We need to make a decision wrt distribution names (second try)
Obviously, we are having a debate about what forms distribution names can take. I think we need a decision. Does anyone know if there are existing rules for distribution names? I can't find them if there are. Up until now, I think we've been in somewhat of a prototyping mode, but I think it's time to move beyond that. I strongly suggest that we need an official specification that says: - what's a legal distribution name and - what the equivalence rules for distribution names are. Whatever we decide needs to be well supported by setuptools and PyPI. I can live with whatever we decide as long as we decide something and make sure it is well communicated and implemented. In particular, I could live with the equivalence rules that setuptools uses if they are documented and if they are supported correctly and efficiently by the index (including mirrors). IMO, a decision is extremely important. If we can't reach consensus, then we need to call in the BDFL. Jim -- Jim Fulton mailto:jim@zope.com Python Powered! CTO (540) 361-1714 http://www.python.org Zope Corporation http://www.zope.com http://www.zope.org
Hi,
Obviously, we are having a debate about what forms distribution names can take. I think we need a decision.
Thanks for bringing this up.
Does anyone know if there are existing rules for distribution names? I can't find them if there are. Up until now, I think we've been in somewhat of a prototyping mode, but I think it's time to move beyond that.
I strongly suggest that we need an official specification that says:
- what's a legal distribution name and
- what the equivalence rules for distribution names are.
Comparison rules are also important: Is artin-1.2-rc2 < artin-1.2? IMO, it's perfectly fine to just state: comparisons are lexicographical (ASCII only). But I'd like to see this mentioned somewhere. - Lars
On Jul 24, 2007, at 1:28 PM, Lars Immisch wrote:
Hi,
Obviously, we are having a debate about what forms distribution names can take. I think we need a decision.
Thanks for bringing this up.
Does anyone know if there are existing rules for distribution names? I can't find them if there are. Up until now, I think we've been in somewhat of a prototyping mode, but I think it's time to move beyond that. I strongly suggest that we need an official specification that says: - what's a legal distribution name and - what the equivalence rules for distribution names are.
Comparison rules are also important:
Is artin-1.2-rc2 < artin-1.2?
Note that these are not distribution names. Well, that depends on how you define "distribution names". Sigh. The dsitribution names I'm trying to talk about don't have version numbers. I don't see a particular reason why these distribution names have to be ordered, Jim -- Jim Fulton mailto:jim@zope.com Python Powered! CTO (540) 361-1714 http://www.python.org Zope Corporation http://www.zope.com http://www.zope.org
At 01:33 PM 7/24/2007 -0400, Jim Fulton wrote:
Note that these are not distribution names. Well, that depends on how you define "distribution names". Sigh. The dsitribution names I'm trying to talk about don't have version numbers.
Setuptools uses the term "project name" for what you're calling a distribution name, if that helps any. :)
On Jul 24, 2007, at 1:48 PM, Phillip J. Eby wrote:
At 01:33 PM 7/24/2007 -0400, Jim Fulton wrote:
Note that these are not distribution names. Well, that depends on how you define "distribution names". Sigh. The dsitribution names I'm trying to talk about don't have version numbers.
Setuptools uses the term "project name" for what you're calling a distribution name, if that helps any. :)
Right. I'm happy to use that. Does anyone want to disagree? BTW, to up the ante, I volunteer to try to update the distutils document to reflect what we decide. Jim -- Jim Fulton mailto:jim@zope.com Python Powered! CTO (540) 361-1714 http://www.python.org Zope Corporation http://www.zope.com http://www.zope.org
Hi,
Obviously, we are having a debate about what forms distribution names can take. I think we need a decision.
Thanks for bringing this up.
Does anyone know if there are existing rules for distribution names? I can't find them if there are. Up until now, I think we've been in somewhat of a prototyping mode, but I think it's time to move beyond that. I strongly suggest that we need an official specification that says: - what's a legal distribution name and - what the equivalence rules for distribution names are.
Comparison rules are also important:
Is artin-1.2-rc2 < artin-1.2?
Note that these are not distribution names. Well, that depends on how you define "distribution names". Sigh. The dsitribution names I'm trying to talk about don't have version numbers. I don't see a particular reason why these distribution names have to be ordered,
I see. Sorry for the drive-by-shooting. Still, I'd like a stated convention how version numbers are compared. I believe this would be good for setuptools also. But the issue is separable from project naming conventions. - Lars
On Jul 24, 2007, at 2:11 PM, Lars Immisch wrote:
Hi,
Obviously, we are having a debate about what forms distribution names can take. I think we need a decision.
Thanks for bringing this up.
Does anyone know if there are existing rules for distribution names? I can't find them if there are. Up until now, I think we've been in somewhat of a prototyping mode, but I think it's time to move beyond that. I strongly suggest that we need an official specification that says: - what's a legal distribution name and - what the equivalence rules for distribution names are.
Comparison rules are also important:
Is artin-1.2-rc2 < artin-1.2? Note that these are not distribution names. Well, that depends on how you define "distribution names". Sigh. The dsitribution names I'm trying to talk about don't have version numbers. I don't see a particular reason why these distribution names have to be ordered,
I see. Sorry for the drive-by-shooting.
np
Still, I'd like a stated convention how version numbers are compared. I believe this would be good for setuptools also.
setuptools has this. It would be nice bless it in a PEP.
But the issue is separable from project naming conventions.
Yup. Jim -- Jim Fulton mailto:jim@zope.com Python Powered! CTO (540) 361-1714 http://www.python.org Zope Corporation http://www.zope.com http://www.zope.org
On Wed, 25 Jul 2007, Lars Immisch wrote:
Still, I'd like a stated convention how version numbers are compared. I believe this would be good for setuptools also.
Currently PyPI sorts releases using distutils.version.LooseVersion It uses distutils.version.StrictVersion when parsing "provides, "requires" and "obsoletes" setup.py package meta-data. Richard
participants (4)
-
Jim Fulton -
Lars Immisch -
Phillip J. Eby -
Richard Jones