view mercurial/cffi/osutil.py @ 45637:ad6ebb6f0dfe

copies: make two version of the changeset centric algorithm They are two main ways to run the changeset-centric copy-tracing algorithm. One fed from data stored in side-data and still in development, and one based on data stored in extra (with a "compatibility" mode). The `extra` based is used in production at Google, but still experimental in code. It is mostly unsuitable for other users because it affects the hash. The side-data based storage and algorithm have been evolving to store more data, cover more cases (mostly around merge, that Google do not really care about) and use lower level storage for efficiency. All this changes make is increasingly hard to maintain de common code base, without impacting code complexity and performance. For example, the compatibility mode requires to keep things at different level than what we need for side-data. So, I am duplicating the involved functions. The newly added `_extra` variants will be kept as today, while I will do some deeper rework of the side data versions. Long terms, the side-data version should be more featureful and performant than the extra based version, so I expect the duplicated `_extra` functions to eventually get dropped. Differential Revision: https://phab.mercurial-scm.org/D9114
author Pierre-Yves David <pierre-yves.david@octobus.net>
date Fri, 25 Sep 2020 14:39:04 +0200
parents 687b865b95ad
children 521ac0d7047f
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(b"struct timespec", b"tv_sec")
    buf = ffi.new(b"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(b"val_attrs_t*", buf)
            for i in range(r):
                lgt = cur.length
                assert lgt == ffi.cast(b'uint32_t*', cur)[0]
                ofs = cur.name_info.attr_dataoffset
                str_lgt = cur.name_info.attr_length
                base_ofs = ffi.offsetof(b'val_attrs_t', b'name_info')
                name = str(
                    ffi.buffer(
                        ffi.cast(b"char*", cur) + base_ofs + ofs, str_lgt - 1
                    )
                )
                tp = attrkinds[cur.obj_type]
                if name == b"." or name == b"..":
                    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(
                    b"val_attrs_t*", int(ffi.cast(b"intptr_t", cur)) + lgt
                )
        return ret

    def listdir(path, stat=False, skip=None):
        req = ffi.new(b"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