+++ /dev/null
-"""A parser for HTML and XHTML."""\r
-\r
-# This file is based on sgmllib.py, but the API is slightly different.\r
-\r
-# XXX There should be a way to distinguish between PCDATA (parsed\r
-# character data -- the normal case), RCDATA (replaceable character\r
-# data -- only char and entity references and end tags are special)\r
-# and CDATA (character data -- only end tags are special).\r
-\r
-\r
-import markupbase\r
-import re\r
-\r
-# Regular expressions used for parsing\r
-\r
-interesting_normal = re.compile('[&<]')\r
-interesting_cdata = re.compile(r'<(/|\Z)')\r
-incomplete = re.compile('&[a-zA-Z#]')\r
-\r
-entityref = re.compile('&([a-zA-Z][-.a-zA-Z0-9]*)[^a-zA-Z0-9]')\r
-charref = re.compile('&#(?:[0-9]+|[xX][0-9a-fA-F]+)[^0-9a-fA-F]')\r
-\r
-starttagopen = re.compile('<[a-zA-Z]')\r
-piclose = re.compile('>')\r
-commentclose = re.compile(r'--\s*>')\r
-tagfind = re.compile('[a-zA-Z][-.a-zA-Z0-9:_]*')\r
-attrfind = re.compile(\r
- r'\s*([a-zA-Z_][-.:a-zA-Z_0-9]*)(\s*=\s*'\r
- r'(\'[^\']*\'|"[^"]*"|[^\s"\'=<>`]*))?')\r
-\r
-locatestarttagend = re.compile(r"""\r
- <[a-zA-Z][-.a-zA-Z0-9:_]* # tag name\r
- (?:\s+ # whitespace before attribute name\r
- (?:[a-zA-Z_][-.:a-zA-Z0-9_]* # attribute name\r
- (?:\s*=\s* # value indicator\r
- (?:'[^']*' # LITA-enclosed value\r
- |\"[^\"]*\" # LIT-enclosed value\r
- |[^'\">\s]+ # bare value\r
- )\r
- )?\r
- )\r
- )*\r
- \s* # trailing whitespace\r
-""", re.VERBOSE)\r
-endendtag = re.compile('>')\r
-endtagfind = re.compile('</\s*([a-zA-Z][-.a-zA-Z0-9:_]*)\s*>')\r
-\r
-\r
-class HTMLParseError(Exception):\r
- """Exception raised for all parse errors."""\r
-\r
- def __init__(self, msg, position=(None, None)):\r
- assert msg\r
- self.msg = msg\r
- self.lineno = position[0]\r
- self.offset = position[1]\r
-\r
- def __str__(self):\r
- result = self.msg\r
- if self.lineno is not None:\r
- result = result + ", at line %d" % self.lineno\r
- if self.offset is not None:\r
- result = result + ", column %d" % (self.offset + 1)\r
- return result\r
-\r
-\r
-class HTMLParser(markupbase.ParserBase):\r
- """Find tags and other markup and call handler functions.\r
-\r
- Usage:\r
- p = HTMLParser()\r
- p.feed(data)\r
- ...\r
- p.close()\r
-\r
- Start tags are handled by calling self.handle_starttag() or\r
- self.handle_startendtag(); end tags by self.handle_endtag(). The\r
- data between tags is passed from the parser to the derived class\r
- by calling self.handle_data() with the data as argument (the data\r
- may be split up in arbitrary chunks). Entity references are\r
- passed by calling self.handle_entityref() with the entity\r
- reference as the argument. Numeric character references are\r
- passed to self.handle_charref() with the string containing the\r
- reference as the argument.\r
- """\r
-\r
- CDATA_CONTENT_ELEMENTS = ("script", "style")\r
-\r
-\r
- def __init__(self):\r
- """Initialize and reset this instance."""\r
- self.reset()\r
-\r
- def reset(self):\r
- """Reset this instance. Loses all unprocessed data."""\r
- self.rawdata = ''\r
- self.lasttag = '???'\r
- self.interesting = interesting_normal\r
- markupbase.ParserBase.reset(self)\r
-\r
- def feed(self, data):\r
- r"""Feed data to the parser.\r
-\r
- Call this as often as you want, with as little or as much text\r
- as you want (may include '\n').\r
- """\r
- self.rawdata = self.rawdata + data\r
- self.goahead(0)\r
-\r
- def close(self):\r
- """Handle any buffered data."""\r
- self.goahead(1)\r
-\r
- def error(self, message):\r
- raise HTMLParseError(message, self.getpos())\r
-\r
- __starttag_text = None\r
-\r
- def get_starttag_text(self):\r
- """Return full source of start tag: '<...>'."""\r
- return self.__starttag_text\r
-\r
- def set_cdata_mode(self):\r
- self.interesting = interesting_cdata\r
-\r
- def clear_cdata_mode(self):\r
- self.interesting = interesting_normal\r
