+def htmlentity_transform(matchobj):
+ """Transforms an HTML entity to a character.
+
+ This function receives a match object and is intended to be used with
+ the re.sub() function.
+ """
+ entity = matchobj.group(1)
+
+ # Known non-numeric HTML entity
+ if entity in compat_html_entities.name2codepoint:
+ return compat_chr(compat_html_entities.name2codepoint[entity])
+
+ mobj = re.match(u'(?u)#(x?\\d+)', entity)
+ if mobj is not None:
+ numstr = mobj.group(1)
+ if numstr.startswith(u'x'):
+ base = 16
+ numstr = u'0%s' % numstr
+ else:
+ base = 10
+ return compat_chr(int(numstr, base))
+
+ # Unknown entity in name, return its literal representation
+ return (u'&%s;' % entity)
+
+compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix
+class IDParser(compat_html_parser.HTMLParser):
+ """Modified HTMLParser that isolates a tag with the specified id"""
+ def __init__(self, id):
+ self.id = id
+ self.result = None
+ self.started = False
+ self.depth = {}
+ self.html = None
+ self.watch_startpos = False
+ self.error_count = 0
+ compat_html_parser.HTMLParser.__init__(self)
+
+ def error(self, message):
+ if self.error_count > 10 or self.started:
+ raise compat_html_parser.HTMLParseError(message, self.getpos())
+ self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line
+ self.error_count += 1
+ self.goahead(1)
+
+ def loads(self, html):
+ self.html = html
+ self.feed(html)
+ self.close()
+
+ def handle_starttag(self, tag, attrs):
+ attrs = dict(attrs)
+ if self.started:
+ self.find_startpos(None)
+ if 'id' in attrs and attrs['id'] == self.id:
+ self.result = [tag]
+ self.started = True
+ self.watch_startpos = True
+ if self.started:
+ if not tag in self.depth: self.depth[tag] = 0
+ self.depth[tag] += 1
+
+ def handle_endtag(self, tag):
+ if self.started:
+ if tag in self.depth: self.depth[tag] -= 1
+ if self.depth[self.result[0]] == 0:
+ self.started = False
+ self.result.append(self.getpos())
+
+ def find_startpos(self, x):
+ """Needed to put the start position of the result (self.result[1])
+ after the opening tag with the requested id"""
+ if self.watch_startpos:
+ self.watch_startpos = False
+ self.result.append(self.getpos())
+ handle_entityref = handle_charref = handle_data = handle_comment = \
+ handle_decl = handle_pi = unknown_decl = find_startpos
+
+ def get_result(self):
+ if self.result is None:
+ return None
+ if len(self.result) != 3:
+ return None
+ lines = self.html.split('\n')
+ lines = lines[self.result[1][0]-1:self.result[2][0]]
+ lines[0] = lines[0][self.result[1][1]:]
+ if len(lines) == 1:
+ lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]]
+ lines[-1] = lines[-1][:self.result[2][1]]
+ return '\n'.join(lines).strip()
+
+def get_element_by_id(id, html):
+ """Return the content of the tag with the specified id in the passed HTML document"""
+ parser = IDParser(id)
+ try:
+ parser.loads(html)
+ except compat_html_parser.HTMLParseError:
+ pass
+ return parser.get_result()
+
+
+def clean_html(html):
+ """Clean an HTML snippet into a readable string"""
+ # Newline vs <br />
+ html = html.replace('\n', ' ')
+ html = re.sub('\s*<\s*br\s*/?\s*>\s*', '\n', html)
+ # Strip html tags
+ html = re.sub('<.*?>', '', html)
+ # Replace html entities
+ html = unescapeHTML(html)
+ return html