kallithea Changeset - 99ffd118f6af

Changeset - 99ffd118f6af

Parent rev.

Child rev.

[Not reviewed]

default

0 1 0

Mads Kiilerich (mads) - 5 years ago 2021-02-16 22:54:27
mads@kiilerich.com

diffs: DiffParser should only HTML escape lines when generating HTML

Move DiffParser HTML escaping so it only is applied when generating HTML.

1 file changed with 9 insertions and 9 deletions:

kallithea/lib/diffs.py

0 comments (0 inline, 0 general)

kallithea/lib/diffs.py

➞

Show inline comments

@@ @@ -399,15 +399,17 @@ class DiffProcessor(object): @@
                 'stats':            stats,
             })
         if not html:
             return _files
         # highlight inline changes when one del is followed by one add
         for diff_data in _files:
             for chunk in diff_data['chunks']:
                 for change in chunk:
                     change['line'] = _escaper(change['line'])
                 # highlight inline changes when one del is followed by one add
                 lineiter = iter(chunk)
                 try:
                     peekline = next(lineiter)
                     while True:
                         # find a first del line
                         while peekline['action'] != 'del':
@@ @@ -444,22 +446,22 @@ class DiffProcessor(object): @@
 _escape_re = re.compile(r'(&)|(<)|(>)|(\t)($)?|(\r)|( $)')
 def _escaper(diff_line):
     r"""
-    Do HTML escaping/markup of a single diff line (including first +/- column)
+    Do HTML escaping/markup of a single diff line (excluding first +/- column)
     >>> _escaper('foobar')
     'foobar'
     >>> _escaper('@foo & bar')
     '@foo &amp; bar'
     >>> _escaper('+foo < bar')
     '+foo &lt; bar'
     >>> _escaper('-foo > bar')
     '-foo &gt; bar'
     >>> _escaper('foo < bar')
     'foo &lt; bar'
     >>> _escaper('foo > bar')
     'foo &gt; bar'
     >>> _escaper(' <foo>')
     ' &lt;foo&gt;'
     >>> _escaper(' foo\tbar')
     ' foo<u>\t</u>bar'
     >>> _escaper(' foo\rbar\r')
     ' foo<u class="cr"></u>bar<u class="cr"></u>'
@@ @@ -499,14 +501,12 @@ def _escaper(diff_line): @@
             if groups[4] is not None:  # end of line
                 return '<u>\t</u><i></i>'
             return '<u>\t</u>'
         if groups[5]:
             return '<u class="cr"></u>'
         if groups[6]:
             if m.start() == 0:
                 return ' '  # first column space shouldn't make empty lines show up as trailing space
             return ' <i></i>'
         assert False
     return _escape_re.sub(substitute, diff_line)
@@ @@ -591,13 +591,13 @@ def _get_header(vcs, diff_chunk): @@
     if rest:
         if _header_next_check.match(rest):
             raise Exception('cannot parse %s diff header: %r followed by %r' % (vcs, safe_str(bytes(diff_chunk[:match.end()])), safe_str(bytes(rest[:1000]))))
         if rest[-1:] != b'\n':
             # The diff will generally already have trailing \n (and be a memoryview). It might also be huge so we don't want to allocate it twice. But in this very rare case, we don't care.
             rest = bytes(rest) + b'\n'
-    diff_lines = (_escaper(safe_str(m.group(1))) for m in re.finditer(br'(.*)\n', rest))
     diff_lines = (safe_str(m.group(1)) for m in re.finditer(br'(.*)\n', rest))
     return meta_info, diff_lines
 _chunk_re = re.compile(r'^@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*)')
 _newline_marker = re.compile(r'^\\ No newline at end of file')

0 comments (0 inline, 0 general)