GNOME Bugzilla – Bug 754006
validate: media-descriptor-writer: Don't create media info when stream info is not present.
Last modified: 2015-11-17 09:12:23 UTC
When a file does not contain any stream info, then there is no need to create the media info file as, it is not considered to be a valid file and no validate checks are done for the same. This skips unnecessary files like .txt, .dump files
Created attachment 309902 [details] [review] skip media info creation when streaminfo is not present
ping :)
Review of attachment 309902 [details] [review]: You should fail hard in that case. Skipping the files is OK but in the case we are verifying that a transcoding happened properly we should fail in that function.
(In reply to Thibault Saunier from comment #3) > Review of attachment 309902 [details] [review] [review]: > > You should fail hard in that case. Skipping the files is OK but in the case > we are verifying that a transcoding happened properly we should fail in that > function. i am a little bit confused here.. What do you mean by fail hard? When stream info is not present, then i am not creating *.media_info file at all. So none of the validate tests including transcoding will/should not be run for these files
(In reply to Vineeth from comment #4) > i am a little bit confused here.. What do you mean by fail hard? > When stream info is not present, then i am not creating *.media_info file at > all. > So none of the validate tests including transcoding will/should not be run > for these files We are using that same code path after a transcoding is doone to check the resulting transcoded file, we should make sure to notify the caller that the discovery failed on that new file.
(In reply to Thibault Saunier from comment #5) > (In reply to Vineeth from comment #4) > > i am a little bit confused here.. What do you mean by fail hard? > > When stream info is not present, then i am not creating *.media_info file at > > all. > > So none of the validate tests including transcoding will/should not be run > > for these files > > We are using that same code path after a transcoding is doone to check the > resulting transcoded file, we should make sure to notify the caller that the > discovery failed on that new file. I searched the whole code and could not find any place where this is being referenced. This piece of code is used while creating *.media_info file before starting the actual tests. You mean after transcoding is done, we create media_info files for the transcoded media file and check the result of the transcoded media file with the media_info file? can you please point me to the location where this is being done?
Here: https://phabricator.freedesktop.org/diffusion/GSTDEV/browse/master/validate/launcher/baseclasses.py;0f7105d317328808272e554a9a138b5c8b47dc68$805
ohhh yeah... that is already being handled.. When there is no streaminfo, then gst_media_descriptor_writer_new_discover will return NULL. so media-check has this logic if (writer == NULL) { g_print ("Could not discover file: %s\n", argv[1]); ret = 1; goto out; } so the media-check-1.0 will return 1 for failure cases. For failure cases of transcode, the below error will be printed. [1 / 2] validate.file.transcode.to_mp3_and_h264_in_mp4.small_ogv: Failed (Could not discover encoded file file:///home/vineethtm/gst-validate/rendered/validate/file/to_mp3_and_h264_in_mp4/small_ogv) Guess this is what you are talking about right?
Ah, oh indeed, sorry for the noise then :)
Review of attachment 309902 [details] [review]: OK