Harald Milz wrote:
Wow - HTTP live streaming to gmplayer works again. 1.3.27 killed it completely.
The only changes in 1.3.27 that could in any way influence this are
- Removed the NPTL check at startup, since several users have reported that VDR now runs fine with NPTL. - Disabled cVideoRepacker in remux.c, because it has caused several problems during recording. If you want to test (and maybe debug) it, activate the line
//#define TEST_cVideoRepacker
in remux.c.
The first one allows VDR to run on systems that use NPTL (there were several reports that this finally works now), while the second one just disabled the cVideoRepacker, since it has caused some problems in 1.3.26.
You can easily test 1.3.27 with either of these changes reverted.
Klaus
Klaus Schmidinger Klaus.Schmidinger@cadsoft.de wrote:
- Removed the NPTL check at startup, since several users have reported that VDR now runs fine with NPTL.
Could anyone give some insight on which kernel/glibc/other libs combo they are running?
osdpip-0.0.8 w/ streamdev-cvs works fine with 1.3.26 ...
Harald Milz wrote:
Klaus Schmidinger Klaus.Schmidinger@cadsoft.de wrote:
- Removed the NPTL check at startup, since several users have reported that VDR now runs fine with NPTL.
Could anyone give some insight on which kernel/glibc/other libs combo they are running?
osdpip-0.0.8 w/ streamdev-cvs works fine with 1.3.26 ...
You should consider testing the heavily patched streamdev-0.3.3-pre3-geni maintained by Peter Weber.
http://mitglied.lycos.de/peterweber69/streamdev_for_VDR-1.3.23.tar.gz
This version above, along with this change works perfectly with vdr-1.3.27:
--- streamdev/server/connectionVTP.c.orig 2005-05-29 12:20:40.000000000 -0700 +++ streamdev/server/connectionVTP.c 2005-05-29 12:25:30.000000000 -0700 @@ -369,8 +369,8 @@ if (isnumber(Option)) { cRecording *recording = Recordings.Get(strtol(Option, NULL, 10) - 1); if (recording) { - if (recording->Summary()) { - char *summary = strdup(recording->Summary()); + if (recording->Info()->Description()) { + char *summary = strdup(recording->Info()->Description()); Reply(250, "%s", strreplace(summary,'\n','|')); free(summary); }
The last time I tested cvs of streamdev, the internal demuxer was broken (although it may be fixed by now). IIRC, streamdev (cvs) has moved away from the use of the internal demuxer and piped it to an external one which does not work as well.
Regards, C.
"If you give someone a program, you will frustrate them for a day; if you teach them how to program, you will frustrate them for a lifetime"
C.Y.M syphir@syphir.sytes.net wrote:
You should consider testing the heavily patched streamdev-0.3.3-pre3-geni maintained by Peter Weber.
So - just another streamdev? The "official" version by Sascha which seems to have been abandoned, the cvs version by whom? And the one by Peter Weber? Is there also somebody who really feels _responsible_ for streamdev?
On Saturday 23 July 2005 18:38, Harald Milz wrote:
So - just another streamdev? The "official" version by Sascha which seems to have been abandoned, the cvs version by whom? And the one by Peter Weber? Is there also somebody who really feels _responsible_ for streamdev?
The CVS is maintained by me :-):-)
I'm sorry I don't have enough time currently to maintain everything as it should be, but... I'm still just one man. The current status of CVS is as follows:
Should work with 1.3.27, external remuxers are possible (Remux Type "Extern", script /root/externremux.sh, must read TS from stdin and output anything to stdout), the client menu (modifying server timers from a client et al) is completely deactivated (as there have been too many changes to such code in the core of 1.3.x) and there is one known problem when switching channels from a client.
I'll get to all of this, and I already have some untested code that's not checked-in yet, but currently I'm afraid I just don't have enough time :-(.
Greetings, Sascha
-
C.Y.M -
Harald Milz -
Klaus Schmidinger -
Sascha Volkenandt