convert/darcs: support changelogs with bytes 0x7F-0xFF (issue2411)

This is a followup to 4481f8a93c7a, which only fixed the conversion of
patches with UTF-8 metadata.

This patch allows a changelog to have any bytes with values
0x7F-0xFF. It parses the XML changelog as Latin-1 and uses
converter_source.recode() to decode the data as UTF-8/Latin-1.

Caveats:

- Since the convert extension doesn't provide any way to specify the
  source encoding, users are still limited to UTF-8 and Latin-1.

- etree will still complain if the changelog has bytes with values
  0x00-0x19. XML only allows printable characters.

hgext/convert/darcs.py

@@ -1,188 +1,200 b''
 # 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
+from mercurial import encoding, util
 import os, shutil, tempfile, re
 
 # The naming drift of ElementTree is fun!
 
 try:
-    from xml.etree.cElementTree import ElementTree
+    from xml.etree.cElementTree import ElementTree, XMLParser
 except ImportError:
     try:
-        from xml.etree.ElementTree import ElementTree
+        from xml.etree.ElementTree import ElementTree, XMLParser
     except ImportError:
         try:
-            from elementtree.cElementTree import ElementTree
+            from elementtree.cElementTree import ElementTree, XMLParser
         except ImportError:
             try:
-                from elementtree.ElementTree import ElementTree
+                from elementtree.ElementTree import ElementTree, XMLParser
             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 so that we can easily skip
         # test-convert-darcs if ElementTree is not around
         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 = {}
 
         # Check darcs repository format
         format = self.format()
         if format:
             if format in ('darcs-1.0', 'hashed'):
                 raise NoRepo(_("%s repository format is unsupported, "
                                "please upgrade") % format)
         else:
             self.ui.warn(_('failed to detect repository format!'))
 
     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 recode(self, s, encoding=None):
+        if isinstance(s, unicode):
+            # XMLParser returns unicode objects for anything it can't
+            # encode into ASCII. We convert them back to str to get
+            # recode's normal conversion behavior.
+            s = s.encode('latin-1')
+        return super(darcs_source, self).recode(s, encoding)
+
     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()
+        # While we are decoding the XML as latin-1 to be as liberal as
+        # possible, etree will still raise an exception if any
+        # non-printable characters are in the XML changelog.
+        parser = XMLParser(encoding='latin-1')
         fp = self._run(cmd, **kwargs)
-        etree.parse(fp)
+        etree.parse(fp, parser=parser)
         self.checkexit(fp.close())
         return etree.getroot()
 
     def format(self):
         output, status = self.run('show', 'repo', no_files=True,
                                   repodir=self.path)
         self.checkexit(status)
         m = re.search(r'^\s*Format:\s*(.*)$', output, re.MULTILINE)
         if not m:
             return None
         return ','.join(sorted(f.strip() for f in m.group(1).split(',')))
 
     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

tests/test-convert-darcs

@@ -1,83 +1,90 b''
 #!/bin/sh
 
 "$TESTDIR/hghave" darcs || exit 80
 
 echo "[extensions]" >> $HGRCPATH
 echo "convert=" >> $HGRCPATH
 echo 'graphlog =' >> $HGRCPATH
 
 DARCS_EMAIL='[e-mail]'; export DARCS_EMAIL
 HOME=`pwd`/do_not_use_HOME_darcs; export HOME
 
 # skip if we can't import elementtree
 mkdir dummy
 mkdir dummy/_darcs
 if hg convert dummy 2>&1 | grep ElementTree > /dev/null; then
     echo 'skipped: missing feature: elementtree module'
     exit 80
 fi
 
 echo '% try converting darcs1 repository'
 hg convert -s darcs "$TESTDIR/darcs/darcs1" 2>&1 | grep darcs-1.0
 
 echo % initialize darcs repo
 mkdir darcs-repo
 cd darcs-repo
 darcs init
 echo a > a
 darcs record -a -l -m p0
 cd ..
 
 echo % branch and update
 darcs get darcs-repo darcs-clone >/dev/null
 cd darcs-clone
 echo c >> a
 echo c > c
 darcs record -a -l -m p1.1
 cd ..
 
