On Monday, November 05, 2018 05:31:36 PM Ondrej Novy wrote: > Hi, > > po 5. 11. 2018 v 9:09 odesílatel Thomas Goirand <z...@debian.org> napsal: > > Do other fellow DD also think it's kind of ok to keep 3.7 in Buster? > > +1 for keeping 3.7 in Buster and switch defaults to 3.7.
I recently got burned by changes in the way re.sub works in python3.7 (see [1] and search for re.sub, [2], and [3]). From the whatsnew: > re.sub() now replaces empty matches adjacent to a previous non-empty match. > For example re.sub('x*', '-', 'abxd') returns now '-a-b--d-' instead of > '-a-b-d-' (the first minus between ‘b’ and ‘d’ replaces ‘x’, and the second > minus replaces an empty string between ‘x’ and ‘d’). This change broke canonicalization in dkimpy. Once I knew about it, I fixed it, but the default 'import foo' autopkgtests we have on so many modules won't find this (I don't think, it didn't in my case). I only found out about it due to an upstream bug report on OS X. No one in Debian (or Ubuntu) reported it. Does anyone have an idea about how common problems from this change might be? That particular package uses re.sub 8 times and only one use was problematic. In any case, I don't have anything other than a vague concern based on one of my packages that there are more hidden problems. Scott K [1] https://docs.python.org/3/whatsnew/3.7.html [2] https://bugs.python.org/issue25054 [3] https://bugs.python.org/issue32308