After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 371390 - HelloWorld Error
HelloWorld Error
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: dont know
0.10.10
Other All
: Normal major
: 0.10.11
Assigned To: GStreamer Maintainers
GStreamer Maintainers
: 372885 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2006-11-06 06:27 UTC by David Wan
Modified: 2006-11-30 22:15 UTC
See Also:
GNOME target: ---
GNOME version: 2.1/2.2


Attachments
log0 (857.82 KB, text/plain)
2006-11-14 09:16 UTC, David Wan
Details
log1 (831.50 KB, text/plain)
2006-11-14 09:20 UTC, David Wan
Details
log2 (824.57 KB, text/plain)
2006-11-14 09:22 UTC, David Wan
Details
the source file (3.40 KB, text/plain)
2006-11-14 12:35 UTC, David Wan
Details

Description David Wan 2006-11-06 06:27:57 UTC
Please describe the problem:
Hi, Group, 
I got the example Helloworld from http://gstreamer.freedesktop.org/data/doc/gstreamer/head/manual/html/chapter-helloworld.html#section-helloworld
and made it compiled on linux-2.4.20 and gcc-3.2.2-5, but when I run it as
./helloworld ../../purelove.mp3

an error comes out as:

pipeline        =       0x08068060
source  =       0x080b6020
decoder =       0x080be0e0
conv    =       0x080c3018
sink    =       0x080c8010
Setting to PLAYING
Running
[Invalid UTF-8] Error: \xce\xde\xb7\xa8\xbd\xe2\xc2\xeb\xc1\xf7\xa1\xa3
Returned, stopping playback
Deleting pipeline

why? 

Steps to reproduce:
1. make helloworld compiled on the same enviroment
2. run it
3. 


Actual results:


Expected results:


Does this happen every time?
yes

Other information:
Comment 1 Tim-Philipp Müller 2006-11-09 16:59:11 UTC
*** Bug 372885 has been marked as a duplicate of this bug. ***
Comment 2 Tim-Philipp Müller 2006-11-09 22:21:17 UTC
That hello world example will only work for Ogg/Vorbis audio files. It will not work with mp3 files as it is.

If you try to play an mp3 file with the example, you will get the following error:

 tpm@flap:~/tmp/helloworld$ ./hello ../../foo.mp3
 Setting to PLAYING
 Running
 Error: Could not demultiplex stream.
 Returned, stopping playback
 Deleting pipeline

That is to be expected. I don't know why you're not getting a proper error message  though, it could be either a garbage pointer or some locale mixup, hard to say like this.

Could you run your example like this:

  $ export GST_DEBUG=*:5
  $ ./helloworld ../../purelove.mp3  2>dbg.log

and then attach dbg.log to this bug report please?
Comment 3 David Wan 2006-11-14 09:16:06 UTC
Created attachment 76548 [details]
log0

the first part of the log
Comment 4 David Wan 2006-11-14 09:20:29 UTC
Created attachment 76550 [details]
log1

second part
Comment 5 David Wan 2006-11-14 09:22:31 UTC
Created attachment 76551 [details]
log2

third part
Comment 6 David Wan 2006-11-14 09:35:49 UTC
I have tried it with an ogg file like as :
  ./oggplayer ../../01givemeasign.ogg

Setting to PLAYING
Running
Dynamic pad created, linking parser/decoder
Error: Internal data stream error.
Returned, stopping playback
Deleting pipeline


and after set GST_DEBUG=5, I got an error log file, but It's too large, so I
splitted it into three attached above. Please check.

By a metion, all above are tested on VMware. Does it have any impact?
Comment 7 Tim-Philipp Müller 2006-11-14 12:16:29 UTC
Things should work fine in VMWare, that should not be a problem.

Could you please attach the _exact_ version of the source code that you used to produce the log file?

In the log file
  - there is evidence of someone trying to link filesrc to vorbisdec (!)
  - there is no evidence of anyone every trying to link
    vorbisdec <=>  audioconvert <=> osssink

These two observations do not match the source code in the example, so it looks like you fiddled with the source code and got a not-linked flow return because you removed the code that does the linking. This is to be expected.

FWIW, there should probably also be an audioresample element in front of the osssink, but your Ogg/Vorbis file seems to have a sample rate of 44100 which is likely to be accepted by osssink by defaul in any case.

Comment 8 David Wan 2006-11-14 12:34:27 UTC
  In this case, the oggplayer.c is just changed from helloworld case got from http://gstreamer.freedesktop.org/data/doc/gstreamer/head/manual/html/chapter-helloworld.html. 
  I just change its name. Nothing else has been changed,either.
Comment 9 David Wan 2006-11-14 12:35:30 UTC
Created attachment 76560 [details]
the source file
Comment 10 Tim-Philipp Müller 2006-11-14 12:47:06 UTC
If I click the link above, I see this

  gst_element_link (source, parser);
  gst_element_link_many (decoder, conv, sink, NULL);

in the code, and not this (as in your attachement):

  gst_element_link (source, parser);
  gst_element_link_many (source, decoder, conv, sink, NULL);

Maybe you're looking at an outdated version of the documentation or the program?

Could you try to change the above line in the source code and see if it makes a difference?
Comment 11 David Wan 2006-11-15 05:59:52 UTC
I'm sorry that some changes have been made by me carelessly. But, I can't hear nothing except 'papapa...' after correction yet. what's matter? The log is too large. 
Comment 12 David Wan 2006-11-15 06:20:26 UTC
It becomes okay after osssink has been changed to esdsink. But I could hear 'do....' with them to test audiotestsrc.