 echo % update source
 cd darcs-repo
 echo b >> a
 echo b > b
 darcs record -a -l -m p1.2
 
 echo % merge branch
 darcs pull -a ../darcs-clone
 sleep 1
 echo e > a
 echo f > f
 mkdir dir
 echo d > dir/d
 echo d > dir/d2
 darcs record -a -l -m p2
 
 echo % test file and directory move
 darcs mv f ff
 # Test remove + move
 darcs remove dir/d2
 rm dir/d2
 darcs mv dir dir2
 darcs record -a -l -m p3
 
 echo % test utf-8 commit message and author
 echo g > g
 # darcs is encoding agnostic, so it takes whatever bytes it's given
 darcs record -a -l -m 'p4: desc ñ' -A 'author ñ'
 
+echo % test latin-1 commit message
+echo h > h
+printf "p5: desc " > ../p5
+python -c 'print "".join([chr(i) for i in range(128, 256)])' >> ../p5
+darcs record -a -l --logfile ../p5
+
 glog()
 {
-    HGENCODING=utf-8 hg glog --template '{rev} "{desc|firstline}" ({author}) files: {files}\n' "$@"
+    hg glog --template '{rev} "{desc|firstline}" ({author}) files: {files}\n' "$@"
 }
 
 cd ..
 hg convert darcs-repo darcs-repo-hg
 # The converter does not currently handle patch conflicts very well.
 # When they occur, it reverts *all* changes and moves forward,
 # letting the conflict resolving patch fix collisions.
 # Unfortunately, non-conflicting changes, like the addition of the
 # "c" file in p1.1 patch are reverted too.
 # Just to say that manifest not listing "c" here is a bug.
-glog -R darcs-repo-hg
+HGENCODING=latin-1 glog -R darcs-repo-hg -r 6 | "$TESTDIR"/printrepr.py
+HGENCODING=utf-8 glog -R darcs-repo-hg -r 0:5 | "$TESTDIR"/printrepr.py
 hg up -q -R darcs-repo-hg
 hg -R darcs-repo-hg manifest --debug

tests/test-convert-darcs.out

@@ -1,45 +1,51 b''
 % try converting darcs1 repository
 darcs-1.0 repository format is unsupported, please upgrade
 % initialize darcs repo
 Finished recording patch 'p0'
 % branch and update
 Finished recording patch 'p1.1'
 % update source
 Finished recording patch 'p1.2'
 % merge branch
 Backing up ./a(-darcs-backup0)
 We have conflicts in the following files:
 ./a
 Finished pulling and applying.
 Finished recording patch 'p2'
 % test file and directory move
 Finished recording patch 'p3'
 % test utf-8 commit message and author
 Finished recording patch 'p4: desc ñ'
+% test latin-1 commit message
+Finished recording patch 'p5: desc ��������������������������������������������������������������������������������������������������������������������������������'
 initializing destination darcs-repo-hg repository
 scanning source...
 sorting...
 converting...
-5 p0
-4 p1.2
-3 p1.1
-2 p2
-1 p3
-0 p4: desc ?
-o  5 "p4: desc ñ" (author ñ) files: g
+6 p0
+5 p1.2
+4 p1.1
+3 p2
+2 p3
+1 p4: desc ?
+0 p5: desc ????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????
+o  6 "p5: desc \x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff" ([e-mail]) files: h
+|
+o  5 "p4: desc \xc3\xb1" (author \xc3\xb1) files: g
 |
 o  4 "p3" ([e-mail]) files: dir/d dir/d2 dir2/d f ff
 |
 o  3 "p2" ([e-mail]) files: a dir/d dir/d2 f
 |
 o  2 "p1.1" ([e-mail]) files:
 |
 o  1 "p1.2" ([e-mail]) files: a b
 |
 o  0 "p0" ([e-mail]) files: a
 
 7225b30cdf38257d5cc7780772c051b6f33e6d6b 644   a
 1e88685f5ddec574a34c70af492f95b6debc8741 644   b
 37406831adc447ec2385014019599dfec953c806 644   dir2/d
 b783a337463792a5c7d548ad85a7d3253c16ba8c 644   ff
 0973eb1b2ecc4de7fafe7447ce1b7462108b4848 644   g
+fe6f8b4f507fe3eb524c527192a84920a4288dac 644   h