def _get_matches(self, page):
matches = []
# Search in page name
results = TitleSearch(self._pattern, use_re=self.use_re, case=self.case)._get_matches(page)
if results:
matches.extend(results)
# Search in page body
body = page.get_raw_body()
for match in self.search_re.finditer(body):
matches.append(TextMatch(re_match=match))
return matches
def _get_matches(self, page):
# Get matches in page links
matches = []
# XXX in python 2.5 any() may be used.
found = False
for link in page.getPageLinks(page.request):
if self.search_re.match(link):
found = True
break
if found:
# Search in page text
results = self.textsearch.search(page)
if results:
matches.extend(results)
else: # This happens e.g. for pages that use navigation macros
matches.append(TextMatch(0, 0))
return matches
def _get_matches(self, page):
""" match categories like this:
... some page text ...
----
## optionally some comments, e.g. about possible categories:
## CategoryFoo
CategoryTheRealAndOnly
Note: there might be multiple comment lines, but all real categories
must be on a single line either directly below the ---- or
directly below some comment lines.
"""
matches = []
pattern = r'(?m)(^-----*\s*\r?\n)(^##.*\r?\n)*^(?!##)(.*)\b%s\b' % self.pattern
search_re = self._build_re(pattern, use_re=self.use_re, case=self.case)[1] # we need only a regexp, but not a pattern
body = page.get_raw_body()
for match in search_re.finditer(body):
matches.append(TextMatch(re_match=match))
return matches