<div dir="ltr"><div style>Ok, I found a possible hugly fix to timestamping problem! Note that I'm working with theora video, not with audio like vorbis in my wrong mail object.<br></div><div style><br></div><div style>
Actually I make a query_position on pipeline just before unlink dynamic bin, then send a new_segment on the queue src pad with offset=query_position so the resulting file dump is the following:</div><div style><br></div><div style>
<div>$ oggz-dump video02.ogv |grep pos|head -5</div><div>00:00:00.000: serialno 1209677529, granulepos 0|0, packetno 0 *** bos: 42 bytes</div><div>00:00:00.000: serialno 1209677529, calc. gpos 0|0, packetno 1: 58 bytes</div>
<div>00:00:00.000: serialno 1209677529, granulepos 0|0, packetno 2: 2.552 kB</div><div>00:00:00.375: serialno 1209677529, granulepos 4|0, packetno 3: 34.163 kB</div><div>00:00:00.500: serialno 1209677529, granulepos 4|1, packetno 4: 3.213 kB</div>
<div><br></div><div style>Packet 3 which I think is the first video packet driving the timeline now start with a gap of 0.375 seconds and it is always the same in every following videoN.ogv file. I don't know if this gap could be fixed totally, but until now this is a good workaround and it's ok for my project.</div>
<div style><br></div><div style>Anyway this solution is working on my old test machine with gstreamer 0.10.28 but on my target machine with gstreamer-sdk 2012.11 the function query_position always fail without further info, so new mail request is coming :)</div>
<div style><br></div><div style>Feel free to reply with suggestion or further consideration about current duration workaround.</div><div style><br></div><div style>Cheers </div></div></div><div class="gmail_extra"><br><br>
<div class="gmail_quote">On Sat, May 25, 2013 at 6:08 PM, Tim-Philipp Müller <span dir="ltr"><<a href="mailto:t.i.m@zen.co.uk" target="_blank">t.i.m@zen.co.uk</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On Fri, 2013-05-24 at 09:40 +0200, lestoilfante wrote:<br>
<br>
Hi,<br>
<div class="im"><br>
> I tried oggz-dump on my files and the output is a bit different from<br>
> that on radio.ogg you suggest where gpos start from xxxxx and continue<br>
> to grow:<br>
</div>> (...)<br>
<div class="im">> #oggz-dump video03.ogg |grep gpos|head pos<br>
><br>
> 00:00:00.000: serialno 1850673644, granulepos 0|0, packetno 0 *** bos<br>
</div><div class="im">> 00:00:00.000: serialno 1850673644, calc. gpos 0|0, packetno 1<br>
</div><div class="im">> 00:00:00.000: serialno 1850673644, granulepos 0|0, packetno 2<br>
</div><div class="im">> 00:00:31.600: serialno 1850673644, granulepos 159|0, packetno 3<br>
</div><div class="im">> 00:00:31.800: serialno 1850673644, granulepos 159|1, packetno 4<br>
><br>
><br>
</div><div class="im">> So I have only one gpos for each file and a lot of granulepos with<br>
> timing start at 0 and jumping to 15 sec on second file and so on. I'm<br>
> not so much confident with ogg but why first packets on video02 and<br>
> video03 start from 00:00:00.000 and jump to 00:00:15.000 and<br>
> 00:00:31.600?<br>
<br>
</div>This is to be expected, I think. The first few packets are header<br>
packets, which are supposed to have a granulepos of 0 according to<br>
<a href="http://xiph.org/vorbis/doc/Vorbis_I_spec.html#x1-130000A.2" target="_blank">http://xiph.org/vorbis/doc/Vorbis_I_spec.html#x1-130000A.2</a> (at least for<br>
vorbis).<br>
<div class="HOEnZb"><div class="h5"><br>
Cheers<br>
-Tim<br>
<br>
<br>
_______________________________________________<br>
gstreamer-devel mailing list<br>
<a href="mailto:gstreamer-devel@lists.freedesktop.org">gstreamer-devel@lists.freedesktop.org</a><br>
<a href="http://lists.freedesktop.org/mailman/listinfo/gstreamer-devel" target="_blank">http://lists.freedesktop.org/mailman/listinfo/gstreamer-devel</a><br>
</div></div></blockquote></div><br></div>