On Feb 15, 2004, at 1:06 PM, Remi Mommsen wrote:
How many *packages* is it? (stick to ONE tree)
Why? If these packages need to be changed, you have to fix/test 149 of them. Some might be easy as they are identical in all 4 trees, others have maybe 4 different versions. IMO it doesn't matter what percentage of the total packages this is. (I guess you will be happy to walk 50 miles as these is only 1% of 5000 miles, isn't it?)
Er, no, not really. When debating the rarity of large patch files, we really don't care if one package happens to be in the stable tree as well as the unstable tree. The question is 'how common is it that packages have large patch files, and what should the limit be?' At the 30k level the answer is 'not really very common at all'.
There aren't that many from 20k-30k either. If anything, the limit should be lowered to that range, not raised.
My problem is, that you came up with an arbitrary number (30kB) without any discussion. There is no clear guidance on how to proceed for those packages already in cvs. Does this new limit apply only to new packages? Who is going to make the changes? I guess we will have to switch back to the 'old' naming scheme for patches stored on a server to make them unique.
We set a guideline for the future, you should fix your packages to follow it. Eventually, if someone feels like it, they can go through the current packages and fix those as well or bug those people to do so.
If you want your patch uploaded, you need to talk to fink-core.
-Ben
------------------------------------------------------- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click _______________________________________________ Fink-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/fink-devel
