comparison mercurial/store.py @ 45942:89a2afe31e82

formating: upgrade to black 20.8b1

This required a couple of small tweaks to un-confuse black, but now it
works. Big formatting changes come from:

 * Dramatically improved collection-splitting logic upstream
 * Black having a strong (correct IMO) opinion that """ is better than '''

Differential Revision: https://phab.mercurial-scm.org/D9430

45941:346af7687c6f

# This avoids a collision between a file named foo and a dir named
# foo.i or foo.d
def _encodedir(path):
    '''
    >>> _encodedir(b'data/foo.i')
    'data/foo.i'
    >>> _encodedir(b'data/foo.i/bla.i')
    'data/foo.i.hg/bla.i'
    >>> _encodedir(b'data/foo.i.hg/bla.i')
    'data/foo.i.hg.hg/bla.i'
    >>> _encodedir(b'data/foo.i\\ndata/foo.i/bla.i\\ndata/foo.i.hg/bla.i\\n')
    'data/foo.i\\ndata/foo.i.hg/bla.i\\ndata/foo.i.hg.hg/bla.i\\n'
    '''
    return (
        path.replace(b".hg/", b".hg.hg/")
        .replace(b".i/", b".i.hg/")
        .replace(b".d/", b".d.hg/")
    )

encodedir = getattr(parsers, 'encodedir', _encodedir)


def decodedir(path):
    '''
    >>> decodedir(b'data/foo.i')
    'data/foo.i'
    >>> decodedir(b'data/foo.i.hg/bla.i')
    'data/foo.i/bla.i'
    >>> decodedir(b'data/foo.i.hg.hg/bla.i')
    'data/foo.i.hg/bla.i'
    '''
    if b".hg/" not in path:
        return path
    return (
        path.replace(b".d.hg/", b".d/")
        .replace(b".i.hg/", b".i/")
        .replace(b".hg.hg/", b".hg/")
    )


def _reserved():
    ''' characters that are problematic for filesystems

    * ascii escapes (0..31)
    * ascii hi (126..255)
    * windows specials

    these characters will be escaped by encodefunctions
    '''
    winreserved = [ord(x) for x in u'\\:*?"<>|']
    for x in range(32):
        yield x
    for x in range(126, 256):
        yield x
    for x in winreserved:
        yield x


def _buildencodefun():
    '''
    >>> enc, dec = _buildencodefun()

    >>> enc(b'nothing/special.txt')
    'nothing/special.txt'
    >>> dec(b'nothing/special.txt')

    >>> enc(b'the\\x07quick\\xADshot')
    'the~07quick~adshot'
    >>> dec(b'the~07quick~adshot')
    'the\\x07quick\\xadshot'
    '''
    e = b'_'
    xchr = pycompat.bytechr
    asciistr = list(map(xchr, range(127)))
    capitals = list(range(ord(b"A"), ord(b"Z") + 1))

_encodefname, _decodefname = _buildencodefun()


def encodefilename(s):
    '''
    >>> encodefilename(b'foo.i/bar.d/bla.hg/hi:world?/HELLO')
    'foo.i.hg/bar.d.hg/bla.hg.hg/hi~3aworld~3f/_h_e_l_l_o'
    '''
    return _encodefname(encodedir(s))


def decodefilename(s):
    '''
    >>> decodefilename(b'foo.i.hg/bar.d.hg/bla.hg.hg/hi~3aworld~3f/_h_e_l_l_o')
    'foo.i/bar.d/bla.hg/hi:world?/HELLO'
    '''
    return decodedir(_decodefname(s))


def _buildlowerencodefun():
    '''
    >>> f = _buildlowerencodefun()
    >>> f(b'nothing/special.txt')
    'nothing/special.txt'
    >>> f(b'HELLO')
    'hello'
    >>> f(b'hello:world?')
    'hello~3aworld~3f'
    >>> f(b'the\\x07quick\\xADshot')
    'the~07quick~adshot'
    '''
    xchr = pycompat.bytechr
    cmap = {xchr(x): xchr(x) for x in pycompat.xrange(127)}
    for x in _reserved():
        cmap[xchr(x)] = b"~%02x" % x
    for x in range(ord(b"A"), ord(b"Z") + 1):

_winres3 = (b'aux', b'con', b'prn', b'nul')  # length 3
_winres4 = (b'com', b'lpt')  # length 4 (with trailing 1..9)


def _auxencode(path, dotencode):
    '''
    Encodes filenames containing names reserved by Windows or which end in
    period or space. Does not touch other single reserved characters c.
    Specifically, c in '\\:*?"<>|' or ord(c) <= 31 are *not* encoded here.
    Additionally encodes space or period at the beginning, if dotencode is
    True. Parameter path is assumed to be all lowercase.

