196 /* read 'name' of subtitle */
197 name_length = GST_READ_UINT32_LE (map.data + 5 + 2);
198 GST_LOG_OBJECT (sub, "length of name: %u", name_length);
199 if (map.size <= 17 + name_length)
200 goto wrong_name_length;
201
202 name_utf8 =
203 g_convert ((gchar *) map.data + 11, name_length, "UTF-8", "UTF-16LE",
204 NULL, NULL, NULL);
205
206 if (name_utf8) {
207 GST_LOG_OBJECT (sub, "subtitle name: %s", name_utf8);
208 gst_avi_subtitle_title_tag (sub, name_utf8);
209 g_free (name_utf8);
210 }
211
212 /* next word must be 4 */
213 if (GST_READ_UINT16_LE (map.data + 11 + name_length) != 0x4)
214 goto wrong_fixed_word_2;
215
216 file_length = GST_READ_UINT32_LE (map.data + 13 + name_length);
217 GST_LOG_OBJECT (sub, "length srt/ssa file: %u", file_length);
218
219 if (map.size < (17 + name_length + file_length))
220 goto wrong_total_length;
221
222 /* store this, so we can send it again after a seek; note that we shouldn't
223 * assume all the remaining data in the chunk is subtitle data, there may
224 * be padding at the end for some reason, so only parse file_length bytes */
225 sub->subfile =
226 gst_avi_subtitle_extract_file (sub, buf, 17 + name_length, file_length);
227
228 if (sub->subfile == NULL)
229 goto extract_failed;
230
231 gst_buffer_unmap (buf, &map);
232
233 return GST_FLOW_OK;
234
235 /* ERRORS */
236 wrong_magic_word:
237 {
238 GST_ELEMENT_ERROR (sub, STREAM, DECODE, (NULL), ("Wrong magic word"));
239 gst_buffer_unmap (buf, &map);
240 return GST_FLOW_ERROR;
241 }
242 wrong_name_length:
243 {
244 GST_ELEMENT_ERROR (sub, STREAM, DECODE, (NULL),
245 ("name doesn't fit in buffer (%" G_GSIZE_FORMAT " < %d)", map.size,
246 17 + name_length));
247 gst_buffer_unmap (buf, &map);
248 return GST_FLOW_ERROR;
249 }
250 wrong_fixed_word_2:
251 {
252 GST_ELEMENT_ERROR (sub, STREAM, DECODE, (NULL),
253 ("wrong fixed word: expected %u, got %u", 4,
254 GST_READ_UINT16_LE (map.data + 11 + name_length)));
255 gst_buffer_unmap (buf, &map);
256 return GST_FLOW_ERROR;
257 }
258 wrong_total_length:
259 {
260 GST_ELEMENT_ERROR (sub, STREAM, DECODE, (NULL),
261 ("buffer size is wrong: need %d bytes, have %" G_GSIZE_FORMAT " bytes",
262 17 + name_length + file_length, map.size));
263 gst_buffer_unmap (buf, &map);
264 return GST_FLOW_ERROR;
265 }
266 extract_failed:
267 {
268 GST_ELEMENT_ERROR (sub, STREAM, DECODE, (NULL),
269 ("could not extract subtitles"));
270 gst_buffer_unmap (buf, &map);
271 return GST_FLOW_ERROR;
272 }
273 }
274
275 static GstFlowReturn
gst_avi_subtitle_chain(GstPad * pad,GstObject * parent,GstBuffer * buffer)276 gst_avi_subtitle_chain (GstPad * pad, GstObject * parent, GstBuffer * buffer)
277 {
278 GstAviSubtitle *sub = GST_AVI_SUBTITLE (parent);
279 GstFlowReturn ret;
280
281 if (sub->subfile != NULL) {
282 GST_WARNING_OBJECT (sub, "Got more buffers than expected, dropping");
283 ret = GST_FLOW_EOS;
284 goto done;
285 }
286
287 /* we expect exactly one buffer with the whole srt/ssa file in it */
288 ret = gst_avi_subtitle_parse_gab2_chunk (sub, buffer);
289 if (ret != GST_FLOW_OK)
290 goto done;
291
