Re: [alsa-devel] [PATCH 14/21] ASoC: soc-core: have legacy_dai_naming at snd_soc_register_dai()

[Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>]

On 10/10/19 8:07 PM, Kuninori Morimoto wrote:
> 
> Hi Pierre-Louis
> 
> Thank you for your feedback
> 
>>> snd_soc_register_dai() is now used from topology.
>>> But, to reduce duplicated code, it will be used from soc-core, too.
>>> To prepare for it, this patch adds missing parameter legacy_dai_naming
>>> to snd_soc_register_dai().
>>
>> It doesn't look like this series reduces the confusion between
>> snd_soc_register_dai() and snd_soc_register_dais() ?
>>
>> maybe for the latter case since it's a static function we don't want
>> the entire prefix then?
> 
> Maybe my explain is not so good...
> The point is that, in general people think like below from naming.
> Other functions are this style.
> 
> =>	int snd_soc_register_dai()
> 	{
> 		...
> 	}
> 
> 	int snd_soc_register_dais()
> 	{
> 		for(..) {
> =>			snd_soc_register_dai()
> 		}		
> 	}
> 
> But in reality is like this
> 
> 	int snd_soc_register_dai()
> 	{
> 		/* almost same but different code */
> 	}
> 
> 	snd_soc_register_dais()
> 	{
> 		/* almost same but different code */
> 	}
> 
> To avoid duplicate code and confusion,
> this patchset try to implement "general" style.
> But needs some preparation.
> 
> I will fix log and English.

I get your point, what I was trying to suggest is that we only use the 
full snd_soc prefix for non-static functions that can be called by other 
files, for static functions we can use just the soc prefix. It's a 
convention that helps the reader understand what is local and what is 
common/shared.

_______________________________________________
Alsa-devel mailing list
Alsa-devel@alsa-project.org
https://mailman.alsa-project.org/mailman/listinfo/alsa-devel