The patch that Ridecell/rope introduces (python-rope/rope@python-rope:54433b8...Ridecell:e7bfd4f) seems fairly small, so I'm wondering if it's really (still) needed:
At least one part of it seems to be about changing Rope's MoveModule so that it becomes possible to specify a module as the destination rather than a package. Can't the need for this be circumvented by just creating a package above the new location using rope.contrib.generate.create_package before performing the move? I wrote a proof-of-concept script which does just that before I found pymv1 and it seemed to work fine.
1 Pretty funny: I searched for all kinds of keywords on Google to no avail, until I finally gave up, decided to write such a tool myself, and had the perfect idea for what it should be called: pymv. But I guessed someone had already taken that name on PyPI for something different, so I checked, and no - it's the exact kind of tool I was going to write!
The patch that
Ridecell/ropeintroduces (python-rope/rope@python-rope:54433b8...Ridecell:e7bfd4f) seems fairly small, so I'm wondering if it's really (still) needed:At least one part of it seems to be about changing Rope's
MoveModuleso that it becomes possible to specify a module as the destination rather than a package. Can't the need for this be circumvented by just creating a package above the new location usingrope.contrib.generate.create_packagebefore performing the move? I wrote a proof-of-concept script which does just that before I foundpymv1 and it seemed to work fine.1 Pretty funny: I searched for all kinds of keywords on Google to no avail, until I finally gave up, decided to write such a tool myself, and had the perfect idea for what it should be called:
pymv. But I guessed someone had already taken that name on PyPI for something different, so I checked, and no - it's the exact kind of tool I was going to write!