This is an archive of the discontinued Mercurial Phabricator instance.

Differential D12346

util: remove iterfile() variant for buggy EINTR handling
ClosedPublic

Authored by indygreg on Mar 3 2022, 10:30 PM.

Details

Reviewers
Alphare
Group Reviewers
hg-reviewers
Commits
rHG2974cdda819b: util: remove iterfile() variant for buggy EINTR handling
Summary

The workaround for Python 2 is no longer needed. So we can delete some
code.

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

indygreg created this revision.Mar 3 2022, 10:30 PM
Alphare accepted this revision.Mar 8 2022, 4:43 AM
This revision is now accepted and ready to land.Mar 8 2022, 4:43 AM
indygreg added a commit: rHG2974cdda819b: util: remove iterfile() variant for buggy EINTR handling.Mar 8 2022, 6:03 AM
Closed by commit rHG2974cdda819b: util: remove iterfile() variant for buggy EINTR handling (authored by indygreg). · Explain Why
This revision was automatically updated to reflect the committed changes.

Revision Contents
Changeset List

PathPackages
Mmercurial/util.py (49 lines)

Diff 32597

mercurial/util.py

import errno import errno
import gc import gc
import hashlib import hashlib
import io import io
import itertools import itertools
import locale import locale
import mmap import mmap
import os import os
import platform as pyplatform
import re as remod import re as remod
import shutil import shutil
import stat import stat
import sys import sys
import time import time
import traceback import traceback
import warnings import warnings
tonativeeol = tocrlf tonativeeol = tocrlf
fromnativeeol = tolf fromnativeeol = tolf
nativeeolwriter = _crlfwriter nativeeolwriter = _crlfwriter
else: else:
tonativeeol = pycompat.identity tonativeeol = pycompat.identity
fromnativeeol = pycompat.identity fromnativeeol = pycompat.identity
nativeeolwriter = pycompat.identity nativeeolwriter = pycompat.identity
if pyplatform.python_implementation() == b'CPython' and sys.version_info < (
3,
0,
):
# There is an issue in CPython that some IO methods do not handle EINTR
# correctly. The following table shows what CPython version (and functions)
# are affected (buggy: has the EINTR bug, okay: otherwise):
#
# | < 2.7.4 | 2.7.4 to 2.7.12 | >= 3.0
# --------------------------------------------------
# fp.__iter__ | buggy | buggy | okay
# fp.read* | buggy | okay [1] | okay
#
# [1]: fixed by changeset 67dc99a989cd in the cpython hg repo.
#
# Here we workaround the EINTR issue for fileobj.__iter__. Other methods
# like "read*" work fine, as we do not support Python < 2.7.4.
#
# Although we can workaround the EINTR issue for fp.__iter__, it is slower:
# "for x in fp" is 4x faster than "for x in iter(fp.readline, '')" in
# CPython 2, because CPython 2 maintains an internal readahead buffer for
# fp.__iter__ but not other fp.read* methods.
#
# On modern systems like Linux, the "read" syscall cannot be interrupted
# when reading "fast" files like on-disk files. So the EINTR issue only
# affects things like pipes, sockets, ttys etc. We treat "normal" (S_ISREG)
# files approximately as "fast" files and use the fast (unsafe) code path,
# to minimize the performance impact.
def iterfile(fp): # TODO delete since workaround variant for Python 2 no longer needed.
fastpath = True
if type(fp) is file:
fastpath = stat.S_ISREG(os.fstat(fp.fileno()).st_mode)
if fastpath:
return fp
else:
# fp.readline deals with EINTR correctly, use it as a workaround.
return iter(fp.readline, b'')
else:
# PyPy and CPython 3 do not have the EINTR issue thus no workaround needed.
def iterfile(fp): def iterfile(fp):
return fp return fp
def iterlines(iterator): def iterlines(iterator):
# type: (Iterator[bytes]) -> Iterator[bytes] # type: (Iterator[bytes]) -> Iterator[bytes]
for chunk in iterator: for chunk in iterator:
for line in chunk.splitlines(): for line in chunk.splitlines():
yield line yield line