[Distutils] semver (again)

Tarek Ziadé tarek at ziade.org
Sun Oct 21 16:14:25 CEST 2012


On 10/21/12 3:17 PM, Vinay Sajip wrote:
> Tarek Ziadé <tarek <at> ziade.org> writes:
>
>> Have you tried to process failed version in PEP 386 with
>> suggest_normalized_version() ?
>>
>> that usually raises the % considerably
>>
> Fair point. Including the results for "proposed" meaning normalized + suggested,
> I got the following:
>
> Packages processed: 24891
> Packages with no versions: 217
> Packages with versions: 24674
> Number of packages clean for all schemes: 19010 (77%)
> Number of packages clean for normalized: 21072 (85%)
> Number of packages clean for proposed: 23685 (96%)

yeah that's close to what I remembered. and the 4% left are non-sense 
versions

that said  suggest_normalized_version() would probably work well with 
semver as well

> Number of packages clean for legacy: 24674 (100%, by you would expect)
> Number of packages clean for semantic: 19278 (78%)
>
> Regards,
>
> Vinay Sajip
>
> _______________________________________________
> Distutils-SIG maillist  -  Distutils-SIG at python.org
> http://mail.python.org/mailman/listinfo/distutils-sig



More information about the Distutils-SIG mailing list