( )⚙ D7650 match: normalize `cwd` early

This is an archive of the discontinued Mercurial Phabricator instance.

Differential D7650

match: normalize `cwd` early
ClosedPublic

Authored by martinvonz on Dec 13 2019, 2:50 PM.

Details

Reviewers
pulkit
Group Reviewers
hg-reviewers
Commits
rHG5685ce2ea3bf: match: normalize `cwd` early
Summary

By having cwd in absolute form, we won't have to adjust it when
passing it to subrepo matchers. This will matter for a coming patch.

Diff Detail

Repository
rHG Mercurial
Branch
default
Lint
No Linters Available
Unit
No Unit Test Coverage

Event Timeline

martinvonz created this revision.Dec 13 2019, 2:50 PM
pulkit added a subscriber: pulkit.Dec 20 2019, 7:17 AM
pulkit added inline comments.
mercurial/match.py
232

util.normpath instead of os.path.normpath

martinvonz updated this revision to Diff 18894.Dec 20 2019, 9:52 AM
martinvonz marked an inline comment as done.Dec 20 2019, 9:52 AM
pulkit accepted this revision.Dec 20 2019, 11:33 AM
This revision is now accepted and ready to land.Dec 20 2019, 11:33 AM
martinvonz added a commit: rHG5685ce2ea3bf: match: normalize `cwd` early.Dec 21 2019, 9:41 AM
Closed by commit rHG5685ce2ea3bf: match: normalize `cwd` early (authored by martinvonz). · Explain Why
This revision was automatically updated to reflect the committed changes.
mharbison72 added a comment.Dec 23 2019, 3:34 PM
In D7650#113661, @mharbison72 wrote:

This ends up making Windows unhappy. I haven't had a chance to look into it yet.
https://ci.hg.gregoryszorc.com/job-info/hg-committed-5685ce2ea3bfe2f82ba0119788e2fef853bcf97a-windows-cpython-2.7-x64-0#failed-tests

(The doctest failures are from D7649. The ../$TESTTMP stuff is this one.)

Revision Contents
Changeset List

PathPackages
Mmercurial/match.py (1 line)

Diff 18695

mercurial/match.py

2. Using the exact() method only returns True if the file name matches one 2. Using the exact() method only returns True if the file name matches one
of the exact patterns (i.e. not re: or glob: patterns): of the exact patterns (i.e. not re: or glob: patterns):
>>> m.exact(b'a') >>> m.exact(b'a')
True True
>>> m.exact(b'main.c') >>> m.exact(b'main.c')
False False
""" """
assert os.path.isabs(root) assert os.path.isabs(root)
cwd = os.path.normpath(os.path.join(root, cwd))
pulkitUnsubmitted
Done

util.normpath instead of os.path.normpath

pulkit: `util.normpath` instead of `os.path.normpath`
normalize = _donormalize normalize = _donormalize
if icasefs: if icasefs:
dirstate = ctx.repo().dirstate dirstate = ctx.repo().dirstate
dsnormalize = dirstate.normalize dsnormalize = dirstate.normalize
def normalize(patterns, default, root, cwd, auditor, warn): def normalize(patterns, default, root, cwd, auditor, warn):
kp = _donormalize(patterns, default, root, cwd, auditor, warn) kp = _donormalize(patterns, default, root, cwd, auditor, warn)
kindpats = [] kindpats = []