Handle symlinks in req_uninstall.py #5442
Conversation
HaraldNordgren
force-pushed
the
issues/5439
branch
2 times, most recently
from
875937a
to
ac20718
Compare
HaraldNordgren
changed the title
HaraldNordgren
force-pushed
the
issues/5439
branch
from
ac20718
to
df48b85
Compare
HaraldNordgren
changed the title
| @@ -359,7 +359,7 @@ def from_dist(cls, dist): | |||
| # develop egg | |||
| with open(develop_egg_link, 'r') as fh: | |||
| link_pointer = os.path.normcase(fh.readline().strip()) | |||
| assert (link_pointer == dist.location), ( | |||
| assert (os.stat(link_pointer) == os.stat(dist.location)), ( | |||
There was a problem hiding this comment.
Should use os.path.samefile, as this is defined to do precisely this job. You may need to fall back to something else on Python 2.7, as samefile only added Windows support in Python 3.2, but I'm not sure if a stat comparison is sufficient on before Python 3.4 on Windows (the docs to samefile say "Changed in version 3.4: Windows now uses the same implementation as all other platforms" which implies that prior to that, Windows needed special treatment).
OTOH, this is "only" an assertion, not a normal error - so maybe it could be removed if it's asserting something that isn't always true? What do the other @pypa/pip-committers think?
HaraldNordgren
force-pushed
the
issues/5439
branch
2 times, most recently
from
ed00a03
to
3e948a3
Compare
Handle symlinks in req_uninstall.py
HaraldNordgren
force-pushed
the
issues/5439
branch
from
3e948a3
to
57a0231
Compare
|
@pfmoore Updated. |
pfmoore
dismissed
their stale review
Review comments addressed. I'm not going so far as to "approve" this change, though. There's still the question of do we need the assertion anyway. Unfortunately, github doesn't offer a plain "Confirm review comments have been addressed without approving" button...
|
This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs. |
lock
bot
added
the
auto-locked
Fixes #5439.