comments.update((st.comment_id, st.comment)

                                for st in ChangesetStatusModel()

                                .get_statuses(c.db_repo.repo_id,

                                              changeset.raw_id, with_revisions=True)

                                if st.comment_id is not None)

                inlines = ChangesetCommentsModel() \

                            .get_inline_comments(c.db_repo.repo_id,

                                                 revision=changeset.raw_id)

                c.inline_comments.extend(inlines)

            cs2 = changeset.raw_id

            cs1 = changeset.parents[0].raw_id if changeset.parents else EmptyChangeset().raw_id

            context_lcl = get_line_ctx('', request.GET)

            ign_whitespace_lcl = get_ignore_ws('', request.GET)

            raw_diff = c.db_repo_scm_instance.get_diff(cs1, cs2,

                ignore_whitespace=ign_whitespace_lcl, context=context_lcl)

            diff_limit = None if c.fulldiff else self.cut_off_limit

            file_diff_data = []

            if method == 'show':

                diff_processor = diffs.DiffProcessor(raw_diff,

                                                     vcs=c.db_repo_scm_instance.alias,

                                                     diff_limit=diff_limit)

                c.limited_diff = False

                    c.limited_diff = True

                    st = f['stats']

                    c.lines_added += st['added']

                    c.lines_deleted += st['deleted']

                    filename = f['filename']

                    fid = h.FID(changeset.raw_id, filename)

                    url_fid = h.FID('', filename)

                    diff = diff_processor.as_html(enable_comments=enable_comments,

                                                  parsed_lines=[f])

                    file_diff_data.append((fid, url_fid, f['operation'], f['old_filename'], filename, diff, st))

            else:

                # downloads/raw we only need RAW diff nothing else

                file_diff_data.append(('', None, None, None, raw_diff, None))

            c.changes[changeset.raw_id] = (cs1, cs2, file_diff_data)

        # sort comments in creation order

        c.comments = [com for com_id, com in sorted(comments.items())]

        # count inline comments

        for __, lines in c.inline_comments:

            for comments in lines.values():

                c.inline_cnt += len(comments)

        if len(c.cs_ranges) == 1:

            c.changeset = c.cs_ranges[0]

                msg = _('Multiple merge ancestors found for merge compare')

            if rev1 is None:

                h.flash(msg, category='error')

                log.error(msg)

                raise HTTPNotFound

            # case we want a simple diff without incoming changesets,

            # previewing what will be merged.

            # Make the diff on the other repo (which is known to have other_rev)

            log.debug('Using ancestor %s as rev1 instead of %s',

                      rev1, c.a_rev)

            org_repo = other_repo

        else: # comparing tips, not necessarily linearly related

            if org_repo != other_repo:

                # TODO: we could do this by using hg unionrepo

                log.error('cannot compare across repos %s and %s', org_repo, other_repo)

                h.flash(_('Cannot compare repositories without using common ancestor'), category='error')

                raise HTTPBadRequest

            rev1 = c.a_rev

        diff_limit = None if fulldiff else self.cut_off_limit

        log.debug('running diff between %s and %s in %s',

                  rev1, c.cs_rev, org_repo.scm_instance.path)

                                      ignore_whitespace=ignore_whitespace,

                                      context=line_context)

        c.limited_diff = False

            c.limited_diff = True

        c.file_diff_data = []

        c.lines_added = 0

        c.lines_deleted = 0

            st = f['stats']

            c.lines_added += st['added']

            c.lines_deleted += st['deleted']

            filename = f['filename']

            fid = h.FID('', filename)

            diff = diff_processor.as_html(enable_comments=False,

                                          parsed_lines=[f])

            c.file_diff_data.append((fid, None, f['operation'], f['old_filename'], filename, diff, st))

        return render('compare/compare_diff.html')

    @LoginRequired(api_access=True)