    ['.com1com2', 'lp~749.lpt4.lpt1', 'conprn', 'com0', 'lpt0', 'foo~2e']
    >>> _auxencode([b'foo. '], True)
    ['foo.~20']
    >>> _auxencode([b' .foo'], True)
    ['~20.foo']
    '''
    for i, n in enumerate(path):
        if not n:
            continue
        if dotencode and n[0] in b'. ':
            n = b"~%02x" % ord(n[0:1]) + n[1:]

        res = b'dh/' + dirs + filler + digest + ext
    return res


def _hybridencode(path, dotencode):
    '''encodes path with a length limit

    Encodes all paths that begin with 'data/', according to the following.

    Default encoding (reversible):

    basename have been taken).
    The extension (e.g. '.i' or '.d') is preserved.

    The string 'data/' at the beginning is replaced with 'dh/', if the hashed
    encoding was used.
    '''
    path = encodedir(path)
    ef = _encodefname(path).split(b'/')
    res = b'/'.join(_auxencode(ef, dotencode))
    if len(res) > _maxstorepathlen:
        res = _hashencode(path, dotencode)

    def topfiles(self):
        # yield manifest before changelog
        return reversed(self._walk(b'', False))

    def walk(self, matcher=None):
        '''yields (unencoded, encoded, size)

        if a matcher is passed, storage files of only those tracked paths
        are passed with matches the matcher
        '''
        # yield data files first
        for x in self.datafiles(matcher):
            yield x
        for x in self.topfiles():
            yield x

        self._dirty = False
        # set of new additions to fncache
        self.addls = set()

    def ensureloaded(self, warn=None):
        '''read the fncache file if not already read.

        If the file on disk is corrupted, raise. If warn is provided,
        warn and keep going instead.'''
        if self.entries is None:
            self._load(warn)

    def _load(self, warn=None):
        '''fill the entries from the fncache file'''

45942:89a2afe31e82

# This avoids a collision between a file named foo and a dir named
# foo.i or foo.d
def _encodedir(path):
    """
    >>> _encodedir(b'data/foo.i')
    'data/foo.i'
    >>> _encodedir(b'data/foo.i/bla.i')
    'data/foo.i.hg/bla.i'
    >>> _encodedir(b'data/foo.i.hg/bla.i')
    'data/foo.i.hg.hg/bla.i'
    >>> _encodedir(b'data/foo.i\\ndata/foo.i/bla.i\\ndata/foo.i.hg/bla.i\\n')
    'data/foo.i\\ndata/foo.i.hg/bla.i\\ndata/foo.i.hg.hg/bla.i\\n'
    """
    return (
        path.replace(b".hg/", b".hg.hg/")
        .replace(b".i/", b".i.hg/")
        .replace(b".d/", b".d.hg/")
    )

encodedir = getattr(parsers, 'encodedir', _encodedir)


def decodedir(path):
    """
    >>> decodedir(b'data/foo.i')
    'data/foo.i'
    >>> decodedir(b'data/foo.i.hg/bla.i')
    'data/foo.i/bla.i'
    >>> decodedir(b'data/foo.i.hg.hg/bla.i')
    'data/foo.i.hg/bla.i'
    """
    if b".hg/" not in path:
        return path
    return (
        path.replace(b".d.hg/", b".d/")
        .replace(b".i.hg/", b".i/")
        .replace(b".hg.hg/", b".hg/")
    )


def _reserved():
    """characters that are problematic for filesystems

    * ascii escapes (0..31)
    * ascii hi (126..255)
    * windows specials

    these characters will be escaped by encodefunctions
    """
    winreserved = [ord(x) for x in u'\\:*?"<>|']
    for x in range(32):
        yield x
    for x in range(126, 256):
        yield x
    for x in winreserved:
        yield x


def _buildencodefun():
    """
    >>> enc, dec = _buildencodefun()

