Mercurial > hg
view mercurial/cffi/osutil.py @ 39037:ede768cfe83e
mail: always fall back to iso-8859-1 if us-ascii won't work (BC)
It looks like this was a well-intentioned backwards compat hack for
previewing the output of `hg email` in a stable way. Unfortunately I
think this hack's time has come, because Python 3 does a much better
job of ensuring it actually emits *valid* email messages. In
particular, Python 2 would blindly trust us that the bytes we handed
it were valid for the encoding we claimed, but Python 3 has some more
sniff-tests that we end up failing.
As a result, if we're going to print an email to the terminal, try
us-ascii first, but if that fails go straight to iso-8859-1 which
should be reasonably readable for ascii-compatible patch bodies. This
*will* be a breaking change for ascii-incompatible textual patch
content, but I don't think that's avoidable if we want to continue
using the email library from the stdlib.
.. bc::
Emails from the patchbomb extension will always be printed as though
they are iso-8859-1 if they're not valid us-ascii. Previously,
previewed emails were always claimed to be us-ascii and might
contain invalid byte sequences.
Differential Revision: https://phab.mercurial-scm.org/D4231
author | Augie Fackler <augie@google.com> |
---|---|
date | Thu, 09 Aug 2018 21:04:15 -0400 |
parents | dacfcdd8b94e |
children | 2372284d9457 |
line wrap: on
line source
# osutil.py - CFFI version of osutil.c # # Copyright 2016 Maciej Fijalkowski <fijall@gmail.com> # # This software may be used and distributed according to the terms of the # GNU General Public License version 2 or any later version. from __future__ import absolute_import import os import stat as statmod from ..pure.osutil import * from .. import ( pycompat, ) if pycompat.isdarwin: from . import _osutil ffi = _osutil.ffi lib = _osutil.lib listdir_batch_size = 4096 # tweakable number, only affects performance, which chunks # of bytes do we get back from getattrlistbulk attrkinds = [None] * 20 # we need the max no for enum VXXX, 20 is plenty attrkinds[lib.VREG] = statmod.S_IFREG attrkinds[lib.VDIR] = statmod.S_IFDIR attrkinds[lib.VLNK] = statmod.S_IFLNK attrkinds[lib.VBLK] = statmod.S_IFBLK attrkinds[lib.VCHR] = statmod.S_IFCHR attrkinds[lib.VFIFO] = statmod.S_IFIFO attrkinds[lib.VSOCK] = statmod.S_IFSOCK class stat_res(object): def __init__(self, st_mode, st_mtime, st_size): self.st_mode = st_mode self.st_mtime = st_mtime self.st_size = st_size tv_sec_ofs = ffi.offsetof("struct timespec", "tv_sec") buf = ffi.new("char[]", listdir_batch_size) def listdirinternal(dfd, req, stat, skip): ret = [] while True: r = lib.getattrlistbulk(dfd, req, buf, listdir_batch_size, 0) if r == 0: break if r == -1: raise OSError(ffi.errno, os.strerror(ffi.errno)) cur = ffi.cast("val_attrs_t*", buf) for i in range(r): lgt = cur.length assert lgt == ffi.cast('uint32_t*', cur)[0] ofs = cur.name_info.attr_dataoffset str_lgt = cur.name_info.attr_length base_ofs = ffi.offsetof('val_attrs_t', 'name_info') name = str(ffi.buffer(ffi.cast("char*", cur) + base_ofs + ofs, str_lgt - 1)) tp = attrkinds[cur.obj_type] if name == "." or name == "..": continue if skip == name and tp == statmod.S_ISDIR: return [] if stat: mtime = cur.mtime.tv_sec mode = (cur.accessmask & ~lib.S_IFMT)| tp ret.append((name, tp, stat_res(st_mode=mode, st_mtime=mtime, st_size=cur.datalength))) else: ret.append((name, tp)) cur = ffi.cast("val_attrs_t*", int(ffi.cast("intptr_t", cur)) + lgt) return ret def listdir(path, stat=False, skip=None): req = ffi.new("struct attrlist*") req.bitmapcount = lib.ATTR_BIT_MAP_COUNT req.commonattr = (lib.ATTR_CMN_RETURNED_ATTRS | lib.ATTR_CMN_NAME | lib.ATTR_CMN_OBJTYPE | lib.ATTR_CMN_ACCESSMASK | lib.ATTR_CMN_MODTIME) req.fileattr = lib.ATTR_FILE_DATALENGTH dfd = lib.open(path, lib.O_RDONLY, 0) if dfd == -1: raise OSError(ffi.errno, os.strerror(ffi.errno)) try: ret = listdirinternal(dfd, req, stat, skip) finally: try: lib.close(dfd) except BaseException: pass # we ignore all the errors from closing, not # much we can do about that return ret