    @HasRepoPermissionLevelDecorator('read')

    def _before(self, *args, **kwargs):

        super(FeedController, self)._before(*args, **kwargs)

    def _get_title(self, cs):

        return h.shorter(cs.message, 160)

    def __get_desc(self, cs):

        desc_msg = [(_('%s committed on %s')

                     % (h.person(cs.author), h.fmt_date(cs.date))) + '<br/>']

        # branches, tags, bookmarks

        if cs.branch:

            desc_msg.append('branch: %s<br/>' % cs.branch)

        for book in cs.bookmarks:

            desc_msg.append('bookmark: %s<br/>' % book)

        for tag in cs.tags:

            desc_msg.append('tag: %s<br/>' % tag)

        changes = []

        diff_limit = safe_int(CONFIG.get('rss_cut_off_limit', 32 * 1024))

        raw_diff = cs.diff()

        diff_processor = DiffProcessor(raw_diff,

        limited_diff = False

            limited_diff = True

            st.update({'added': st['stats']['added'],

                       'removed': st['stats']['deleted']})

            changes.append('\n %(operation)s %(filename)s '

                           '(%(added)s lines added, %(removed)s lines removed)'

                            % st)

        if limited_diff:

            changes = changes + ['\n ' +

                                 _('Changeset was too big and was cut off...')]

        # rev link

        _url = h.canonical_url('changeset_home', repo_name=c.db_repo.repo_name,

                   revision=cs.raw_id)

        desc_msg.append('changeset: <a href="%s">%s</a>' % (_url, cs.raw_id[:8]))

        desc_msg.append('<pre>')

        desc_msg.append(h.urlify_text(cs.message))

        desc_msg.append('\n')

        desc_msg.extend(changes)

        if str2bool(CONFIG.get('rss_include_diff', False)):

            desc_msg.append('\n\n')

            desc_msg.append(raw_diff)

        desc_msg.append('</pre>')

        return map(safe_unicode, desc_msg)

                c.cs_repo.scm_instance.get_changeset(c.cs_rev) # check it exists - raise ChangesetDoesNotExistError if not

                c.update_msg = _("Git pull requests don't support iterating yet.")

        except ChangesetDoesNotExistError:

            c.update_msg = _('Error: some changesets not found when displaying pull request from %s.') % c.cs_rev

        c.avail_revs = avail_revs

        c.avail_cs = [org_scm_instance.get_changeset(r) for r in avail_show]

        c.avail_jsdata = graph_data(org_scm_instance, avail_show)

        raw_ids = [x.raw_id for x in c.cs_ranges]

        c.cs_comments = c.cs_repo.get_comments(raw_ids)

        c.cs_statuses = c.cs_repo.statuses(raw_ids)

        ignore_whitespace = request.GET.get('ignorews') == '1'

        line_context = safe_int(request.GET.get('context'), 3)

        c.ignorews_url = _ignorews_url

        c.context_url = _context_url

        fulldiff = request.GET.get('fulldiff')

        diff_limit = None if fulldiff else self.cut_off_limit

        # we swap org/other ref since we run a simple diff on one repo

        log.debug('running diff between %s and %s in %s',

                  c.a_rev, c.cs_rev, org_scm_instance.path)

        try:

                                                ignore_whitespace=ignore_whitespace,

                                                context=line_context)

        except ChangesetDoesNotExistError:

        c.limited_diff = False

            c.limited_diff = True

        c.file_diff_data = []

        c.lines_added = 0

        c.lines_deleted = 0

            st = f['stats']

            c.lines_added += st['added']

            c.lines_deleted += st['deleted']

            filename = f['filename']

            fid = h.FID('', filename)

            diff = diff_processor.as_html(enable_comments=True,

                                          parsed_lines=[f])

            c.file_diff_data.append((fid, None, f['operation'], f['old_filename'], filename, diff, st))

        # inline comments

        c.inline_cnt = 0

        c.inline_comments = cc_model.get_inline_comments(

                                c.db_repo.repo_id,

                                pull_request=pull_request_id)

