#!/usr/bin/python """ wiki.py - A Simple Python Wiki Author: Sean B. Palmer, inamidst.com """ import cgitb; cgitb.enable() import sys, os, re, cgi, glob, time from HTMLParser import HTMLParser try: from cStringIO import StringIO except ImportError: from StringIO import StringIO __version__ = str(time.time()) class Parser(object): EOF = 0 def __init__(self, write=None, error=None): self.text = None self.pos = 0 if write is None: write = sys.stdout.write self.write = write if error is None: # example: sys.stderr.write("%s: %s" % (e, msg)) error = lambda e, msg: None self.error = error def find(self, tokens): """For each token in tokens, if the current position matches one of those tokens, return True. Return False otherwise.""" for token in tokens: if token == self.EOF: if self.pos == len(self.text): return True elif self.text[self.pos:].startswith(token): return True return False def eat(self, token): """Eat the length of token if token's an int, or the token itself.""" if type(token) is int: if (self.pos + token) > len(self.text): self.error("UnexpectedEOF", "Reached end of file") return None s = self.text[self.pos:self.pos+token] self.pos += token return s else: assert self.find([token]) self.pos += len(token) if self.pos > len(self.text): self.error("UnexpectedEOF", "Reached end of file") return None return token def get(self, tokens, start=None, finish=None): if start is not None: self.eat(start) content = '' while not self.find(tokens): s = self.eat(1) if s is not None: content += s else: return content # reached EOF if finish is not None: self.eat(finish) return content r_tag = re.compile(r'(?"]+$') r_emdash = re.compile(r'[A-Za-z0-9"]--(?=[A-Za-z0-9"{])') r_alpha = re.compile(r'[A-Za-z]+') def makeID(s, current): s = (''.join(r_alpha.findall(s)) or 'id') + str(len(s)) while s in current: s += 'n' return s class TextParser(Parser): LIST = 0 HEADING = 1 PRE = 2 QUOT = 3 PARAGRAPH = 4 LI_START = '* ' LI_OPEN = '\n* ' PRE_START = '{{{\n' PRE_END = '\n}}}' QUOT_START = '[[[\n' QUOT_END = '\n]]]' H_START = '@ ' SEPERATOR = '\n\n' def __init__(self, write=None, error=None): Parser.__init__(self, write=write, error=error) self.rawlinks = [] self.ids = [] def __call__(self, s): self.text = s self.normalize() self.parse() def normalize(self): self.text = self.text.strip() # ('\t\r\n ') self.text = self.text.replace('\r\n', '\n') self.text = self.text.replace('\r', '\n') self.text = re.sub(r'(?sm)\n[ \t]*\n', '\n\n', self.text) def parse(self): blocks = [] while 1: blocks.append(self.blockElement()) if self.find([Parser.EOF]): break for block in blocks: blocktype, values = block[0], block[1:] {self.LIST: self.listElement, self.HEADING: self.headingElement, self.PRE: self.preElement, self.QUOT: self.quotElement, self.PARAGRAPH: self.paragraphElement }[blocktype](*values) def blockElement(self): self.whitespace() if self.find([self.LI_START]): content = self.get([self.SEPERATOR, Parser.EOF], self.LI_START) content = tuple(content.split('\n* ')) return (self.LIST,) + content elif self.find([self.H_START]): content = self.get(['\n', Parser.EOF], self.H_START) return (self.HEADING, content) elif self.find([self.PRE_START]): content = self.get([self.PRE_END], self.PRE_START, self.PRE_END) return (self.PRE, content) elif self.find([self.QUOT_START]): content = self.get([self.QUOT_END], self.QUOT_START, self.QUOT_END) if self.find([' - ']): citation = self.get(['\n', Parser.EOF], ' - ') if not (r_uri.match(citation) and citation): self.error('CitationURIError', # @@ allow other stuff? 'Citation (%s) must be a URI.' % citation) else: citation = None return (self.QUOT, content, citation) else: return (self.PARAGRAPH, self.get([self.SEPERATOR, Parser.EOF])) def whitespace(self): while self.find(' \t\n'): self.eat(1) def listElement(self, *items): self.write('
') self.write('\n') self.write(self.wikiParse(content, level=0)) self.write('\n') self.write('') self.write('\n') def quotElement(self, content, cite): self.write('
') self.write('\n') self.write('') self.write('\n') def paragraphElement(self, content): self.write('') # @@ self.write('\n') self.write(self.wikiParse(content, level=0)) self.write('\n') self.write('') self.write('\n') self.write('
') self.write(self.wikiParse(content)) self.write('
