Date: Sun, 06 Feb 2011 12:55:48 -0200 From: Raphael Kubo da Costa <kubito@gmail.com> To: freebsd-ports@freebsd.org Subject: Re: Call for testers: projectM comes to FreeBSD! [re ports/152823] Message-ID: <87zkq91abv.fsf@gmail.com> References: <AANLkTimtuCRzdk4MjWH2KaQTHHPsYqenq94f3Eiu%2BUdW@mail.gmail.com> <8762syroh6.fsf@gmail.com> <AANLkTimYQ4wPGL2WzNY9wA33eSt9R=hfHkLeDwer5H5V@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Chris Rees <utisoft@gmail.com> writes:
> and the authors are _dreadful_ at making releases. I have a feeling
> this may be the last.
:(
Then there's not much to do indeed...
>> + @${REINPLACE_CMD} -e '/return NULL/d' ${WRKSRC}/actor_projectM.cpp
>>
>> Is this one from upstream? If so, isn't it better to put it in files/?
>
> No, this is one I made. They've written a dummy function in, but it's
> type struct * and tries to return NULL.
Hmm, if it's a function with a return type other than void, it should
return at least something.
> I have a serious hatred of files/patch-*; they break with most
> updates, they bloat the repository and make it slower for everyone to
> csup as well. I thought that a ${REINPLACE_CMD} was a cleaner way of
> doing so.
To be honest (note I'm not a ports committer, so there might be reasons
for this), I don't personally like the current way patches are added:
it's better than when they were just called patch-aa and nobody would
ever be able to know what their purpose was (the CVS log normally
wouldn't help much), however it's still hard to tell when something was
obtained from upstream or is something local, if upstream has been
contacted about the patch or what exactly it is supposed to fix. Adding
some notes about this in the file before the diff itself would certainly
help a lot.
However, I feel this is still better than sed'ing in the Makefile is
worse than that -- you are effectively patching the source in a way that
is harder to separate than the building of the port itself, and since it
is still a form of "patching", it will break the same way it would if it
was in files/.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?87zkq91abv.fsf>