        # count inline comments

        for __, lines in c.inline_comments:

            for comments in lines.values():

                c.inline_cnt += len(comments)

        # comments

        c.comments = cc_model.get_comments(c.db_repo.repo_id, pull_request=pull_request_id)

        # (badly named) pull-request status calculation based on reviewer votes

        (c.pull_request_reviewers,

         c.pull_request_pending_reviewers,

                </tr>

              </table>''' % html

def wrapped_diff(filenode_old, filenode_new, diff_limit=None,

                ignore_whitespace=True, line_context=3,

                enable_comments=False):

    """

    Returns a file diff wrapped into a table.

    Checks for diff_limit and presents a message if the diff is too big.

    """

    if filenode_old is None:

        filenode_old = FileNode(filenode_new.path, '', EmptyChangeset())

    op = None

    a_path = filenode_old.path # default, might be overriden by actual rename in diff

    if filenode_old.is_binary or filenode_new.is_binary:

        diff = wrap_to_table(_('Binary file'))

        stats = (0, 0)

    elif diff_limit != -1 and (

            diff_limit is None or

            (filenode_old.size < diff_limit and filenode_new.size < diff_limit)):

                                ignore_whitespace=ignore_whitespace,

                                context=line_context)

            op = f['operation']

            a_path = f['old_filename']

        diff = diff_processor.as_html(enable_comments=enable_comments)

        stats = diff_processor.stat()

    else:

        diff = wrap_to_table(_('Changeset was too big and was cut off, use '

                               'diff menu to display this diff'))

        stats = (0, 0)

    if not diff:

        submodules = filter(lambda o: isinstance(o, SubModuleNode),

                            [filenode_new, filenode_old])

        if submodules:

            diff = wrap_to_table(escape('Submodule %r' % submodules[0]))

        else:

            diff = wrap_to_table(_('No changes detected'))

    cs1 = filenode_old.changeset.raw_id

    cs2 = filenode_new.changeset.raw_id

    return cs1, cs2, a_path, diff, stats, op

    _hg_header_re = re.compile(r"""

        ^diff[ ]--git

            [ ]a/(?P<a_path>.+?)[ ]b/(?P<b_path>.+?)\n

        (?:^old[ ]mode[ ](?P<old_mode>\d+)\n

           ^new[ ]mode[ ](?P<new_mode>\d+)(?:\n|$))?

        (?:^similarity[ ]index[ ](?P<similarity_index>\d+)%(?:\n|$))?

        (?:^rename[ ]from[ ](?P<rename_from>.+)\n

           ^rename[ ]to[ ](?P<rename_to>.+)(?:\n|$))?

        (?:^copy[ ]from[ ](?P<copy_from>.+)\n

           ^copy[ ]to[ ](?P<copy_to>.+)(?:\n|$))?

        (?:^new[ ]file[ ]mode[ ](?P<new_file_mode>.+)(?:\n|$))?

        (?:^deleted[ ]file[ ]mode[ ](?P<deleted_file_mode>.+)(?:\n|$))?

        (?:^index[ ](?P<a_blob_id>[0-9A-Fa-f]+)

            \.\.(?P<b_blob_id>[0-9A-Fa-f]+)[ ]?(?P<b_mode>.+)?(?:\n|$))?

        (?:^(?P<bin_patch>GIT[ ]binary[ ]patch)(?:\n|$))?

        (?:^---[ ](a/(?P<a_file>.+?)|/dev/null)\t?(?:\n|$))?

        (?:^\+\+\+[ ](b/(?P<b_file>.+?)|/dev/null)\t?(?:\n|$))?

    """, re.VERBOSE | re.MULTILINE)

    # Used for inline highlighter word split, must match the substitutions in _escaper

    _token_re = re.compile(r'()(&amp;|&lt;|&gt;|<u>\t</u>|<u class="cr"></u>| <i></i>|\W+?)')

