mirrors/gstreamer
mirrors/gstreamer
1
1
Fork 0
mirror of https://gitlab.freedesktop.org/gstreamer/gstreamer.git synced 2025-02-17 03:35:21 +00:00
Code Issues 1 Projects Releases Wiki Activity

matroskademux: UTF-8 subtitles may have markup

Browse source 
Fixes #616936.
This commit is contained in:
Mark Nauwelaerts 2010-06-23 11:12:00 +02:00 committed by Sebastian Dröge
parent 38b596337f
commit 74e0c05ff7
2 changed files with 61 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

64
gst/matroska/matroska-demux.c
View file

@ -112,8 +112,8 @@ static GstStaticPadTemplate subtitle_src_templ =
GST_STATIC_PAD_TEMPLATE ("subtitle_%02d",
GST_PAD_SRC,
GST_PAD_SOMETIMES,
GST_STATIC_CAPS ("text/plain; application/x-ssa; application/x-ass; "
"application/x-usf; video/x-dvd-subpicture; "
GST_STATIC_CAPS ("text/x-pango-markup; application/x-ssa; "
"application/x-ass;application/x-usf; video/x-dvd-subpicture; "
"subpicture/x-pgs; subtitle/x-kate; " "application/x-subtitle-unknown")
);
@ -3797,6 +3797,38 @@ gst_matroska_demux_add_wvpk_header (GstElement * element,
return GST_FLOW_OK;
}
/* @text must be null-terminated */
static gboolean
gst_matroska_demux_subtitle_chunk_has_tag (GstElement * element,
const gchar * text)
{
gchar *tag;
/* yes, this might all lead to false positives ... */
tag = (gchar *) text;
while ((tag = strchr (tag, '<'))) {
tag++;
if (*tag != '\0' && *(tag + 1) == '>') {
/* some common convenience ones */
/* maybe any character will do here ? */
switch (*tag) {
case 'b':
case 'i':
case 'u':
case 's':
return TRUE;
default:
return FALSE;
}
}
}
if (strstr (text, "<span"))
return TRUE;
return FALSE;
}
static GstFlowReturn
gst_matroska_demux_check_subtitle_buffer (GstElement * element,
GstMatroskaTrackContext * stream, GstBuffer ** buf)
@ -3815,7 +3847,7 @@ gst_matroska_demux_check_subtitle_buffer (GstElement * element,
if (!sub_stream->invalid_utf8) {
if (g_utf8_validate (data, size, NULL)) {
return GST_FLOW_OK;
goto next;
}
GST_WARNING_OBJECT (element, "subtitle stream %d is not valid UTF-8, this "
"is broken according to the matroska specification", stream->num);
@ -3862,6 +3894,29 @@ gst_matroska_demux_check_subtitle_buffer (GstElement * element,
gst_buffer_unref (*buf);
*buf = newbuf;
data = (const gchar *) GST_BUFFER_DATA (*buf);
size = GST_BUFFER_SIZE (*buf);
next:
/* caps claim markup text, so we need to escape text,
* except if text is already markup and then needs no further escaping */
sub_stream->seen_markup_tag = sub_stream->seen_markup_tag ||
gst_matroska_demux_subtitle_chunk_has_tag (element, data);
if (!sub_stream->seen_markup_tag) {
utf8 = g_markup_escape_text (data, size);
newbuf = gst_buffer_new ();
GST_BUFFER_MALLOCDATA (newbuf) = (guint8 *) utf8;
GST_BUFFER_DATA (newbuf) = (guint8 *) utf8;
GST_BUFFER_SIZE (newbuf) = strlen (utf8);
gst_buffer_copy_metadata (newbuf, *buf,
GST_BUFFER_COPY_TIMESTAMPS | GST_BUFFER_COPY_FLAGS);
gst_buffer_unref (*buf);
*buf = newbuf;
}
return GST_FLOW_OK;
}
@ -6249,7 +6304,8 @@ gst_matroska_demux_subtitle_caps (GstMatroskaTrackSubtitleContext *
/* TODO: Add GST_MATROSKA_CODEC_ID_SUBTITLE_BMP support
* Check if we have to do something with codec_private */
if (!strcmp (codec_id, GST_MATROSKA_CODEC_ID_SUBTITLE_UTF8)) {
caps = gst_caps_new_simple ("text/plain", NULL);
/* well, plain text simply does not have a lot of markup ... */
caps = gst_caps_new_simple ("text/x-pango-markup", NULL);
context->postprocess_frame = gst_matroska_demux_check_subtitle_buffer;
} else if (!strcmp (codec_id, GST_MATROSKA_CODEC_ID_SUBTITLE_SSA)) {
caps = gst_caps_new_simple ("application/x-ssa", NULL);

1
gst/matroska/matroska-ids.h
View file

@ -567,6 +567,7 @@ typedef struct _GstMatroskaTrackSubtitleContext {
gboolean check_utf8; /* buffers should be valid UTF-8 */
gboolean invalid_utf8; /* work around broken files */
gboolean seen_markup_tag; /* markup found in text */
} GstMatroskaTrackSubtitleContext;
typedef struct _GstMatroskaIndex {