On Tue, Aug 23, 2016 at 09:44:16AM -0700, Glyph Lefkowitz wrote: > > We can drop support for 3.3 of course, but that's a separate discussion.
I'm marginally in favor of it. I appreciate Adi's concern about the build infrastructure. And thanks to the inimitable Donald Stufft, it's easy to get a sense of how popular Python 3.3 is among Twisted users and whether or not dropping support for it will affect a large percentage of them. He provided the following BigTable Query to run against the PyPI downloads dataset (available at https://bigquery.cloud.google.com/dataset/the-psf:pypi?pli=1 ): SELECT REGEXP_EXTRACT(details.python, r"^([^\.]+\.[^\.]+)") as python_version, COUNT(*) as downloads, FROM TABLE_DATE_RANGE( [the-psf:pypi.downloads], DATE_ADD(CURRENT_TIMESTAMP(), -31, "day"), DATE_ADD(CURRENT_TIMESTAMP(), -1, "day") ) WHERE file.project = 'twisted' GROUP BY python_version ORDER BY downloads DESC Here're the results: python_version downloads 2.7 446869 null 21959 3.5 15204 3.4 11245 2.6 5060 3.3 1060 3.2 271 3.6 241 2.4 5 1.17 3 2.5 2 (Donald explained that `null` represents mirrors and caches like bandersnatch.) It appears that dropping Python 3.3 wouldn't exclude very many of our users. _______________________________________________ Twisted-Python mailing list Twisted-Python@twistedmatrix.com http://twistedmatrix.com/cgi-bin/mailman/listinfo/twisted-python
