[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Packages preventing other packages from going into testing (fwd)



J.H.M. Dassen (Ray) wrote:
>On Sun, Jun 15, 2003 at 18:44:12 +0200, Jacob Hallén wrote:
>> I have written a little Python program that analyses
>> http://ftp-master.debian.org/testing/update_excuses.html
>> to find out where the bottlenecks are for packages going into testing.
>> Essentially, it is a count of packages that directly and indirectly depend 
on 
>> a package with at least one RC bug before they can go into testing.
>
>Maybe I'm missing something, but it looks like you're reinventing
>        http://bjorn.haxx.se/debian/

Yes, you are missing a small but important thing. Björns program is designed 
to answer the question "Why is package X not in testing?". It does this very 
well and has lots of nifty features I didn't put into mine.

My program answers the question "Where will an extra effort produce drastic 
results?" I'm sure I could have modified Björns program to yield this output, 
but I don't read and write Perl programs for pleasure.

I think it is important to have a tool that answers my question and does so 
in a terse form that is pruned of all irrelevant data. 

If Testing is much closer to resleasable than it is now, it will be so much 
easier to make decisions about what to do with the few packages that are not 
ready for a release and so much easier to make a time plan for when to 
release.
I know that releases are made when "everything is ready", but this still 
means that some packages have to be dropped from the release and you have to 
make an extra effort for some time (bug squash parties etc) before a release 
is possible. Having luked on the Debian lists for many years, I think the 
next release will be more of a challenge than any of the earlier ones. Too 
few of the developers are really focused on what it takes to make a release. 
If there isn't change, I'm not sure there will ever be another stable release.

Jacob Hallén




Reply to: