import: parse email messages

mercurial/commands.py

@@ -12,7 +12,7 b' demandload(globals(), "os re sys signal '
 demandload(globals(), "fancyopts ui hg util lock revlog templater bundlerepo")
 demandload(globals(), "fnmatch mdiff random signal tempfile time")
 demandload(globals(), "traceback errno socket version struct atexit sets bz2")
-demandload(globals(), "archival changegroup")
+demandload(globals(), "archival cStringIO changegroup email.Parser")
 demandload(globals(), "hgweb.server sshserver")
 
 class UnknownCommand(Exception):
@@ -1719,11 +1719,15 b' def import_(ui, repo, patch1, *patches, '
     If there are outstanding changes in the working directory, import
     will abort unless given the -f flag.
 
-    If a patch looks like a mail message (its first line starts with
-    "From " or looks like an RFC822 header), it will not be applied
-    unless the -f option is used.  The importer neither parses nor
-    discards mail headers, so use -f only to override the "mailness"
-    safety check, not to import a real mail message.
+    You can import a patch straight from a mail message.  Even patches
+    as attachments work (body part must be type text/plain or
+    text/x-patch to be used).  Sender and subject line of email
+    message are used as default committer and commit message.  Any
+    text/plain body part before first diff is added to commit message.
+
+    If imported patch was generated by hg export, user and description
+    from patch override values from message headers and body.  Values
+    given on command line with -m and -u override these.
 
     To read a patch from standard input, use patch name "-".
     """
@@ -1739,79 +1743,93 b' def import_(ui, repo, patch1, *patches, '
 
     # attempt to detect the start of a patch
     # (this heuristic is borrowed from quilt)
-    diffre = re.compile(r'(?:Index:[ \t]|diff[ \t]|RCS file: |' +
+    diffre = re.compile(r'^(?:Index:[ \t]|diff[ \t]|RCS file: |' +
                         'retrieving revision [0-9]+(\.[0-9]+)*$|' +
-                        '(---|\*\*\*)[ \t])')
+                        '(---|\*\*\*)[ \t])', re.MULTILINE)
 
     for patch in patches:
         pf = os.path.join(d, patch)
 
-        message = []
+        message = None
         user = None
         date = None
         hgpatch = False
+
+        p = email.Parser.Parser()
         if pf == '-':
-            f = sys.stdin
-            fd, tmpname = tempfile.mkstemp(prefix='hg-patch-')
-            pf = tmpname
-            tmpfp = os.fdopen(fd, 'w')
+            msg = p.parse(sys.stdin)
             ui.status(_("applying patch from stdin\n"))
         else:
-            f = open(pf)
-            tmpfp, tmpname = None, None
+            msg = p.parse(file(pf))
             ui.status(_("applying %s\n") % patch)
+
+        fd, tmpname = tempfile.mkstemp(prefix='hg-patch-')
+        tmpfp = os.fdopen(fd, 'w')
         try:
-            while True:
-                line = f.readline()
-                if not line: break
-                if tmpfp: tmpfp.write(line)
-                line = line.rstrip()
-                if (not message and not hgpatch and
-                       mailre.match(line) and not opts['force']):
-                    if len(line) > 35:
-                        line = line[:32] + '...'
-                    raise util.Abort(_('first line looks like a '
-                                       'mail header: ') + line)
-                if diffre.match(line):
+            message = msg['Subject']
+            if message:
+                message = message.replace('\n\t', ' ')
+                ui.debug('Subject: %s\n' % message)
+            user = msg['From']
+            if user:
+                ui.debug('From: %s\n' % user)
+            diffs_seen = 0
+            ok_types = ('text/plain', 'text/x-patch')
+            for part in msg.walk():
+                content_type = part.get_content_type()
+                ui.debug('Content-Type: %s\n' % content_type)
+                if content_type not in ok_types:
+                    continue
+                payload = part.get_payload(decode=True)
+                m = diffre.search(payload)
+                if m:
+                    ui.debug(_('found patch at byte %d\n') % m.start(0))
+                    diffs_seen += 1
+                    hgpatch = False
+                    fp = cStringIO.StringIO()
+                    for line in payload[:m.start(0)].splitlines():
+                        if line.startswith('# HG changeset patch'):
+                            ui.debug(_('patch generated by hg export\n'))
+                            hgpatch = True
+                        elif hgpatch:
+                            if line.startswith('# User '):
+                                user = line[7:]
+                                ui.debug('From: %s\n' % user)
+                            elif line.startswith("# Date "):
+                                date = line[7:]
+                        if not line.startswith('# '):
+                            fp.write(line)
+                            fp.write('\n')
+                            hgpatch = False
+                    message = fp.getvalue() or message
                     if tmpfp:
-                        for chunk in util.filechunkiter(f):
-                            tmpfp.write(chunk)
-                    break
-                elif hgpatch:
-                    # parse values when importing the result of an hg export
-                    if line.startswith("# User "):
-                        user = line[7:]
-                        ui.debug(_('User: %s\n') % user)
-                    elif line.startswith("# Date "):
-                        date = line[7:]
-                    elif not line.startswith("# ") and line:
-                        message.append(line)
-                        hgpatch = False
-                elif line == '# HG changeset patch':
-                    hgpatch = True
-                    message = []       # We may have collected garbage
-                elif message or line:
-                    message.append(line)
+                        tmpfp.write(payload)
+                        if not payload.endswith('\n'):
+                            tmpfp.write('\n')
+                elif not diffs_seen and message and content_type == 'text/plain':
+                    message += '\n' + payload
 
             if opts['message']:
                 # pickup the cmdline msg
                 message = opts['message']
             elif message:
                 # pickup the patch msg
-                message = '\n'.join(message).rstrip()
+                message = message.strip()
             else:
                 # launch the editor
                 message = None
             ui.debug(_('message:\n%s\n') % message)
 
-            if tmpfp: tmpfp.close()
-            files = util.patch(strip, pf, ui)
-
+            tmpfp.close()
+            if not diffs_seen:
+                raise util.Abort(_('no diffs found'))
+
+            files = util.patch(strip, tmpname, ui)
             if len(files) > 0:
                 addremove_lock(ui, repo, files, {})
             repo.commit(files, message, user, date)
         finally:
-            if tmpname: os.unlink(tmpname)
+            os.unlink(tmpname)
 
 def incoming(ui, repo, source="default", **opts):
     """show new changesets found in source