Hi Bruno, Ronny, On Sat, Jun 24, 2017 at 14:51 +0000, Bruno Oliveira wrote: > Hi Ronny, > > Not sure about removing as it will break things that are working even if by > accident. I don't see any problem about deprecating it and warning about > not using it in new code.
... maybe simply documenting the new methods (write_bytes/write_text i guess) and not mentioning the old methods anymore but not removing/changing them. That has happened several times and i think it's a good approach. I suspect there is quite some code (in tests and outside of tests) that implicitely uses path.local.write so removing it would break a lot of things that currently work even with a messy api. holgr > Cheers > > On Sat, Jun 24, 2017 at 10:39 AM RonnyPfannschmidt < > [email protected]> wrote: > > > Hi everyone, > > > > after investigating the issue i pretty much had to deduce that a > > horrible mistake happened while doing the python3 port, and the write > > method was a complete and broken mess ever since working only by > > accident on ascii inputs > > > > any sane fix needs a break in api and so i propose deprecating/removing > > it as both write_text and write_bytes exist and always do the correct > > thing and the magic mess method could really use a removal > > > > i strictly oppose fixing it with even more hacks ^^ > > > > -- Ronny > > > > _______________________________________________ > > pytest-dev mailing list > > [email protected] > > https://mail.python.org/mailman/listinfo/pytest-dev > > > _______________________________________________ > pytest-dev mailing list > [email protected] > https://mail.python.org/mailman/listinfo/pytest-dev _______________________________________________ pytest-dev mailing list [email protected] https://mail.python.org/mailman/listinfo/pytest-dev