    >>> enc(b'nothing/special.txt')
    'nothing/special.txt'
    >>> dec(b'nothing/special.txt')

    >>> enc(b'the\\x07quick\\xADshot')
    'the~07quick~adshot'
    >>> dec(b'the~07quick~adshot')
    'the\\x07quick\\xadshot'
    """
    e = b'_'
    xchr = pycompat.bytechr
    asciistr = list(map(xchr, range(127)))
    capitals = list(range(ord(b"A"), ord(b"Z") + 1))

_encodefname, _decodefname = _buildencodefun()


def encodefilename(s):
    """
    >>> encodefilename(b'foo.i/bar.d/bla.hg/hi:world?/HELLO')
    'foo.i.hg/bar.d.hg/bla.hg.hg/hi~3aworld~3f/_h_e_l_l_o'
    """
    return _encodefname(encodedir(s))


def decodefilename(s):
    """
    >>> decodefilename(b'foo.i.hg/bar.d.hg/bla.hg.hg/hi~3aworld~3f/_h_e_l_l_o')
    'foo.i/bar.d/bla.hg/hi:world?/HELLO'
    """
    return decodedir(_decodefname(s))


def _buildlowerencodefun():
    """
    >>> f = _buildlowerencodefun()
    >>> f(b'nothing/special.txt')
    'nothing/special.txt'
    >>> f(b'HELLO')
    'hello'
    >>> f(b'hello:world?')
    'hello~3aworld~3f'
    >>> f(b'the\\x07quick\\xADshot')
    'the~07quick~adshot'
    """
    xchr = pycompat.bytechr
    cmap = {xchr(x): xchr(x) for x in pycompat.xrange(127)}
    for x in _reserved():
        cmap[xchr(x)] = b"~%02x" % x
    for x in range(ord(b"A"), ord(b"Z") + 1):

_winres3 = (b'aux', b'con', b'prn', b'nul')  # length 3
_winres4 = (b'com', b'lpt')  # length 4 (with trailing 1..9)


def _auxencode(path, dotencode):
    """
    Encodes filenames containing names reserved by Windows or which end in
    period or space. Does not touch other single reserved characters c.
    Specifically, c in '\\:*?"<>|' or ord(c) <= 31 are *not* encoded here.
    Additionally encodes space or period at the beginning, if dotencode is
    True. Parameter path is assumed to be all lowercase.

    ['.com1com2', 'lp~749.lpt4.lpt1', 'conprn', 'com0', 'lpt0', 'foo~2e']
    >>> _auxencode([b'foo. '], True)
    ['foo.~20']
    >>> _auxencode([b' .foo'], True)
    ['~20.foo']
    """
    for i, n in enumerate(path):
        if not n:
            continue
        if dotencode and n[0] in b'. ':
            n = b"~%02x" % ord(n[0:1]) + n[1:]

        res = b'dh/' + dirs + filler + digest + ext
    return res


def _hybridencode(path, dotencode):
    """encodes path with a length limit

    Encodes all paths that begin with 'data/', according to the following.

    Default encoding (reversible):

    basename have been taken).
    The extension (e.g. '.i' or '.d') is preserved.

    The string 'data/' at the beginning is replaced with 'dh/', if the hashed
    encoding was used.
    """
    path = encodedir(path)
    ef = _encodefname(path).split(b'/')
    res = b'/'.join(_auxencode(ef, dotencode))
    if len(res) > _maxstorepathlen:
        res = _hashencode(path, dotencode)

    def topfiles(self):
        # yield manifest before changelog
        return reversed(self._walk(b'', False))

    def walk(self, matcher=None):
        """yields (unencoded, encoded, size)

        if a matcher is passed, storage files of only those tracked paths
        are passed with matches the matcher
        """
        # yield data files first
        for x in self.datafiles(matcher):
            yield x
        for x in self.topfiles():
            yield x

        self._dirty = False
        # set of new additions to fncache
        self.addls = set()

    def ensureloaded(self, warn=None):
        """read the fncache file if not already read.

        If the file on disk is corrupted, raise. If warn is provided,
        warn and keep going instead."""
        if self.entries is None:
            self._load(warn)

    def _load(self, warn=None):
        '''fill the entries from the fncache file'''