-\r
- # Internal -- handle data as far as reasonable. May leave state\r
- # and data to be processed by a subsequent call. If 'end' is\r
- # true, force handling all data as if followed by EOF marker.\r
- def goahead(self, end):\r
- rawdata = self.rawdata\r
- i = 0\r
- n = len(rawdata)\r
- while i < n:\r
- match = self.interesting.search(rawdata, i) # < or &\r
- if match:\r
- j = match.start()\r
- else:\r
- j = n\r
- if i < j: self.handle_data(rawdata[i:j])\r
- i = self.updatepos(i, j)\r
- if i == n: break\r
- startswith = rawdata.startswith\r
- if startswith('<', i):\r
- if starttagopen.match(rawdata, i): # < + letter\r
- k = self.parse_starttag(i)\r
- elif startswith("</", i):\r
- k = self.parse_endtag(i)\r
- elif startswith("<!--", i):\r
- k = self.parse_comment(i)\r
- elif startswith("<?", i):\r
- k = self.parse_pi(i)\r
- elif startswith("<!", i):\r
- k = self.parse_declaration(i)\r
- elif (i + 1) < n:\r
- self.handle_data("<")\r
- k = i + 1\r
- else:\r
- break\r
- if k < 0:\r
- if end:\r
- self.error("EOF in middle of construct")\r
- break\r
- i = self.updatepos(i, k)\r
- elif startswith("&#", i):\r
- match = charref.match(rawdata, i)\r
- if match:\r
- name = match.group()[2:-1]\r
- self.handle_charref(name)\r
- k = match.end()\r
- if not startswith(';', k-1):\r
- k = k - 1\r
- i = self.updatepos(i, k)\r
- continue\r
- else:\r
- if ";" in rawdata[i:]: #bail by consuming &#\r
- self.handle_data(rawdata[0:2])\r
- i = self.updatepos(i, 2)\r
- break\r
- elif startswith('&', i):\r
- match = entityref.match(rawdata, i)\r
- if match:\r
- name = match.group(1)\r
- self.handle_entityref(name)\r
- k = match.end()\r
- if not startswith(';', k-1):\r
- k = k - 1\r
- i = self.updatepos(i, k)\r
- continue\r
- match = incomplete.match(rawdata, i)\r
- if match:\r
- # match.group() will contain at least 2 chars\r
- if end and match.group() == rawdata[i:]:\r
- self.error("EOF in middle of entity or char ref")\r
- # incomplete\r
- break\r
- elif (i + 1) < n:\r
- # not the end of the buffer, and can't be confused\r
- # with some other construct\r
- self.handle_data("&")\r
- i = self.updatepos(i, i + 1)\r
- else:\r
- break\r
- else:\r
- assert 0, "interesting.search() lied"\r
- # end while\r
- if end and i < n:\r
- self.handle_data(rawdata[i:n])\r
- i = self.updatepos(i, n)\r
- self.rawdata = rawdata[i:]\r
-\r
- # Internal -- parse processing instr, return end or -1 if not terminated\r
- def parse_pi(self, i):\r
- rawdata = self.rawdata\r
- assert rawdata[i:i+2] == '<?', 'unexpected call to parse_pi()'\r
- match = piclose.search(rawdata, i+2) # >\r
- if not match:\r
- return -1\r
- j = match.start()\r
- self.handle_pi(rawdata[i+2: j])\r
- j = match.end()\r
- return j\r
-\r
- # Internal -- handle starttag, return end or -1 if not terminated\r
- def parse_starttag(self, i):\r
- self.__starttag_text = None\r
- endpos = self.check_for_whole_start_tag(i)\r
- if endpos < 0:\r
- return endpos\r
- rawdata = self.rawdata\r
- self.__starttag_text = rawdata[i:endpos]\r
-\r
- # Now parse the data between i+1 and j into a tag and attrs\r
- attrs = []\r
- match = tagfind.match(rawdata, i+1)\r
- assert match, 'unexpected call to parse_starttag()'\r
- k = match.end()\r
- self.lasttag = tag = rawdata[i+1:k].lower()\r
-\r
- while k < endpos:\r
- m = attrfind.match(rawdata, k)\r
- if not m:\r
- break\r
- attrname, rest, attrvalue = m.group(1, 2, 3)\r
- if not rest:\r
- attrvalue = None\r
- elif attrvalue[:1] == '\'' == attrvalue[-1:] or \\r
- attrvalue[:1] == '"' == attrvalue[-1:]:\r
- attrvalue = attrvalue[1:-1]\r
- attrvalue = self.unescape(attrvalue)\r
- attrs.append((attrname.lower(), attrvalue))\r
- k = m.end()\r
-\r
- end = rawdata[k:endpos].strip()\r
- if end not in (">", "/>"):\r
- lineno, offset = self.getpos()\r
- if "\n" in self.__starttag_text:\r