292 /* now push the subtitle data downstream */
293 ret = gst_pad_push (sub->src, gst_buffer_ref (sub->subfile));
294
295 done:
296
297 gst_buffer_unref (buffer);
298 return ret;
299 }
300
301 static gboolean
gst_avi_subtitle_send_event(GstElement * element,GstEvent * event)302 gst_avi_subtitle_send_event (GstElement * element, GstEvent * event)
303 {
304 GstAviSubtitle *avisubtitle = GST_AVI_SUBTITLE (element);
305 gboolean ret = FALSE;
306
307 if (GST_EVENT_TYPE (event) == GST_EVENT_SEEK) {
308 if (avisubtitle->subfile) {
309 if (gst_pad_push (avisubtitle->src,
310 gst_buffer_ref (avisubtitle->subfile)) == GST_FLOW_OK)
311 ret = TRUE;
312 }
313 }
314 gst_event_unref (event);
315 return ret;
316 }
317
318 static void
gst_avi_subtitle_class_init(GstAviSubtitleClass * klass)319 gst_avi_subtitle_class_init (GstAviSubtitleClass * klass)
320 {
321 GstElementClass *gstelement_class = (GstElementClass *) klass;
322
323 GST_DEBUG_CATEGORY_INIT (avisubtitle_debug, "avisubtitle", 0,
324 "parse avi subtitle stream");
325
326 gstelement_class->change_state =
327 GST_DEBUG_FUNCPTR (gst_avi_subtitle_change_state);
328 gstelement_class->send_event =
329 GST_DEBUG_FUNCPTR (gst_avi_subtitle_send_event);
330
331 gst_element_class_add_static_pad_template (gstelement_class, &sink_template);
332 gst_element_class_add_static_pad_template (gstelement_class, &src_template);
333
334 gst_element_class_set_static_metadata (gstelement_class,
335 "Avi subtitle parser", "Codec/Parser/Subtitle",
336 "Parse avi subtitle stream", "Thijs Vermeir <thijsvermeir@gmail.com>");
337 }
338
339 static void
gst_avi_subtitle_init(GstAviSubtitle * self)340 gst_avi_subtitle_init (GstAviSubtitle * self)
341 {
342 GstCaps *caps;
343
344 self->src = gst_pad_new_from_static_template (&src_template, "src");
345 gst_element_add_pad (GST_ELEMENT (self), self->src);
346
347 self->sink = gst_pad_new_from_static_template (&sink_template, "sink");
348 gst_pad_set_chain_function (self->sink,
349 GST_DEBUG_FUNCPTR (gst_avi_subtitle_chain));
350
351 caps = gst_static_pad_template_get_caps (&src_template);
352 gst_pad_set_caps (self->src, caps);
353 gst_caps_unref (caps);
354
355 gst_pad_use_fixed_caps (self->src);
356 gst_element_add_pad (GST_ELEMENT (self), self->sink);
357
358 self->subfile = NULL;
359 }
360
361 static GstStateChangeReturn
gst_avi_subtitle_change_state(GstElement * element,GstStateChange transition)362 gst_avi_subtitle_change_state (GstElement * element, GstStateChange transition)
363 {
364 GstStateChangeReturn ret;
365 GstAviSubtitle *sub = GST_AVI_SUBTITLE (element);
366
367 switch (transition) {
368 case GST_STATE_CHANGE_NULL_TO_READY:
369 case GST_STATE_CHANGE_READY_TO_PAUSED:
370 default:
371 break;
372 }
373
374 ret = GST_ELEMENT_CLASS (parent_class)->change_state (element, transition);
375 if (ret == GST_STATE_CHANGE_FAILURE)
376 return ret;
377
378 switch (transition) {
379 case GST_STATE_CHANGE_PAUSED_TO_READY:
380 if (sub->subfile) {
381 gst_buffer_unref (sub->subfile);
382 sub->subfile = NULL;
383 }
384 break;
385 default:
386 break;
387 }
388
389 return ret;
390 }
391