[FFmpeg-devel] [PATCH] avformat/hls: remove unused function

[Hendrik Leppkes]

On Sun, Sep 27, 2015 at 12:32 PM, compn <tempn at mi.rr.com> wrote:
> On Sat, 26 Sep 2015 19:20:57 +0200
> Hendrik Leppkes <h.leppkes at gmail.com> wrote:
>
>> On Sat, Sep 26, 2015 at 7:00 PM, wm4 <nfxjfg at googlemail.com> wrote:
>> > On Sat, 26 Sep 2015 12:53:58 -0400
>> > Ganesh Ajjanagadde <gajjanag at mit.edu> wrote:
>> >
>> >> On Sat, Sep 26, 2015 at 12:37 PM, wm4 <nfxjfg at googlemail.com>
>> >> wrote:
>> >> > On Sat, 26 Sep 2015 12:12:44 -0400
>> >> > Ganesh Ajjanagadde <gajjanag at mit.edu> wrote:
>> >> >
>> >> >> On Sat, Sep 26, 2015 at 12:06 PM, wm4 <nfxjfg at googlemail.com>
>> >> >> wrote:
>> >> >> > On Sat, 26 Sep 2015 10:59:44 -0400
>> >> >> > Ganesh Ajjanagadde <gajjanag at mit.edu> wrote:
>> >> >> >
>> >> >> >> On Fri, Sep 18, 2015 at 11:16 AM, Ganesh Ajjanagadde
>> >> >> >> <gajjanag at mit.edu> wrote:
>> >> >> >> > On Fri, Sep 18, 2015 at 11:08 AM, wm4
>> >> >> >> > <nfxjfg at googlemail.com> wrote:
>> >> >> >> >> On 18.09.2015 15:09, Ganesh Ajjanagadde wrote:
>> >> >> >> >>>
>> >> >> >> >>> On Wed, Sep 16, 2015 at 8:26 PM, Ganesh Ajjanagadde
>> >> >> >> >>> <gajjanagadde at gmail.com> wrote:
>> >> >> >> >>>>
>> >> >> >> >>>> Fixes -Wunused-function from
>> >> >> >> >>>>
>> >> >> >> >>>> http://fate.ffmpeg.org/report.cgi?time=20150820031140&slot=arm64-darwin-clang-apple-5.1
>> >> >> >> >>>>
>> >> >> >> >>>> Signed-off-by: Ganesh Ajjanagadde
>> >> >> >> >>>> <gajjanagadde at gmail.com> ---
>> >> >> >> >>>>   libavformat/hls.c | 13 -------------
>> >> >> >> >>>>   1 file changed, 13 deletions(-)
>> >> >> >> >>>>
>> >> >> >> >>>> diff --git a/libavformat/hls.c b/libavformat/hls.c
>> >> >> >> >>>> index c16c770..2ea3a22 100644
>> >> >> >> >>>> --- a/libavformat/hls.c
>> >> >> >> >>>> +++ b/libavformat/hls.c
>> >> >> >> >>>> @@ -495,19 +495,6 @@ static int
>> >> >> >> >>>> ensure_playlist(HLSContext *c, struct playlist **pls,
>> >> >> >> >>>> const char *url return 0;
>> >> >> >> >>>>   }
>> >> >> >> >>>>
>> >> >> >> >>>> -static int open_in(HLSContext *c, AVIOContext **in,
>> >> >> >> >>>> const char *url) -{
>> >> >> >> >>>> -    AVDictionary *tmp = NULL;
>> >> >> >> >>>> -    int ret;
>> >> >> >> >>>> -
>> >> >> >> >>>> -    av_dict_copy(&tmp, c->avio_opts, 0);
>> >> >> >> >>>> -
>> >> >> >> >>>> -    ret = avio_open2(in, url, AVIO_FLAG_READ,
>> >> >> >> >>>> c->interrupt_callback, &tmp);
>> >> >> >> >>>> -
>> >> >> >> >>>> -    av_dict_free(&tmp);
>> >> >> >> >>>> -    return ret;
>> >> >> >> >>>> -}
>> >> >> >> >>>> -
>> >> >> >> >>>>   static int url_connect(struct playlist *pls,
>> >> >> >> >>>> AVDictionary *opts, AVDictionary *opts2)
>> >> >> >> >>>>   {
>> >> >> >> >>>>       AVDictionary *tmp = NULL;
>> >> >> >> >>>> --
>> >> >> >> >>>> 2.5.2
>> >> >> >> >>>>
>> >> >> >> >>>
>> >> >> >> >>> ping; wm4 basically did not like my old patch: see
>> >> >> >> >>> https://ffmpeg.org/pipermail/ffmpeg-devel/2015-August/177769.html
>> >> >> >> >>> for his thoughts and my response. Here, I delete as
>> >> >> >> >>> opposed to comment out the lines. I personally do not
>> >> >> >> >>> care about deleting vs commenting out; but one of the
>> >> >> >> >>> two should be used.
>> >> >> >> >>
>> >> >> >> >>
>> >> >> >> >> If you think the inactive code is worth keeping, put it
>> >> >> >> >> into a git branch or tag the commit removing the code.t
>> >> >> >> >
>> >> >> >> > I do not have write access, so I will leave it up to you
>> >> >> >> > (or anyone else pushing) to either apply the old patch or
>> >> >> >> > this one (and possibly tag it if desired).
>> >> >> >>
>> >> >> >> See above for two possiblities on resolving this: either
>> >> >> >> this patch (optionally tagging it), or simply commenting out
>> >> >> >> the function (see above for link to old patch).
>> >> >> >> Ping.
>> >> >> >
>> >> >> > Just get rid of this. I had to deal with the HLS code lately,
>> >> >> > and it was utterly confusing that there was this unused
>> >> >> > function.
>> >> >> >
>> >> >> > Also, this function is so trivial, that anyone really needing
>> >> >> > it could rewrite it in under a minute, and it'd probably be
>> >> >> > an improvement too. Even discussing keeping this is a waste
>> >> >> > of energy.
>> >> >>
>> >> >> Please push this patch then; you have write access and I don't.
>> >> >
>> >> > Send your SSH key to Michael Niedermayer. I'll apply this patch
>> >> > in a moment.
>> >>
>> >> Michael offered write access a month back, to which I replied
>> >> saying that I do not trust myself with it at the moment and would
>> >> like to wait a year. I also said that I would reconsider if it is
>> >> leading to wasted maintainer time. Our back and forth here
>> >> suggests that there is some wastage.
>> >>
>> >> If no one objects to me getting write access, I will send it to
>> >> Michael. Please note that the intention right now is not for any
>> >> "maintainership" duties, but simply to ease the application of
>> >> reviewed patches.
>> >
>> > Yes, it's much better for everyone if you just push your own
>> > patches, after they have been OK'ed.
>>
>> Personally I find giving new people push access after only 3 months of
>> activity slightly premature.
>
> so if they have been around for 6 months its better?
>

Maybe, maybe not. There is no written rule for the time. I just feel
like someone with push access should gather a bit of seniority in the
project first, just to get a proper grasp of how it runs and what all
the "unwritten rules" you refer to actually are.

- Hendrik