[FFmpeg-cvslog] r25385 - trunk/libavformat/mpegenc.c

Michael Niedermayer michaelni
Thu Oct 7 21:40:39 CEST 2010


On Thu, Oct 07, 2010 at 08:58:48PM +0200, Reimar D?ffinger wrote:
> On Thu, Oct 07, 2010 at 12:24:15PM +0200, Michael Niedermayer wrote:
> > Its important that people have fun contributing to the project and to me
> > this is not fun. Do people want to string up the leader (or for the matter of
> > fact anyone) in rules that make it not fun for him to work?
> 
> I don't that is the intention at all. But I and I imagine quite a
> few others don't see why adding a few whitespaces would kill the fun.
> Now if it is so that you don't mind trying to do so if in exchange
> nobody will bother you about it except a remainder if you seem
> to have forgotten it completely or something like that I'd be
> fine with that.
> I don't mind it enough that I wouldn't be willing to compromise
> wildly on this, but I do not understand what your issue is with
> this (note: this is meant as a statement of fact, not a criticism).

you should look at some of my handwritten text or how things ive written
back in the day when i still was going to school looked or just look at
my spelling errors and punctuation or lack thereof or capitalization.
I am _bad_ at things like this and for me it is effort to get things
capitalized punctuated and so on correctly, yes i can surely make my code
compliant to indent -kr but it _does_ cost time and it is _not_ fun for me.

To me and most people i knew 5 years ago K&R meant {} placement, here now
things start with someone making the source consistent in some cosmetic aspect
and then expecting everyone to remember just one additional rule to keep it
consistent, this is piling rules up over time.
And at the same time people complain that the patch review process is
too full of bikeshed nonsesne already and others have their own forks and
dont submit changes because its too much hassle to go through patch review.

Also some of the whitespace placement actually reduces readability like in
equations a*b + c*d vs. a * b + c * d
or ive seen switch case reformattings that did nothing but undo sane human
thought out formatting

Ultimately the issue i have with the whole is, i dont mind at all if
someone wants to make things consistent in some way but that doesnt
implicate that its ok to burden everyone from maintainer to the
small one time contributor with all these consistency rules
It wasnt done in the past and we did had contributors that felt strong on
cosmetic issues, be that the AAC and AC3 variables or odded shimons coding
style. If we force consistency at the whitespace level we surely should
force variables to be understandable english words. But had we done that
we would be missing some AAC and AC3 code now. I really dont think that
this would have been better

Anyway i will try to add some whitespace to be closer to indent -kr but
ill only do it as far as it doesnt annoy me too much and i very strongly
oppose that anyones patch is rejected because of such whitespace issues.
that is just because i try to follow it does not mean i expect others to
follow it. I consider it a stupid idea still


[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

it is not once nor twice but times without number that the same ideas make
their appearance in the world. -- Aristotle
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-cvslog/attachments/20101007/e8360186/attachment.pgp>



More information about the ffmpeg-cvslog mailing list