# manifest.py - manifest revision class for mercurial

#

# Copyright 2005-2007 Matt Mackall <mpm@selenic.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 i18n import _

import mdiff, parsers, error, revlog, util

import array, struct

import os

propertycache = util.propertycache

def _parsev1(data):

    # This method does a little bit of excessive-looking

    # precondition checking. This is so that the behavior of this

    # class exactly matches its C counterpart to try and help

    # prevent surprise breakage for anyone that develops against

    # the pure version.

    if data and data[-1] != '\n':

        raise ValueError('Manifest did not end in a newline.')

    prev = None

    for l in data.splitlines():

        if prev is not None and prev > l:

            raise ValueError('Manifest lines not in sorted order.')

        prev = l

        f, n = l.split('\0')

        if len(n) > 40:

            yield f, revlog.bin(n[:40]), n[40:]

        else:

            yield f, revlog.bin(n), ''

def _parsev2(data):

    metadataend = data.find('\n')

    # Just ignore metadata for now

    pos = metadataend + 1

    prevf = ''

    while pos < len(data):

        end = data.find('\n', pos + 1) # +1 to skip stem length byte

        if end == -1:

            raise ValueError('Manifest ended with incomplete file entry.')

        stemlen = ord(data[pos])

        items = data[pos + 1:end].split('\0')

        f = prevf[:stemlen] + items[0]

        if prevf > f:

            raise ValueError('Manifest entries not in sorted order.')

        fl = items[1]

        # Just ignore metadata (items[2:] for now)

        n = data[end + 1:end + 21]

        yield f, n, fl

        pos = end + 22

        prevf = f

def _parse(data):

    """Generates (path, node, flags) tuples from a manifest text"""

    if data.startswith('\0'):

        return iter(_parsev2(data))

    else:

        return iter(_parsev1(data))

def _text(it, usemanifestv2):

    """Given an iterator over (path, node, flags) tuples, returns a manifest

    text"""

    if usemanifestv2:

        return _textv2(it)

    else:

        return _textv1(it)

def _textv1(it):

    files = []

    lines = []

    _hex = revlog.hex

    for f, n, fl in it:

        files.append(f)

        # if this is changed to support newlines in filenames,

        # be sure to check the templates/ dir again (especially *-raw.tmpl)

        lines.append("%s\0%s%s\n" % (f, _hex(n), fl))

    _checkforbidden(files)

    return ''.join(lines)

def _textv2(it):

    files = []

    lines = ['\0\n']

    prevf = ''

    for f, n, fl in it:

        files.append(f)

        stem = os.path.commonprefix([prevf, f])

        stemlen = min(len(stem), 255)

        lines.append("%c%s\0%s\n%s\n" % (stemlen, f[stemlen:], fl, n))

        prevf = f

    _checkforbidden(files)

    return ''.join(lines)

class _lazymanifest(dict):

    """This is the pure implementation of lazymanifest.

    It has not been optimized *at all* and is not lazy.

    """

    def __init__(self, data):

        dict.__init__(self)

        for f, n, fl in _parse(data):

            self[f] = n, fl

    def __setitem__(self, k, v):

        node, flag = v

        assert node is not None

        if len(node) > 21:

            node = node[:21] # match c implementation behavior

        dict.__setitem__(self, k, (node, flag))

    def __iter__(self):

        return iter(sorted(dict.keys(self)))

    def iterkeys(self):

        return iter(sorted(dict.keys(self)))

    def iterentries(self):

        return ((f, e[0], e[1]) for f, e in sorted(self.iteritems()))

    def copy(self):

        c = _lazymanifest('')

        c.update(self)

        return c

    def diff(self, m2, clean=False):

        '''Finds changes between the current manifest and m2.'''

        diff = {}

        for fn, e1 in self.iteritems():

            if fn not in m2:

                diff[fn] = e1, (None, '')

            else:

                e2 = m2[fn]

                if e1 != e2:

                    diff[fn] = e1, e2

                elif clean:

                    diff[fn] = None

        for fn, e2 in m2.iteritems():

            if fn not in self:

                diff[fn] = (None, ''), e2

        return diff

    def filtercopy(self, filterfn):

        c = _lazymanifest('')

        for f, n, fl in self.iterentries():

            if filterfn(f):

                c[f] = n, fl

        return c

    def text(self):

        """Get the full data of this manifest as a bytestring."""

        return _textv1(self.iterentries())

try:

    _lazymanifest = parsers.lazymanifest

except AttributeError:

    pass

class manifestdict(object):

    def __init__(self, data=''):

        if data.startswith('\0'):

            #_lazymanifest can not parse v2

            self._lm = _lazymanifest('')

            for f, n, fl in _parsev2(data):

                self._lm[f] = n, fl

        else:

            self._lm = _lazymanifest(data)

    def __getitem__(self, key):

        return self._lm[key][0]

    def find(self, key):

        return self._lm[key]

    def __len__(self):

        return len(self._lm)

    def __setitem__(self, key, node):

        self._lm[key] = node, self.flags(key, '')

    def __contains__(self, key):

        return key in self._lm

    def __delitem__(self, key):

        del self._lm[key]

    def __iter__(self):

        return self._lm.__iter__()

    def iterkeys(self):

        return self._lm.iterkeys()

    def keys(self):

        return list(self.iterkeys())

    def _intersectfiles(self, files):