    _escape_re = re.compile(r'(&)|(<)|(>)|(\t)|(\r)|(?<=.)( \n| $)')

        """

        :param diff:   a text in diff format

        :param vcs: type of version control hg or git

        :param diff_limit: define the size of diff that is considered "big"

            based on that parameter cut off will be triggered, set to None

            to show full diff

        """

        if not isinstance(diff, basestring):

            raise Exception('Diff must be a basestring got %s instead' % type(diff))

        self._diff = diff

        self.adds = 0

        self.removes = 0

        # calculate diff size

        self.diff_limit = diff_limit

        self.cur_diff_size = 0

        self.vcs = vcs

    def _escaper(self, string):

        """

        Do HTML escaping/markup and check the diff limit

        """

        self.cur_diff_size += len(string)

        # escaper gets iterated on each .next() call and it checks if each

        # parsed line doesn't exceed the diff limit

        if self.diff_limit is not None and self.cur_diff_size > self.diff_limit:

            raise DiffLimitExceeded('Diff Limit Exceeded')

        def substitute(m):

            groups = m.groups()

            if groups[0]:

                return '&'

            if groups[1]:

                return '<'

            if groups[2]:

                return '>'

            if groups[3]:

                return '<u>\t</u>'

            if groups[4]:

                return '<u class="cr"></u>'

        Parses a Git diff for a single file (header and chunks) and returns a tuple with:

        1. A dict with meta info:

            a_path, b_path, similarity_index, rename_from, rename_to,

            old_mode, new_mode, new_file_mode, deleted_file_mode,

            a_blob_id, b_blob_id, b_mode, a_file, b_file

        2. An iterator yielding lines with simple HTML markup.

        """

        match = None

        if self.vcs == 'git':

            match = self._git_header_re.match(diff_chunk)

        elif self.vcs == 'hg':

            match = self._hg_header_re.match(diff_chunk)

        if match is None:

            raise Exception('diff not recognized as valid %s diff' % self.vcs)

        meta_info = match.groupdict()

        rest = diff_chunk[match.end():]

        if rest and not rest.startswith('@') and not rest.startswith('literal ') and not rest.startswith('delta '):

            raise Exception('cannot parse %s diff header: %r followed by %r' % (self.vcs, diff_chunk[:match.end()], rest[:1000]))

        diff_lines = (self._escaper(m.group(0)) for m in re.finditer(r'.*\n|.+$', rest)) # don't split on \r as str.splitlines do

        return meta_info, diff_lines

        """Parse self._diff and return a list of dicts with meta info and chunks for each file.

        If diff is truncated, wrap it in LimitedDiffContainer.

        Optionally, do an extra pass and to extra markup of one-liner changes.

        """

        _files = [] # list of dicts with meta info and chunks

        diff_container = lambda arg: arg

        starts = [m.start() for m in self._diff_git_re.finditer(self._diff)]

        starts.append(len(self._diff))

        for start, end in zip(starts, starts[1:]):

            head, diff_lines = self._get_header(buffer(self._diff, start, end - start))

            op = None

            stats = {

                'added': 0,

                'deleted': 0,

                'binary': False,

                'ops': {},

            }

            if head['deleted_file_mode']:

                op = 'D'

                stats['binary'] = True

        return chunks, added, deleted

    def _safe_id(self, idstring):

        """Make a string safe for including in an id attribute.

        The HTML spec says that id attributes 'must begin with

        a letter ([A-Za-z]) and may be followed by any number

        of letters, digits ([0-9]), hyphens ("-"), underscores

        ("_"), colons (":"), and periods (".")'. These regexps

        are slightly over-zealous, in that they remove colons

        and periods unnecessarily.

        Whitespace is transformed into underscores, and then

        anything which is not a hyphen or a character that

        matches \w (alphanumerics and underscore) is removed.

