Generating DSMCC contents

[artemies]

Hi,

There are times that the DSMCC directory (and the associated contents) are not generated when I run demuxfs using same transport stream file. I run the demuxfs several times for the same file there are cases that I'm getting the objects in the DSMCC.

Thanks!

[lucasvr]

Hi. Which flags are you using when invoking demuxfs? Are you playing the file on loop? Also, are you interacting with the mount point while the file is being parsed? If so, could you let me know if the problem also happens if you wait until the transport stream has been completely read?

[artemies]

Hi,

I'm transmitting the ts file in a loop. Then, on the receiver I record it as a transport stream file for several minutes (more than the length of the video). Then once I'm done recording the ts file I just run demuxfs using the recorded ts file.

[artemies]

Hi,

At some instances I'm getting the DSMCC files correctly. But when trying to rerun the demuxfs, after creating a new mount point there are times that I'm not getting the DSMCC objects

[lucasvr]

That's interesting. The file parser should be deterministic, unless there are interactions with the mount point as the file is being parsed by demuxfs. Let me see if I can reproduce the problem using transport streams I already have at hand. Please expect a feedback by early next week.

[artemies]

Hi,

Thank you. I can share the transport stream file to you also. By the way, I'm running the demuxfs in a Raspberry Pi 3. will there be any problems running multiple instances of demuxfs for with different mount points?

[lucasvr]

There should be no problem running multiple instances of demuxfs, as long as the input comes from a file and not a tuner device. Let me try with the transport streams I already have, first. I'll let you know if I can reproduce the issue.

[lucasvr]

I was able to reproduce the issue. The reason for the disappearance of some DSM-CC applications in the TS that @Kitoko86 provided to me before is that the various applications have the same name (or no name), causing the first one(s) to be overwritten by the most recently application parsed. I'll work on a fix for this one soon.