[FFmpeg-devel] [PATCH] List enabled code in configure output

[Ramiro Polla]

Ramiro Polla escreveu:
> Ramiro Polla wrote:
>   
>> Rich Felker wrote:
>>     
>>> On Wed, Sep 26, 2007 at 04:39:26PM -0300, Ramiro Polla wrote:
>>>       
>>>>>>> The ${space:0:$length}
>>>>>>>               
>>>>>> Replaced with a while loop that adds one space at a time.
>>>>>>             
>>>>> I'm not having anything of the kind.  What's this patch good for
>>>>> anyway?
>>>>>           
>>>> As has already been discussed before in [1], some people would like for 
>>>> configure to print out which codecs and formats have been selected. I 
>>>> would like to have this feature, and Michael said he likes it too.
>>>>
>>>> I didn't like the fact that it would output one really long unreadable 
>>>> line. I also wouldn't like if it was one entry per line, since that 
>>>> would make configure's output really, really long. I suggest something 
>>>> in between: using show_list, print the output tabulated with 8 spaces. 
>>>> But I didn't find a way to make the code portable and elegant at the 
>>>> same time.
>>>>
>>>> show_list_1.diff makes show_list much faster, so I think it should be 
>>>> applied independent of the rest of this discussion.
>>>>
>>>> These are 3 possibilities of implementing listing the selected output:
>>>> show_list_2.diff prints one entry per line (long output).
>>>> show_list_3.diff prints them sequentially (long lines).
>>>> And another option is tabulating the output. Would you be against this 
>>>> kind of output, even if we managed to get the source code to be portable 
>>>> and clean?
>>>>         
>>> This kind of thing can probably be done easily with sed...
>>>       
>> I failed to find a simple way to do it that doesn't involve a loop, 
>> either in the script or in the sed command through labels, or a function 
>> that returns one up to 8 spaces, which is also ugly.
>>
>> Of course, there's no point in me trying further if the tabulated output 
>> is not accepted, although I find it nicer to look at.
>>
>> Mans, we can't (always) guess your preference unless you speak up. 
>> What's your opinion?
>> 1. All entries in same line
>> 2. One per line
>> 3. Tabulated output
>> 4. I'll cook up something I like better later
>> 5. None of the above and no selected features will ever be printed out
>>     
>
> I'll apply show_list_1.diff tomorrow if no one objects.
>
> I'll also apply show_list_3.diff (option 1) tomorrow if no one objects.
>   

Both applied.