        """

        # Transform all whitespace to underscore

        idstring = re.sub(r'\s', "_", idstring)

        # Remove everything that is not a hyphen or a member of \w

        idstring = re.sub(r'(?!-)\W', "", idstring).lower()

        return idstring

    def as_html(self, table_class='code-difftable', line_class='line',

                old_lineno_class='lineno old', new_lineno_class='lineno new',

                no_lineno_class='lineno',

                code_class='code', enable_comments=False, parsed_lines=None):

        """

        Return given diff as html table with customized css classes

        """

        def _link_to_if(condition, label, url):

            """

            Generates a link if condition is meet or just the label if not.

            """

            if condition:

                return '''<a href="%(url)s">%(label)s</a>''' % {

                    'url': url,

                    'label': label

                }

            else:

                return label

        if parsed_lines:

            diff_lines = parsed_lines

        _html_empty = True

        _html = []

        _html.append('''<table class="%(table_class)s">\n''' % {

            'table_class': table_class

        })

        for diff in diff_lines:

            for line in diff['chunks']:

                _html_empty = False

                for change in line:

                    _html.append('''<tr class="%(lc)s %(action)s">\n''' % {

                        'lc': line_class,

                        'action': change['action']

                    })

                    anchor_old_id = ''

                    anchor_new_id = ''

                    anchor_old = "%(filename)s_o%(oldline_no)s" % {

                        'filename': self._safe_id(diff['filename']),

                        'oldline_no': change['old_lineno']

                    }

                    anchor_new = "%(filename)s_n%(oldline_no)s" % {

    ],

    'hg_diff_copy_chmod_and_edit_file.diff': [

        ('file5', 'M',

         {'added': 2,

          'deleted': 1,

          'binary': False,

          'ops': {COPIED_FILENODE: 'file copied from file4 to file5',

                  CHMOD_FILENODE: 'modified file chmod 100755 => 100644',

                  MOD_FILENODE: 'modified file'}}),

    ],

    'hg_diff_rename_space_cr.diff': [

        ('oh yes', 'R',

         {'added': 3,

          'deleted': 2,

          'binary': False,

          'ops': {RENAMED_FILENODE: 'file renamed from oh no to oh yes'}}),

    ],

}

class TestDiffLib(TestController):

    @parametrize('diff_fixture', DIFF_FIXTURES)

    def test_diff(self, diff_fixture):

        vcs = 'hg'

        if diff_fixture.startswith('git_'):

            vcs = 'git'

        expected_data = DIFF_FIXTURES[diff_fixture]

        assert expected_data == data

    def test_diff_markup(self):

        assert not chunks[0]

        #from pprint import pprint; pprint(chunks[1])

        l = ['\n']

        for d in chunks[1]:

            l.append('%(action)-7s %(new_lineno)3s %(old_lineno)3s %(line)r\n' % d)

        s = ''.join(l)

        print s

        assert s == r'''

context ... ... u'@@ -51,6 +51,13 @@\n'

unmod    51  51 u'<u>\t</u>begin();\n'

unmod    52  52 u'<u>\t</u>\n'

add      53     u'<u>\t</u>int foo;<u class="cr"></u>\n'

add      54     u'<u>\t</u>int bar; <u class="cr"></u>\n'

add      55     u'<u>\t</u>int baz;<u>\t</u><u class="cr"></u>\n'

add      56     u'<u>\t</u>int space; <i></i>'

add      57     u'<u>\t</u>int tab;<u>\t</u>\n'

add      58     u'<u>\t</u>\n'

unmod    59  53 u' <i></i>'

del          54 u'<u>\t</u>#define MAX_STEPS (48)\n'

add      60     u'<u>\t</u><u class="cr"></u>\n'

add      61     u'<u>\t</u>#define MAX_STEPS (64)<u class="cr"></u>\n'

unmod    62  55 u'\n'

del          56 u'<u>\t</u>#define MIN_STEPS (<del>48</del>)\n'

add      63     u'<u>\t</u>#define MIN_STEPS (<ins>42</ins>)\n'