[Twisted-Python] Release Procedure, Patches and Bugs, oh my!
We have about 20 items in the TODO file, we have loads of patches, and I have a suspicion we are not reporting bugs well enough because it's too frustrating. I suggest using some kind of bug tracking system. I suggest roundup, but please don't fight about this -- anything is better then nothing. Thanks. -- The Official Moshe Zadka FAQ: http://moshez.geek The Official Moshe Zadka FAQ For Dummies: http://moshez.org Read the FAQ
Moshe Zadka wrote:
I agree! I'm just not sure how to add CGIs, or I'd add it. Plus it needs a dbm module, and bsddb3 is *probably* better than python's bsddb (based off bsddb 1.8), so maybe someone with root should install it (http://pybsddb.sf.net). Or we could use SF... But roundup looks prettier, and will probably be faster. But we need *something*, and we need it soon. The patches are: 1. DNS resolver used for lookups (moshez) 2. multiple accept() (moshez) 3. New API for adding delayeds and shutdowns (itamar) 4. --python and --plugin for mktap (moshez) And I have a few local patches and I'm sure other people have too.
On Mon, Sep 24, 2001 at 12:54:35AM +0200, Itamar Shtull-Trauring wrote:
Agreed. Roundup sounds like a good place to start, especially since I have very little bug tracking system experience.
I agree! I'm just not sure how to add CGIs, or I'd add it.
Make yourself a --personal --static webserver, and you can run CGIs.
Can we use dirdbm or would that explode too quickly? I'd rather have stable than fast, and I've been burned by all varieties of bsddb at one point or another...
While I don't disagree that we need a system for tracking submitted patches, I think I've missed out on this corner of open source culture :). Why are we submitting patches to the mailing list? I think that if we agree that something should be done (and it won't break anything) rather than submitting patches to the list, commit it to CVS. I think that all the patches here with the possible exception of the 4th are cool new functionality, and if there's something that turns out not to fit, we can remove or revert it. I'm not against doing things through patches but I don't understand what value it has... (Why do I have this weird feeling that somebody's going to point to a mailing list message I don't remember where I recommend that everyone submit patches to the mailing list first...? -_-) -- ______ __ __ _____ _ _ | ____ | \_/ |_____] |_____| |_____| |_____ | | | | @ t w i s t e d m a t r i x . c o m http://twistedmatrix.com/users/glyph
On Sun, 23 Sep 2001, Glyph Lefkowitz <glyph@twistedmatrix.com> wrote:
I, for one, sometime like to send out a patch before I commit to get some feedback. For example, I can send out half-baked code to flow an idea past you. When we discuss it partly on the mailing list and partly on IRC, there is no place someone can go to see all discussion about a patch. What's more, since most patches are appliable (more or less) against the last release, this gives non-developer an opportunity to comment. I don't think we have a clear guideline, so my guideline is "things I'm the authority for I just submit. Things which are simple bugfixes or patches I ask around, then submit. Changes or big things I ask on the mailing list first". For example, the resolver patch needs some more discussion: Itamar had valid (though wrong, IMHO) objections to it and we talked about a reasonable compromise. -- The Official Moshe Zadka FAQ: http://moshez.geek The Official Moshe Zadka FAQ For Dummies: http://moshez.org Read the FAQ
Moshe Zadka wrote:
I agree! I'm just not sure how to add CGIs, or I'd add it. Plus it needs a dbm module, and bsddb3 is *probably* better than python's bsddb (based off bsddb 1.8), so maybe someone with root should install it (http://pybsddb.sf.net). Or we could use SF... But roundup looks prettier, and will probably be faster. But we need *something*, and we need it soon. The patches are: 1. DNS resolver used for lookups (moshez) 2. multiple accept() (moshez) 3. New API for adding delayeds and shutdowns (itamar) 4. --python and --plugin for mktap (moshez) And I have a few local patches and I'm sure other people have too.
On Mon, Sep 24, 2001 at 12:54:35AM +0200, Itamar Shtull-Trauring wrote:
Agreed. Roundup sounds like a good place to start, especially since I have very little bug tracking system experience.
I agree! I'm just not sure how to add CGIs, or I'd add it.
Make yourself a --personal --static webserver, and you can run CGIs.
Can we use dirdbm or would that explode too quickly? I'd rather have stable than fast, and I've been burned by all varieties of bsddb at one point or another...
While I don't disagree that we need a system for tracking submitted patches, I think I've missed out on this corner of open source culture :). Why are we submitting patches to the mailing list? I think that if we agree that something should be done (and it won't break anything) rather than submitting patches to the list, commit it to CVS. I think that all the patches here with the possible exception of the 4th are cool new functionality, and if there's something that turns out not to fit, we can remove or revert it. I'm not against doing things through patches but I don't understand what value it has... (Why do I have this weird feeling that somebody's going to point to a mailing list message I don't remember where I recommend that everyone submit patches to the mailing list first...? -_-) -- ______ __ __ _____ _ _ | ____ | \_/ |_____] |_____| |_____| |_____ | | | | @ t w i s t e d m a t r i x . c o m http://twistedmatrix.com/users/glyph
On Sun, 23 Sep 2001, Glyph Lefkowitz <glyph@twistedmatrix.com> wrote:
I, for one, sometime like to send out a patch before I commit to get some feedback. For example, I can send out half-baked code to flow an idea past you. When we discuss it partly on the mailing list and partly on IRC, there is no place someone can go to see all discussion about a patch. What's more, since most patches are appliable (more or less) against the last release, this gives non-developer an opportunity to comment. I don't think we have a clear guideline, so my guideline is "things I'm the authority for I just submit. Things which are simple bugfixes or patches I ask around, then submit. Changes or big things I ask on the mailing list first". For example, the resolver patch needs some more discussion: Itamar had valid (though wrong, IMHO) objections to it and we talked about a reasonable compromise. -- The Official Moshe Zadka FAQ: http://moshez.geek The Official Moshe Zadka FAQ For Dummies: http://moshez.org Read the FAQ
participants (3)
-
Glyph Lefkowitz -
Itamar Shtull-Trauring -
Moshe Zadka