hgext/convert/darcs.py

convert/darcs: handle non-ASCII metadata in darcs changelog (issue2354)

Given a commit author or message with non-ASCII characters in a darcs
repo, convert would raise a UnicodeEncodeError when adding changesets
to the hg changelog.

This happened because etree returns back unicode objects for any text
it can't encode into ASCII. convert was passing these objects to
changelog.add(), which would then attempt encoding.fromlocal() on
them.

This patch ensures converter_source.recode() is called on each piece
of commit data returned by etree.

(Also note that darcs is currently encoding agnostic and will print
out whatever is in a patch's metadata byte-for-byte, even in the XML
changelog.)

# darcs.py - darcs support for the convert extension
#
#  Copyright 2007-2009 Matt Mackall <[e-mail]> and others
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

from common import NoRepo, checktool, commandline, commit, converter_source
from mercurial.i18n import _
from mercurial import util
import os, shutil, tempfile

# The naming drift of ElementTree is fun!

try:
    from xml.etree.cElementTree import ElementTree
except ImportError:
    try:
        from xml.etree.ElementTree import ElementTree
    except ImportError:
        try:
            from elementtree.cElementTree import ElementTree
        except ImportError:
            try:
                from elementtree.ElementTree import ElementTree
            except ImportError:
                ElementTree = None

class darcs_source(converter_source, commandline):
    def __init__(self, ui, path, rev=None):
        converter_source.__init__(self, ui, path, rev=rev)
        commandline.__init__(self, ui, 'darcs')

        # check for _darcs, ElementTree, _darcs/inventory so that we can
        # easily skip test-convert-darcs if ElementTree is not around
        if not os.path.exists(os.path.join(path, '_darcs', 'inventories')):
            raise NoRepo(_("%s does not look like a darcs repository") % path)

        if not os.path.exists(os.path.join(path, '_darcs')):
            raise NoRepo(_("%s does not look like a darcs repository") % path)

        checktool('darcs')
        version = self.run0('--version').splitlines()[0].strip()
        if version < '2.1':
            raise util.Abort(_('darcs version 2.1 or newer needed (found %r)') %
                             version)

        if ElementTree is None:
            raise util.Abort(_("Python ElementTree module is not available"))

        self.path = os.path.realpath(path)

        self.lastrev = None
        self.changes = {}
        self.parents = {}
        self.tags = {}

    def before(self):
        self.tmppath = tempfile.mkdtemp(
            prefix='convert-' + os.path.basename(self.path) + '-')
        output, status = self.run('init', repodir=self.tmppath)
        self.checkexit(status)

        tree = self.xml('changes', xml_output=True, summary=True,
                        repodir=self.path)
        tagname = None
        child = None
        for elt in tree.findall('patch'):
            node = elt.get('hash')
            name = elt.findtext('name', '')
            if name.startswith('TAG '):
                tagname = name[4:].strip()
            elif tagname is not None:
                self.tags[tagname] = node
                tagname = None
            self.changes[node] = elt
            self.parents[child] = [node]
            child = node
        self.parents[child] = []

    def after(self):
        self.ui.debug('cleaning up %s\n' % self.tmppath)
        shutil.rmtree(self.tmppath, ignore_errors=True)

    def xml(self, cmd, **kwargs):
        # NOTE: darcs is currently encoding agnostic and will print
        # patch metadata byte-for-byte, even in the XML changelog.
        etree = ElementTree()
        fp = self._run(cmd, **kwargs)
        etree.parse(fp)
        self.checkexit(fp.close())
        return etree.getroot()

    def manifest(self):
        man = []
        output, status = self.run('show', 'files', no_directories=True,
                                  repodir=self.tmppath)
        self.checkexit(status)
        for line in output.split('\n'):
            path = line[2:]
            if path:
                man.append(path)
        return man

    def getheads(self):
        return self.parents[None]

    def getcommit(self, rev):
        elt = self.changes[rev]
        date = util.strdate(elt.get('local_date'), '%a %b %d %H:%M:%S %Z %Y')
        desc = elt.findtext('name') + '\n' + elt.findtext('comment', '')
        # etree can return unicode objects for name, comment, and author,
        # so recode() is used to ensure str objects are emitted.
        return commit(author=self.recode(elt.get('author')),
                      date=util.datestr(date),
                      desc=self.recode(desc).strip(),
                      parents=self.parents[rev])

    def pull(self, rev):
        output, status = self.run('pull', self.path, all=True,
                                  match='hash %s' % rev,
                                  no_test=True, no_posthook=True,
                                  external_merge='/bin/false',
                                  repodir=self.tmppath)
        if status:
            if output.find('We have conflicts in') == -1:
                self.checkexit(status, output)
            output, status = self.run('revert', all=True, repodir=self.tmppath)
            self.checkexit(status, output)

    def getchanges(self, rev):
        copies = {}
        changes = []
        man = None
        for elt in self.changes[rev].find('summary').getchildren():
            if elt.tag in ('add_directory', 'remove_directory'):
                continue
            if elt.tag == 'move':
                if man is None:
                    man = self.manifest()
                source, dest = elt.get('from'), elt.get('to')
                if source in man:
                    # File move
                    changes.append((source, rev))
                    changes.append((dest, rev))
                    copies[dest] = source
                else:
                    # Directory move, deduce file moves from manifest
                    source = source + '/'
                    for f in man:
                        if not f.startswith(source):
                            continue
                        fdest = dest + '/' + f[len(source):]
                        changes.append((f, rev))
                        changes.append((fdest, rev))
                        copies[fdest] = f
            else:
                changes.append((elt.text.strip(), rev))
        self.pull(rev)
        self.lastrev = rev
        return sorted(changes), copies

    def getfile(self, name, rev):
        if rev != self.lastrev:
            raise util.Abort(_('internal calling inconsistency'))
        path = os.path.join(self.tmppath, name)
        data = open(path, 'rb').read()
        mode = os.lstat(path).st_mode
        mode = (mode & 0111) and 'x' or ''
        return data, mode

    def gettags(self):
        return self.tags