lavf: TED Talks JSON captions demuxer.

16001373 · Nicolas George · 18eb3196 · 16001373 · 16001373 · 16001373
Commit 16001373 authored Jun 05, 2012 by Nicolas George
8 changed files
--- a/Changelog
+++ b/Changelog
@@ -39,6 +39,7 @@ version <next>:
 - Paris Audio File demuxer
 - Virtual concatenation demuxer
 - VobSub demuxer
+- JSON captions for TED talks decoding support


 version 1.0:

--- a/doc/demuxers.texi
+++ b/doc/demuxers.texi
@@ -215,4 +215,25 @@ backslash or single quotes.

 @end table

+@section tedcaptions
+
+JSON captions used for @url{http://www.ted.com/, TED Talks}.
+
+TED does not provide links to the captions, but they can be guessed from the
+page. The file @file{tools/bookmarklets.html} from the FFmpeg source tree
+contains a bookmarklet to expose them.
+
+This demuxer accepts the following option:
+@table @option
+@item start_time
+Set the start time of the TED talk, in milliseconds. The default is 15000
+(15s). It is used to sync the captions with the downloadable videos, because
+they include a 15s intro.
+@end table
+
+Example: convert the captions to a format most players understand:
+@example
+ffmpeg -i http://www.ted.com/talks/subtitles/id/1/lang/en talk1-en.srt
+@end example
+
 @c man end INPUT DEVICES
--- a/doc/general.texi
+++ b/doc/general.texi
@@ -924,6 +924,7 @@ performance on systems without hardware floating point support).
 @item SAMI             @tab   @tab X @tab   @tab X
 @item SubRip (SRT)     @tab X @tab X @tab X @tab X
 @item SubViewer        @tab   @tab X @tab   @tab X
+@item TED Talks captions @tab @tab X @tab   @tab X
 @item VobSub (IDX+SUB) @tab   @tab X @tab   @tab X
 @item 3GPP Timed Text  @tab   @tab   @tab X @tab X
 @item WebVTT           @tab   @tab X @tab   @tab X

--- a/libavformat/Makefile
+++ b/libavformat/Makefile
@@ -346,6 +346,7 @@ OBJS-$(CONFIG_SUBVIEWER_DEMUXER)         += subviewerdec.o
 OBJS-$(CONFIG_SWF_DEMUXER)               += swfdec.o swf.o
 OBJS-$(CONFIG_SWF_MUXER)                 += swfenc.o swf.o
 OBJS-$(CONFIG_TAK_DEMUXER)               += takdec.o apetag.o img2.o rawdec.o
+OBJS-$(CONFIG_TEDCAPTIONS_DEMUXER)       += tedcaptionsdec.o
 OBJS-$(CONFIG_THP_DEMUXER)               += thp.o
 OBJS-$(CONFIG_TIERTEXSEQ_DEMUXER)        += tiertexseq.o
 OBJS-$(CONFIG_MKVTIMESTAMP_V2_MUXER)     += mkvtimestamp_v2.o

--- a/libavformat/allformats.c
+++ b/libavformat/allformats.c
@@ -244,6 +244,7 @@ void av_register_all(void)
    REGISTER_DEMUXER  (SUBVIEWER, subviewer);
    REGISTER_MUXDEMUX (SWF, swf);
    REGISTER_DEMUXER  (TAK, tak);
+    REGISTER_DEMUXER  (TEDCAPTIONS, tedcaptions);
    REGISTER_MUXER    (TG2, tg2);
    REGISTER_MUXER    (TGP, tgp);
    REGISTER_DEMUXER  (THP, thp);

--- a/libavformat/tedcaptionsdec.c
+++ b/libavformat/tedcaptionsdec.c
--- a/libavformat/version.h
+++ b/libavformat/version.h
@@ -30,7 +30,7 @@
 #include "libavutil/avutil.h"

 #define LIBAVFORMAT_VERSION_MAJOR 54
-#define LIBAVFORMAT_VERSION_MINOR 48
+#define LIBAVFORMAT_VERSION_MINOR 49
 #define LIBAVFORMAT_VERSION_MICRO 100

 #define LIBAVFORMAT_VERSION_INT AV_VERSION_INT(LIBAVFORMAT_VERSION_MAJOR, \

--- a/tools/bookmarklets.html
+++ b/tools/bookmarklets.html
+<!DOCTYPE html>
+<html>
+<head>
+<!--
+    This file is part of FFmpeg.
+
+    All scripts contained in this file can be considered public domain.
+  -->
+<title>FFmpeg bookmarklets</title>
+<meta charset="UTF-8">
+<script type="text/javascript">
+function convert(js) {
+  js = js.replace(/\/\*.*?\*\//g, ""); /* comments */
+  js = js.replace(/\s+/g, " ");
+  js = js.replace(/\s+\z/, "");
+  js = "(function(){" + js + "})();void 0";
+  return "javascript:" + escape(js);
+}
+function init() {
+  var pre = document.getElementsByTagName("pre");
+  for (var i = 0; pre.length > i; i++) {
+    document.getElementById(pre[i].id + "-link").href = convert(pre[i].textContent);
+  }
+}
+</script>
+<style type="text/css">
+pre { border: solid black 1px; padding: 0.2ex; font-size: 80% }
+</style>
+</head>
+<body onload="init()">
+
+<h1>Introduction</h1>
+
+The scripts in this page are
+<a href="http://en.wikipedia.org/wiki/Bookmarklet">bookmarklets</a>: store
+their link version in a bookmark, and later activate the bookmark on a page
+to run the script.
+
+<h1>TED Talks captions</h1>
+
+<p><a id="ted_talks_captions-link" href="#">Get links to the captions</a></p>
+
+<pre id="ted_talks_captions">
+d = window.open("", "sub", "width=256,height=512,resizable=yes,scrollbars=yes").document;
+l = document.getElementById("languageCode").getElementsByTagName("option");
+for (i = 1; i &lt; l.length ; i++) {
+  d.body.appendChild(p = d.createElement("p"));
+  p.appendChild(a = d.createElement("a"));
+  a.appendChild(d.createTextNode(l[i].textContent));
+  a.href="http://www.ted.com/talks/subtitles/id/" + talkID+"/lang/" + l[i].value;
+}
+</pre>
+
+</body>
+</html>