[Zope-dev] Re: bad "zope.size" to remove from PyPI

[Chris McDonough]

On Aug 4, 2008, at 12:37 AM, Aaron Lehmann wrote:

>
> On Aug 2, 2008, at 11:45 AM, Chris Withers wrote:
>
>> Benji York wrote:
>>>> In case anybody's wondering how this complies with our "no  
>>>> removal of any
>>>> release whatsoever" policy [1], be assured that a 3.4dev-r73090  
>>>> thing isn't
>>>> a release by our standards. This version number not only  
>>>> contains the 'dev'
>>>> marker, meaning it must have come from a development branch  
>>>> (possibly the
>>>> trunk), it also contains the -rXXX suffix meaning it was made  
>>>> right from a
>>>> subversion checkout without having created a tags first (why  
>>>> else would you
>>>> want to include the revision number).
>>> Still, it's likely that someone was using it and their buildouts  
>>> are now
>>> broken.  We should have instead generated a proper release with a  
>>> higher
>>> version number and left the dev release alone.
>>
>> This is silly.
>>
>> Mistakes happen. Buildout and/or setuptools should be tolerant of  
>> accidental releases that are then removed from PyPI.
>>
>> What currently happens in cases like this?
>
>
> If the buildout is nailed to that version or above, and there is  
> none, it breaks.  Worse, if someone now adds another egg of the  
> same version, but consumers have cached a version, their buildout  
> won't download it, because it will already have that version in  
> cache.  I realize that this particular compound error is unlikely  
> to happen in this instance, but the principle holds.
>
> Yes, mistakes happen.  What Benji is saying is that deletion is not  
> the right way to remedy them, as unintuitive as that may seem.

I do agree that easy_install behavior is pretty insane wrt preferring  
dev releases over stable "old" releases.  But I think it's pretty  
reasonable to delete accidental releases, whatever the behavior of  
setuptools/easy_install/buildout.

In the first case you mention where "buildout is nailed to that  
version or above", that would have to be in error, as the release  
shouldn't have been made in the first place.  That also seems true of  
the second case, but I suppose the difference is the offended person  
is more innocent there: they should likely contact the person who  
released it and ask them to not do that again.

- C