[asterisk-dev] [Code Review] Remove dead PLC code

[Russell Bryant]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/683/#review2135
-----------------------------------------------------------

Ship it!


- Russell


On 2010-06-02 12:00:43, Mark Michelson wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/683/
> -----------------------------------------------------------
> 
> (Updated 2010-06-02 12:00:43)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Summary
> -------
> 
> Recently, I made a change that made PLC more reliable. The result was that essentially everything that was PLC-related in translator code became irrelevant. This diff aims to remove the unnecessary code.
> 
> I kind of took the chainsaw approach here, just removing everything that is not used. If I should take a different approach with the ast_translator structure for ABI purposes, that can be done.
> 
> 
> Diffs
> -----
> 
>   /trunk/codecs/codec_adpcm.c 266924 
>   /trunk/codecs/codec_alaw.c 266924 
>   /trunk/codecs/codec_dahdi.c 266924 
>   /trunk/codecs/codec_g722.c 266924 
>   /trunk/codecs/codec_g726.c 266924 
>   /trunk/codecs/codec_gsm.c 266924 
>   /trunk/codecs/codec_lpc10.c 266924 
>   /trunk/codecs/codec_ulaw.c 266924 
>   /trunk/include/asterisk/translate.h 266924 
>   /trunk/main/translate.c 266924 
> 
> Diff: https://reviewboard.asterisk.org/r/683/diff
> 
> 
> Testing
> -------
> 
> I ran calls from one Asterisk box to another Asterisk box that is set to drop 5% of the RTP frames it writes. I used wireshark to ensure that when genericplc is set in codecs.conf that on the first Asterisk box we do fill in for the lost packets before sending frames out. In other words, PLC is still working.
> 
> I also ran the tests under valgrind and ensured that no new warnings were generated. In the interest of full disclosure, I did see valgrind warnings I had not seen before. However, these were clearly not related to my code changes and were related to a previous commit to trunk. These will be taken care of in a separate commit.
> 
> 
> Thanks,
> 
> Mark
> 
>