') self.write('\n') def wikiParse(self, s, level=None): if level is None: level = 1 # @@ use a proper parser, or catch the matches pos, result = 0, '' while pos < len(s): m = r_tag.match(s[pos:]) if m: span = m.span() result += self.tag(s[pos:pos+span[1]], level=level) pos += span[1] - span[0] else: m = r_emdash.match(s[pos:]) if m and (level > 0): # unicode must be explicit inresult += s[pos] + '—' # u'\u2014'.encode('utf-8') pos += 3 elif (s[pos] == '{') and (s[pos+1:pos+2] != '{') and (level > 0): if (s < 10): area = s[0:pos+10] else: area = s[pos-10:pos+10] msg = "The '{' must be escaped as '{{' in %r" % area raise "WikiParseError", msg elif (s[pos:pos+2] == '{{'): # d8uv bug "and (level > 0): " result += '{' pos += 2 elif s[pos] == '&': result += '&' pos += 1 elif s[pos] == '<': result += '<' pos += 1 else: result += s[pos] pos += 1 return result def iriParse(self, uri): r_unicode = re.compile(r'\{U\+([1-9A-F][0-9A-F]{1,5})\}') def escape(m): bytes = unichr(int(m.group(1), 16)).encode('utf-8') return ''.join(['%%%02X' % ord(s) for s in bytes]) return r_unicode.sub(escape, uri) def unicodeify(self, s): if len(s) not in (2, 4, 6): raise ValueError, 'Must be of length 2, 4, or 6' for letter in 'abcdef': if letter in s: raise ValueError, 'Unicode escapes must be lower-case' i = int(s.lstrip('0'), 16) raw = [0x9, 0xA, 0xD] + list(xrange(0x20, 0x7E)) del raw[raw.index(0x2D)], raw[raw.index(0x5D)], raw[raw.index(0x7D)] if i in raw: return chr(i) # printable - '-]}' elif i > 0x10FFFF: raise ValueError, 'Codepoint is out of range' return '%s;' % s def tag(self, s, level=None): if level is None: level = 1 # @@ { {U+..}? s = s[1:-1] # @@ or s.strip('{}') if s.startswith('U+'): try: result = self.unicodeify(s[2:]) except ValueError: result = cgi.escape('{%s}' % s) elif s == '$timenow': result = time.strftime('%Y-%m-%d %H:%M:%S', time.gmtime()) elif s == '$datenow': result = time.strftime('%Y-%m-%d', time.gmtime()) elif level < 1: result = '{' + self.wikiParse('%s}' % s) elif s.startswith('* '): result = '%s' % s[2:] elif s.startswith('#'): i = s.find(' ') href, title = s[:i], s[i+1:] result = '%s' % (href, title) elif not re.compile(r'[A-Za-z0-9/_.-]').match(s): result = cgi.escape('{%s}' % s) else: self.rawlinks.append(s) words = s.split(' ') words = [word.strip() for word in words if word.strip()] if ('/' not in words[0]) and (':' not in words[0]): # @@! wikiname = ''.join(words) uri = './%s' % wikiname else: uri, s = words[0], ' '.join(words[1:]) uri, s = cgi.escape(uri, quote=1), cgi.escape(s) result = '%s' % (uri, s) return result def wikiParse(s): output = StringIO() parse = TextParser(write=output.write) parse(s) output.flush() output.seek(0) return output.read() class Wikify(HTMLParser): def __init__(self, write=None): HTMLParser.__init__(self) if write is None: self.write = sys.stdout.write else: self.write = write self.content = False self.block = False self.blockquote = False self.anchor = False self.current = None self.heading = False def handle_starttag(self, tag, attrs): self.current = tag attrs = dict(attrs) xhtmlxmlns = 'http://www.w3.org/1999/xhtml' if (tag == 'html') and (attrs.get('xmlns') != xhtmlxmlns): raise "ParseError", "document is not XHTML" elif (tag == 'div') and (attrs.get('class') == 'content'): self.content = True if self.content: if tag in ('p', 'li', 'h1', 'h2', 'pre'): self.block = True if tag == 'li': self.write('* ') elif tag in ('h1', 'h2'): self.write('@ ') elif tag == 'pre' and not self.blockquote: self.write('{{{') elif tag == 'blockquote': self.blockquote = attrs or True self.write('[[[') elif tag == 'strong': self.write('{* ') elif tag == 'a': self.anchor = attrs self.anchor['_'] = '' elif tag == 'h1': self.write('@ ') self.heading = True def handle_endtag(self, tag): self.current = None if self.content: if tag in ('p', 'li', 'h1', 'h2', 'pre'): self.block = False if tag in ('p', 'h1', 'h2'): self.write('\n\n') elif tag in ('ul', 'li'): self.write('\n') elif tag == 'pre' and not self.blockquote: self.write('}}}\n\n') elif tag == 'blockquote': self.write(']]]') if hasattr(self.blockquote, 'get'): cite = self.blockquote.get('cite', None) if cite is not None: self.write(' - %s' % cite) self.write('\n\n') self.blockquote = False elif tag == 