This is an archive of the discontinued Mercurial Phabricator instance.

Differential D10798

util: add `nb_bytes` argument to `copyfile` to partially copy a file
ClosedPublic

Authored by marmoute on May 30 2021, 1:30 PM.

Details

Reviewers
Alphare
Group Reviewers
hg-reviewers
Commits
rHG9b841267253c: util: add `nb_bytes` argument to `copyfile` to partially copy a file
Summary

When set, this allow to copy only the first nb_bytes of a file. This will be
useful for censor/strip operation with revlogv2.

Diff Detail

Repository
rHG Mercurial
Lint
Automatic diff as part of commit; lint not applicable.
Unit
Automatic diff as part of commit; unit tests not applicable.

Event Timeline

marmoute created this revision.May 30 2021, 1:30 PM
baymax updated this revision to Diff 28311.May 31 2021, 6:51 AM

✅ refresh by Heptapod after a successful CI run (🐙 💚)

Alphare requested changes to this revision.May 31 2021, 10:49 AM
Alphare added a subscriber: Alphare.
Alphare added inline comments.
mercurial/util.py
1945

I think this message is confused!

This revision now requires changes to proceed.May 31 2021, 10:49 AM
baymax updated this revision to Diff 28326.May 31 2021, 4:21 PM

✅ refresh by Heptapod after a successful CI run (🐙 💚)

baymax updated this revision to Diff 28408.Jun 1 2021, 11:17 PM

✅ refresh by Heptapod after a successful CI run (🐙 💚)

Alphare accepted this revision as: Alphare.Jun 2 2021, 4:21 AM
baymax updated this revision to Diff 28423.Jun 2 2021, 7:43 AM

✅ refresh by Heptapod after a successful CI run (🐙 💚)

baymax updated this revision to Diff 28457.Jun 4 2021, 11:05 AM

✅ refresh by Heptapod after a successful CI run (🐙 💚)

baymax updated this revision to Diff 28479.Jun 7 2021, 5:17 AM

✅ refresh by Heptapod after a successful CI run (🐙 💚)

marmoute added a commit: rHG9b841267253c: util: add `nb_bytes` argument to `copyfile` to partially copy a file.Jun 7 2021, 5:26 AM
This revision was not accepted when it landed; it landed in state Needs Review.
Closed by commit rHG9b841267253c: util: add `nb_bytes` argument to `copyfile` to partially copy a file (authored by marmoute). · Explain Why
This revision was automatically updated to reflect the committed changes.

Revision Contents
Changeset List

PathPackages
Mmercurial/util.py (16 lines)

Diff 28501

mercurial/util.py

b'reiserfs', b'reiserfs',
b'tmpfs', b'tmpfs',
b'ufs', b'ufs',
b'xfs', b'xfs',
b'zfs', b'zfs',
} }
def copyfile(src, dest, hardlink=False, copystat=False, checkambig=False): def copyfile(
src, dest, hardlink=False, copystat=False, checkambig=False, nb_bytes=None
):
"""copy a file, preserving mode and optionally other stat info like """copy a file, preserving mode and optionally other stat info like
atime/mtime atime/mtime
checkambig argument is used with filestat, and is useful only if checkambig argument is used with filestat, and is useful only if
destination file is guarded by any lock (e.g. repo.lock or destination file is guarded by any lock (e.g. repo.lock or
repo.wlock). repo.wlock).
copystat and checkambig should be exclusive. copystat and checkambig should be exclusive.
nb_bytes: if set only copy the first `nb_bytes` of the source file.
""" """
assert not (copystat and checkambig) assert not (copystat and checkambig)
oldstat = None oldstat = None
if os.path.lexists(dest): if os.path.lexists(dest):
if checkambig: if checkambig:
oldstat = checkambig and filestat.frompath(dest) oldstat = checkambig and filestat.frompath(dest)
unlink(dest) unlink(dest)
if hardlink: if hardlink:
# Hardlinks are problematic on CIFS (issue4546), do not allow hardlinks # Hardlinks are problematic on CIFS (issue4546), do not allow hardlinks
# unless we are confident that dest is on a whitelisted filesystem. # unless we are confident that dest is on a whitelisted filesystem.
try: try:
fstype = getfstype(os.path.dirname(dest)) fstype = getfstype(os.path.dirname(dest))
except OSError: except OSError:
fstype = None fstype = None
if fstype not in _hardlinkfswhitelist: if fstype not in _hardlinkfswhitelist:
hardlink = False hardlink = False
if hardlink: if hardlink:
try: try:
oslink(src, dest) oslink(src, dest)
if nb_bytes is not None:
m = "the `nb_bytes` argument is incompatible with `hardlink`"
AlphareUnsubmitted
Not Done

I think this message is confused!

Alphare: I think this message is confused!
raise error.ProgrammingError(m)
return return
except (IOError, OSError): except (IOError, OSError):
pass # fall back to normal copy pass # fall back to normal copy
if os.path.islink(src): if os.path.islink(src):
os.symlink(os.readlink(src), dest) os.symlink(os.readlink(src), dest)
# copytime is ignored for symlinks, but in general copytime isn't needed # copytime is ignored for symlinks, but in general copytime isn't needed
# for them anyway # for them anyway
if nb_bytes is not None:
m = "cannot use `nb_bytes` on a symlink"
raise error.ProgrammingError(m)
else: else:
try: try:
shutil.copyfile(src, dest) shutil.copyfile(src, dest)
if copystat: if copystat:
# copystat also copies mode # copystat also copies mode
shutil.copystat(src, dest) shutil.copystat(src, dest)
else: else:
shutil.copymode(src, dest) shutil.copymode(src, dest)
if oldstat and oldstat.stat: if oldstat and oldstat.stat:
newstat = filestat.frompath(dest) newstat = filestat.frompath(dest)
if newstat.isambig(oldstat): if newstat.isambig(oldstat):
# stat of copied file is ambiguous to original one # stat of copied file is ambiguous to original one
advanced = ( advanced = (
oldstat.stat[stat.ST_MTIME] + 1 oldstat.stat[stat.ST_MTIME] + 1
) & 0x7FFFFFFF ) & 0x7FFFFFFF
os.utime(dest, (advanced, advanced)) os.utime(dest, (advanced, advanced))
# We could do something smarter using `copy_file_range` call or similar
if nb_bytes is not None:
with open(dest, mode='r+') as f:
f.truncate(nb_bytes)
except shutil.Error as inst: except shutil.Error as inst:
raise error.Abort(stringutil.forcebytestr(inst)) raise error.Abort(stringutil.forcebytestr(inst))
def copyfiles(src, dst, hardlink=None, progress=None): def copyfiles(src, dst, hardlink=None, progress=None):
"""Copy a directory tree using hardlinks if possible.""" """Copy a directory tree using hardlinks if possible."""
num = 0 num = 0