Re: FW: Bug in File Meta Source: Repeat does not work for detached headers

[Marcus Müller]

Hi Adrian,

thanks for the report! Yes, if you can, please create a ticket! If you want, you could 
throw your recording and your playback flow graphs into a zip file and attach that to your 
issue text, that makes reproduction 100% reliable.

This is excellent bug reporting, by the way, thank you very much!

Just as perspective:

We have two other issues, https://github.com/gnuradio/gnuradio/issues/3516 and 
https://github.com/gnuradio/gnuradio/issues/5568 , concerning the metadata file sink. Both 
of them use the non-detached header, so they seem to be different from yours.

Best,
Marcus

On 13.03.24 13:27, Adrian Winter wrote:
> Hey,
>
> I'm pretty sure I discovered a bug in GNURadio and am now following the steps 
> to report the bug, starting with sending this mail to the mailing list 
> according to the development instructions.
>
> Short description:
>
> The File Meta Source block does not repeat the file, if a detached header is 
> used.
>
> Used Software:
>
>    *   GNURadio 3.10.9.2
>    *   Python 3.11.8
>    *   Installed through radioconda
>    *   X86-64 running Ubuntu 20.04 LTS
>
> Steps to reproduce:
>
>
>    1.  Create any IQ file using the File Meta Sink block with Detached Header being set 
> to "True"
>    2.  Replay this file with File Meta Source block, Detached Header = True and 
> Repeat = True, with the data being sent to any sink (e.g. Tag Debug)
>
> Observed Behaviour:
>
> The file's contents are being loaded, but when EOF is reached no more samples 
> are being produced, essentially stalling the flowgraph
>
> Expected Behaviour:
>
> The file should reset to the beginning and continue to produce samples 
> indefinitely.
>
> Other Notes:
>
>    *   Replaying the IQ file without the information in the header using File 
> Source and Repeat = True works as expected.
>    *   Replaying a file using the File Meta Source with Repeat = True, but 
> inline headers works as expected.
>    *   The data type does not appear to have an influence (same behaviour for 
> IShort and Complex Float)
>
> Any help would of course be appreciated, and please let me know if you can 
> reproduce the issue and if I should create a ticket in Github.
>
> Thank you in advance and kind regards!
> Adrian Winter
>
> Researcher,
> SINTEF Digital
> Trondheim, Norway
>
> PS: this is the third time I send this message in various states of being 
> subscribed to this mailing list. I really hope that the older two versions are 
> actually filtered and I'm not spamming the mailing list. If I accidentally did, 
> please accept my apologies - first time interacting with a mailing list.