        '''make a new lazymanifest with the intersection of self with files

        The algorithm assumes that files is much smaller than self.'''

        ret = manifestdict()

        lm = self._lm

        for fn in files:

            if fn in lm:

                ret._lm[fn] = self._lm[fn]

        return ret

    def filesnotin(self, m2):

        '''Set of files in this manifest that are not in the other'''

        files = set(self)

        files.difference_update(m2)

        return files

    @propertycache

    def _dirs(self):

        return util.dirs(self)

    def dirs(self):

        return self._dirs

    def hasdir(self, dir):

        return dir in self._dirs

    def matches(self, match):

        '''generate a new manifest filtered by the match argument'''

        if match.always():

            return self.copy()

        files = match.files()

        if (len(files) < 100 and (match.isexact() or

            (not match.anypats() and util.all(fn in self for fn in files)))):

            return self._intersectfiles(files)

        lm = manifestdict('')

        lm._lm = self._lm.filtercopy(match)

        return lm

    def diff(self, m2, clean=False):

        '''Finds changes between the current manifest and m2.

        Args:

          m2: the manifest to which this manifest should be compared.

          clean: if true, include files unchanged between these manifests

                 with a None value in the returned dictionary.

        The result is returned as a dict with filename as key and

        values of the form ((n1,fl1),(n2,fl2)), where n1/n2 is the

        nodeid in the current/other manifest and fl1/fl2 is the flag

        in the current/other manifest. Where the file does not exist,

        the nodeid will be None and the flags will be the empty

        string.

        '''

        return self._lm.diff(m2._lm, clean)

    def setflag(self, key, flag):

        self._lm[key] = self[key], flag

    def get(self, key, default=None):

        try:

            return self._lm[key][0]

        except KeyError:

            return default

    def flags(self, key, default=''):

        try:

            return self._lm[key][1]

        except KeyError:

            return default

    def copy(self):

        c = manifestdict('')

        c._lm = self._lm.copy()

        return c

    def iteritems(self):

        return (x[:2] for x in self._lm.iterentries())

    def text(self, usemanifestv2=False):

        if usemanifestv2:

            return _textv2(self._lm.iterentries())

        else:

            # use (probably) native version for v1

            return self._lm.text()

    def fastdelta(self, base, changes):

        """Given a base manifest text as an array.array and a list of changes

        relative to that text, compute a delta that can be used by revlog.

        """

        delta = []

        dstart = None

        dend = None

        dline = [""]

        start = 0

        # zero copy representation of base as a buffer

        addbuf = util.buffer(base)

        # start with a readonly loop that finds the offset of

        # each line and creates the deltas

        for f, todelete in changes:

            # bs will either be the index of the item or the insert point

            start, end = _msearch(addbuf, f, start)

            if not todelete:

                h, fl = self._lm[f]

                l = "%s\0%s%s\n" % (f, revlog.hex(h), fl)

            else:

                if start == end:

                    # item we want to delete was not found, error out

                    raise AssertionError(

                            _("failed to remove %s from manifest") % f)

                l = ""

            if dstart is not None and dstart <= start and dend >= start:

                if dend < end:

                    dend = end

                if l:

                    dline.append(l)

            else:

                if dstart is not None:

                    delta.append([dstart, dend, "".join(dline)])

                dstart = start

                dend = end

                dline = [l]

        if dstart is not None:

            delta.append([dstart, dend, "".join(dline)])

        # apply the delta to the base, and get a delta for addrevision

        deltatext, arraytext = _addlistdelta(base, delta)

        return arraytext, deltatext

def _msearch(m, s, lo=0, hi=None):

    '''return a tuple (start, end) that says where to find s within m.

    If the string is found m[start:end] are the line containing

    that string.  If start == end the string was not found and

    they indicate the proper sorted insertion point.

    m should be a buffer or a string

    s is a string'''

    def advance(i, c):

        while i < lenm and m[i] != c:

            i += 1

        return i

    if not s:

        return (lo, lo)

    lenm = len(m)

    if not hi:

        hi = lenm

    while lo < hi:

        mid = (lo + hi) // 2

        start = mid

        while start > 0 and m[start - 1] != '\n':

            start -= 1

        end = advance(start, '\0')

        if m[start:end] < s:

            # we know that after the null there are 40 bytes of sha1

            # this translates to the bisect lo = mid + 1

            lo = advance(end + 40, '\n') + 1

        else:

            # this translates to the bisect hi = mid

            hi = start

    end = advance(lo, '\0')

    found = m[lo:end]

    if s == found:

        # we know that after the null there are 40 bytes of sha1

        end = advance(end + 40, '\n')

        return (lo, end + 1)

    else:

        return (lo, lo)

def _checkforbidden(l):

    """Check filenames for illegal characters."""

    for f in l:

        if '\n' in f or '\r' in f:

            raise error.RevlogError(