'a': attrs, dual = self.anchor, True uri, title = attrs.get('href', ''), attrs.get('_', '') stuff = [w.strip() for w in title.split(' ') if w.strip()] stitle = ''.join(stuff) if uri.startswith('./'): wikiname = uri[2:] if r_name.match(wikiname): if wikiname == stitle: dual = False if not dual: self.write('{%s}' % title) else: self.write('{%s %s}' % (uri, title)) self.anchor = False elif tag == 'strong': self.write('}') elif tag == 'div': self.content = False elif tag == 'h1': self.write('\n\n') self.heading = False def handle_data(self, data): if self.current in ('p', 'li', 'h1', 'h2', 'pre'): # d8uv, pre added data = data.replace('{', '{{') if (self.content and self.block) or self.heading: if not self.anchor: self.write(data) else: self.anchor['_'] += data def handle_charref(self, name): if (self.content and self.block) or self.heading: if name.startswith('x'): result = '{U+%s}' % name.lstrip('x') elif name == '8212': result = '--' else: raise "ParseError", "Unknown character reference: %s" % name if not self.anchor: self.write(result) else: self.anchor['_'] += result def handle_entityref(self, name): if (self.content and self.block) or self.heading: entities = {'lt':'<', 'gt':'>', 'amp':'&', 'quot':'"'} result = entities.get(name, '?') if not self.anchor: self.write(result) else: self.anchor['_'] += result def wikify(s): output = StringIO() parser = Wikify(write=output.write) parser.feed(s) # @@ except? output.flush() output.seek(0) return output.read() def html(title, body): return ( '', '', '', '', '') return html('Editing %s' % (wikiname or 'index'), '\n'.join(lines)) def doPost(wikiname, text, append=False): if append: filename = '%s.html' % wikiname if os.path.isfile(filename): f = open(filename) data = f.read() # try: text = wikify(data) + text # except "ParseError", e: # msg = "Error: couldn't wikify existing source! (%s)\n" % e # text = msg + text f.close() if text: # If there's any content to the post... html = [(line + '\n') for line in compile(wikiname, text)] filename = '%s.html' % wikiname f = open(filename, 'w') f.writelines(html) f.close() try: os.chmod(filename, 0646) except OSError: pass elif (not append) and os.path.isfile('%s.html' % wikiname): os.remove('%s.html' % wikiname) else: noop() def post(wikiname): form = cgi.FieldStorage() form.__call__ = lambda s: form[s].value text = form('text') if form.has_key('append'): append = True else: append = False doPost(wikiname, text, append=append) def main(env=None, out=None): if env is None: env = os.environ if out is None: out = sys.stdout method = env.get('REQUEST_METHOD') if method not in frozenset(['GET', 'POST']): raise ValueError('Unsupported method: %s' % method) out.write('Content-Type: text/html; charset=utf-8\r\n') out.write('\r\n') uri = env.get('REQUEST_URI') path, directory, filename = uri.rsplit('/', 2) wikiname = filename or 'index' if not r_name.match(wikiname): raise ValueError('Invalid wikiname: %s' % wikiname) if directory != '@edit': if method == 'POST': post(wikiname) out.writelines(get(wikiname)) else: out.writelines(edit(wikiname)) if __name__=='__main__': main()%s ' % title, '', '', '\n%s\n' % body, '', '' ) def compile(wikiname, text): content = wikiParse(text) s = '' heading = None if content.startswith('') heading = content[:i][len('
')] s += '
\n\n' % heading content = content[i:].lstrip('\r\n') heading = s[:i][j+1:-len('
')] else: s += '%s
\n' % (wikiname) s += '\n%s\n\n\n' % content s += ''' Sean B. Palmer ''' if (heading is not None) and (heading != wikiname): title = '%s - %s' % (heading, wikiname) else: title = wikiname return html(title, s) def noop(): # @@ Should we display some sort of user warning? pass def get(wikiname): filename = '%s.html' % wikiname if os.path.isfile(filename): f = open(filename) for line in f: yield line f.close() else: lines = ('%s
\nThis page does not yet exist. ' % wikiname, 'Create it!
' % wikiname) for line in html('Create %s' % wikiname, '\n'.join(lines)): yield line def edit(wikiname): filename = '%s.html' % wikiname if os.path.isfile(filename): f = open(filename) data = f.read() try: text = wikify(data) except "ParseError", e: text = "Error: couldn't wikify source! (%s)\n" % e f.close() else: text = '' if wikiname == 'index': wikiname = '' lines = ('