[MPlayer-cvslog] r33516 - trunk/stream/stream_cue.c

[Reimar Döffinger]

On 29 May 2011, at 15:54, Ivan Kalvachev <ikalvachev at gmail.com> wrote:
>> Well, I wanted it split. And I had to understand it first.
>> So I just reimplemented it.
> 
> I didn't split the patch so the user could easily test it.

Yes, it was just one of the reasons I didn't use it directly.

> 
>> I find it really difficult to decide how to handle such a case,
>> where I consider to have implemented it myself but on the other
>> hand someone else has done the same thing before and I certainly
>> was influenced by it.
> 
> Always mention previous works.
> If you ask yourself if you should do it, that's a sign that you should.

Yes, I know in principle. I just didn't really care. Sorry, I too sometimes make the stupidest mistakes.

> 
>> And should it happen again please flame for not attributing correctly,
>> because your first message gave me a very wrong idea of what you
>> thought was wrong about what I did :-)
> 
> My first worry was that your commits completely broke my patch.

You couldn't know that so it was badly done, but I would of course have fixed your patch myself if it had been necessary.

> I kind of find it hard to image that you were not aware
> that you are doing something wrong.

Not enough to expect such a strong reaction.

> BTW, The only portion of my patch that is not yet committed is skipping
> cases 0 and 1 when bin_filename is missing.
> I could do (around line 212)
> "for (i=2*!i; ..." or "for(i=i?0:2; ..."

I found that quite ugly.

> or
> "if ( i<= 1 && bin_filename[0] == '\0') continue;"

That would fit with the other check there, however I intended to just leave that out, the IS_REG check does catch it so it seemed rather unnecessary/purely cosmetic to me.