- lineno = lineno + self.__starttag_text.count("\n")\r
- offset = len(self.__starttag_text) \\r
- - self.__starttag_text.rfind("\n")\r
- else:\r
- offset = offset + len(self.__starttag_text)\r
- self.error("junk characters in start tag: %r"\r
- % (rawdata[k:endpos][:20],))\r
- if end.endswith('/>'):\r
- # XHTML-style empty tag: <span attr="value" />\r
- self.handle_startendtag(tag, attrs)\r
- else:\r
- self.handle_starttag(tag, attrs)\r
- if tag in self.CDATA_CONTENT_ELEMENTS:\r
- self.set_cdata_mode()\r
- return endpos\r
-\r
- # Internal -- check to see if we have a complete starttag; return end\r
- # or -1 if incomplete.\r
- def check_for_whole_start_tag(self, i):\r
- rawdata = self.rawdata\r
- m = locatestarttagend.match(rawdata, i)\r
- if m:\r
- j = m.end()\r
- next = rawdata[j:j+1]\r
- if next == ">":\r
- return j + 1\r
- if next == "/":\r
- if rawdata.startswith("/>", j):\r
- return j + 2\r
- if rawdata.startswith("/", j):\r
- # buffer boundary\r
- return -1\r
- # else bogus input\r
- self.updatepos(i, j + 1)\r
- self.error("malformed empty start tag")\r
- if next == "":\r
- # end of input\r
- return -1\r
- if next in ("abcdefghijklmnopqrstuvwxyz=/"\r
- "ABCDEFGHIJKLMNOPQRSTUVWXYZ"):\r
- # end of input in or before attribute value, or we have the\r
- # '/' from a '/>' ending\r
- return -1\r
- self.updatepos(i, j)\r
- self.error("malformed start tag")\r
- raise AssertionError("we should not get here!")\r
-\r
- # Internal -- parse endtag, return end or -1 if incomplete\r
- def parse_endtag(self, i):\r
- rawdata = self.rawdata\r
- assert rawdata[i:i+2] == "</", "unexpected call to parse_endtag"\r
- match = endendtag.search(rawdata, i+1) # >\r
- if not match:\r
- return -1\r
- j = match.end()\r
- match = endtagfind.match(rawdata, i) # </ + tag + >\r
- if not match:\r
- self.error("bad end tag: %r" % (rawdata[i:j],))\r
- tag = match.group(1)\r
- self.handle_endtag(tag.lower())\r
- self.clear_cdata_mode()\r
- return j\r
-\r
- # Overridable -- finish processing of start+end tag: <tag.../>\r
- def handle_startendtag(self, tag, attrs):\r
- self.handle_starttag(tag, attrs)\r
- self.handle_endtag(tag)\r
-\r
- # Overridable -- handle start tag\r
- def handle_starttag(self, tag, attrs):\r
- pass\r
-\r
- # Overridable -- handle end tag\r
- def handle_endtag(self, tag):\r
- pass\r
-\r
- # Overridable -- handle character reference\r
- def handle_charref(self, name):\r
- pass\r
-\r
- # Overridable -- handle entity reference\r
- def handle_entityref(self, name):\r
- pass\r
-\r
- # Overridable -- handle data\r
- def handle_data(self, data):\r
- pass\r
-\r
- # Overridable -- handle comment\r
- def handle_comment(self, data):\r
- pass\r
-\r
- # Overridable -- handle declaration\r
- def handle_decl(self, decl):\r
- pass\r
-\r
- # Overridable -- handle processing instruction\r
- def handle_pi(self, data):\r
- pass\r
-\r
- def unknown_decl(self, data):\r
- self.error("unknown declaration: %r" % (data,))\r
-\r
- # Internal -- helper to remove special character quoting\r
- entitydefs = None\r
- def unescape(self, s):\r
- if '&' not in s:\r
- return s\r
- def replaceEntities(s):\r
- s = s.groups()[0]\r
- try:\r
- if s[0] == "#":\r
- s = s[1:]\r
- if s[0] in ['x','X']:\r
- c = int(s[1:], 16)\r
- else:\r
- c = int(s)\r
- return unichr(c)\r
- except ValueError:\r
- return '&#'+s+';'\r
- else:\r
- # Cannot use name2codepoint directly, because HTMLParser supports apos,\r
- # which is not part of HTML 4\r
- import htmlentitydefs\r
- if HTMLParser.entitydefs is None:\r
- entitydefs = HTMLParser.entitydefs = {'apos':u"'"}\r
- for k, v in htmlentitydefs.name2codepoint.iteritems():\r
- entitydefs[k] = unichr(v)\r
- try:\r
- return self.entitydefs[s]\r
- except KeyError:\r
- return '&'+s+';'\r
-\r
- return re.sub(r"&(#?[xX]?(?:[0-9a-fA-F]+|\w{1,8}));", replaceEntities, s)\r
projects / mirror_edk2.git / blobdiff