Changeset - 7d0476e1f1dc
[Not reviewed]
beta
0 1 0
Marcin Kuzminski - 13 years ago 2012-12-11 01:15:30
marcin@python-works.com
fixes issue #678 Incorrect diff markup when diff contains >, <, or & symbols
- regex by \W did split &amp; and other to 3 tokens, and escaping was broken
1 file changed with 5 insertions and 3 deletions:
0 comments (0 inline, 0 general)
rhodecode/lib/diffs.py
Show inline comments
 
@@ -148,180 +148,182 @@ class LimitedDiffContainer(object):
 
        self.diff = diff
 
        self.diff_limit = diff_limit
 
        self.cur_diff_size = cur_diff_size
 

	
 
    def __iter__(self):
 
        for l in self.diff:
 
            yield l
 

	
 

	
 
class DiffProcessor(object):
 
    """
 
    Give it a unified or git diff and it returns a list of the files that were
 
    mentioned in the diff together with a dict of meta information that
 
    can be used to render it in a HTML template.
 
    """
 
    _chunk_re = re.compile(r'^@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*)')
 
    _newline_marker = re.compile(r'^\\ No newline at end of file')
 
    _git_header_re = re.compile(r"""
 
        #^diff[ ]--git
 
            [ ]a/(?P<a_path>.+?)[ ]b/(?P<b_path>.+?)\n
 
        (?:^similarity[ ]index[ ](?P<similarity_index>\d+)%\n
 
           ^rename[ ]from[ ](?P<rename_from>\S+)\n
 
           ^rename[ ]to[ ](?P<rename_to>\S+)(?:\n|$))?
 
        (?:^old[ ]mode[ ](?P<old_mode>\d+)\n
 
           ^new[ ]mode[ ](?P<new_mode>\d+)(?:\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|$))?
 
        (?:^---[ ](a/(?P<a_file>.+)|/dev/null)(?:\n|$))?
 
        (?:^\+\+\+[ ](b/(?P<b_file>.+)|/dev/null)(?:\n|$))?
 
    """, re.VERBOSE | re.MULTILINE)
 
    _hg_header_re = re.compile(r"""
 
        #^diff[ ]--git
 
            [ ]a/(?P<a_path>.+?)[ ]b/(?P<b_path>.+?)\n
 
        (?:^similarity[ ]index[ ](?P<similarity_index>\d+)%(?:\n|$))?
 
        (?:^rename[ ]from[ ](?P<rename_from>\S+)\n
 
           ^rename[ ]to[ ](?P<rename_to>\S+)(?:\n|$))?
 
        (?:^old[ ]mode[ ](?P<old_mode>\d+)\n
 
           ^new[ ]mode[ ](?P<new_mode>\d+)(?:\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|$))?
 
        (?:^---[ ](a/(?P<a_file>.+)|/dev/null)(?:\n|$))?
 
        (?:^\+\+\+[ ](b/(?P<b_file>.+)|/dev/null)(?:\n|$))?
 
    """, re.VERBOSE | re.MULTILINE)
 

	
 
    #used for inline highlighter word split
 
    _token_re = re.compile(r'()(&gt;|&lt;|&amp;|\W+?)')
 

	
 
    def __init__(self, diff, vcs='hg', format='gitdiff', diff_limit=None):
 
        """
 
        :param diff:   a text in diff format
 
        :param vcs: type of version controll hg or git
 
        :param format: format of diff passed, `udiff` or `gitdiff`
 
        :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._format = format
 
        self.adds = 0
 
        self.removes = 0
 
        # calculate diff size
 
        self.diff_size = len(diff)
 
        self.diff_limit = diff_limit
 
        self.cur_diff_size = 0
 
        self.parsed = False
 
        self.parsed_diff = []
 
        self.vcs = vcs
 

	
 
        if format == 'gitdiff':
 
            self.differ = self._highlight_line_difflib
 
            self._parser = self._parse_gitdiff
 
        else:
 
            self.differ = self._highlight_line_udiff
 
            self._parser = self._parse_udiff
 

	
 
    def _copy_iterator(self):
 
        """
 
        make a fresh copy of generator, we should not iterate thru
 
        an original as it's needed for repeating operations on
 
        this instance of DiffProcessor
 
        """
 
        self.__udiff, iterator_copy = tee(self.__udiff)
 
        return iterator_copy
 

	
 
    def _escaper(self, string):
 
        """
 
        Escaper for diff escapes special chars and checks the diff limit
 

	
 
        :param string:
 
        :type string:
 
        """
 

	
 
        self.cur_diff_size += len(string)
 

	
 
        # escaper get's 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')
 

	
 
        return safe_unicode(string).replace('&', '&amp;')\
 
                .replace('<', '&lt;')\
 
                .replace('>', '&gt;')
 

	
 
    def _line_counter(self, l):
 
        """
 
        Checks each line and bumps total adds/removes for this diff
 

	
 
        :param l:
 
        """
 
        if l.startswith('+') and not l.startswith('+++'):
 
            self.adds += 1
 
        elif l.startswith('-') and not l.startswith('---'):
 
            self.removes += 1
 
        return safe_unicode(l)
 

	
 
    def _highlight_line_difflib(self, line, next_):
 
        """
 
        Highlight inline changes in both lines.
 
        """
 

	
 
        if line['action'] == 'del':
 
            old, new = line, next_
 
        else:
 
            old, new = next_, line
 

	
 
        oldwords = re.split(r'(\W)', old['line'])
 
        newwords = re.split(r'(\W)', new['line'])
 

	
 
        oldwords = self._token_re.split(old['line'])
 
        newwords = self._token_re.split(new['line'])
 
        sequence = difflib.SequenceMatcher(None, oldwords, newwords)
 

	
 
        oldfragments, newfragments = [], []
 
        for tag, i1, i2, j1, j2 in sequence.get_opcodes():
 
            oldfrag = ''.join(oldwords[i1:i2])
 
            newfrag = ''.join(newwords[j1:j2])
 
            if tag != 'equal':
 
                if oldfrag:
 
                    oldfrag = '<del>%s</del>' % oldfrag
 
                if newfrag:
 
                    newfrag = '<ins>%s</ins>' % newfrag
 
            oldfragments.append(oldfrag)
 
            newfragments.append(newfrag)
 

	
 
        old['line'] = "".join(oldfragments)
 
        new['line'] = "".join(newfragments)
 

	
 
    def _highlight_line_udiff(self, line, next_):
 
        """
 
        Highlight inline changes in both lines.
 
        """
 
        start = 0
 
        limit = min(len(line['line']), len(next_['line']))
 
        while start < limit and line['line'][start] == next_['line'][start]:
 
            start += 1
 
        end = -1
 
        limit -= start
 
        while -end <= limit and line['line'][end] == next_['line'][end]:
 
            end -= 1
 
        end += 1
 
        if start or end:
 
            def do(l):
 
                last = end + len(l['line'])
 
                if l['action'] == 'add':
 
                    tag = 'ins'
 
                else:
 
                    tag = 'del'
 
                l['line'] = '%s<%s>%s</%s>%s' % (
 
                    l['line'][:start],
 
                    tag,
 
                    l['line'][start:last],
 
                    tag,
 
                    l['line'][last:]
 
                )
 
            do(line)
 
            do(next_)
 

	
 
    def _get_header(self, diff_chunk):
0 comments (0 inline, 0 general)