2 # -*- coding: utf-8 -*-
4 # Copyright 1999, 2000 Martin Pool <mbp@humbug.org.au>
5 # Copyright 2002 Gerardo Poggiali
6 # Copyright 2007, 2008, 2009 Bernie Innocenti <bernie@codewiz.org>
8 # This program is free software: you can redistribute it and/or modify
9 # it under the terms of the GNU General Public License as published by
10 # the Free Software Foundation, either version 3 of the License, or
11 # (at your option) any later version.
13 # This program is distributed in the hope that it will be useful, but
14 # WITHOUT ANY WARRANTY; without even the implied warranty of
15 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 # General Public License for more details.
18 # You should have received a copy of the GNU General Public License
19 # along with this program. If not, see <http://www.gnu.org/licenses/>.
21 __version__ = '$Id$'[4:12]
23 from time import clock
26 import cgi, sys, os, re, errno, stat
28 # Regular expression defining a WikiWord
29 # (but this definition is also assumed in other places)
30 word_re = re.compile(r"^\b((([A-Z][a-z0-9]+){2,}/)*([A-Z][a-z0-9]+){2,})\b$")
31 # FIXME: we accept stuff like foo/../bar and we shouldn't
32 file_re = re.compile(r"^\b([A-Za-z0-9_\-][A-Za-z0-9_\.\-/]*)\b$")
33 img_re = re.compile(r"^.*\.(png|gif|jpg|jpeg|bmp|ico|ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt)$", re.IGNORECASE)
34 video_re = re.compile(r"^.*\.(ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt)$", re.IGNORECASE)
35 url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$")
36 link_re = re.compile(r"(?:\[\[|{{)([^\s\|]+)(?:\s*\|\s*([^\]]+)|)(?:\]\]|}})")
37 ext_re = re.compile(r"\.([^\./]+)$")
41 # CGI stuff ---------------------------------------------------------
43 return os.environ.get('SCRIPT_NAME', '')
46 path_info = os.environ.get('PATH_INFO', '')
47 if len(path_info) and path_info[0] == '/':
48 return path_info[1:] or 'FrontPage'
50 return os.environ.get('QUERY_STRING', '') or 'FrontPage'
52 def privileged_path():
53 return privileged_url or script_name()
56 user = os.environ.get('REMOTE_USER', '')
57 if user is None or user == '' or user == 'anonymous':
58 user = 'AnonymousCoward'
62 return os.environ.get('REMOTE_ADDR', '')
64 def get_hostname(addr):
66 from socket import gethostbyaddr
67 return gethostbyaddr(addr)[0] + ' (' + addr + ')'
71 def is_external_url(pathname):
72 return (url_re.match(pathname) or pathname.startswith('/'))
74 def relative_url(pathname, privileged=False):
75 if not is_external_url(pathname):
77 url = privileged_path()
80 pathname = url + '/' + pathname
84 return re.sub(' ', '-', re.sub('[^a-z0-9_ ]', '', s.lower()).strip())
86 # Formatting stuff --------------------------------------------------
87 def emit_header(mime_type="text/html"):
88 print "Content-type: " + mime_type + "; charset=utf-8\n"
90 def sendfile(dest_file, src_file):
91 """Efficiently copy file data between file descriptors"""
93 data = src_file.read(65536)
97 def send_guru(msg_text, msg_type):
98 if not msg_text: return
99 print '<pre id="guru" onclick="this.style.display = \'none\'" class="' + msg_type + '">'
100 if msg_type == 'error':
101 print ' Software Failure. Press left mouse button to continue.\n'
103 if msg_type == 'error':
104 print '\n Guru Meditation #DEADBEEF.ABADC0DE'
107 sendfile(sys.stdout, open('gurumeditation.js', 'rb'))
111 def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=False):
113 if title_done: return
117 print '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'
118 print ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">'
119 print '<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">'
121 print "<head><title>%s: %s</title>" % (site_name, text)
122 print ' <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8" />'
124 print ' <meta name="robots" content="noindex,nofollow" />'
126 for meta in meta_urls:
127 http_equiv, content = meta
128 print ' <meta http-equiv="%s" content="%s" />' % (http_equiv, relative_url(content))
130 for link in link_urls:
132 print ' <link rel="%s" href="%s" />' % (rel, relative_url(href))
134 if name and writable and privileged_url is not None:
135 print ' <link rel="alternate" type="application/x-wiki" title="Edit this page" href="%s" />' \
136 % (privileged_path() + '?edit=' + name)
138 if history_url is not None:
139 print ' <link rel="alternate" type="application/rss+xml" title="RSS" href="%s" />' \
140 % relative_url(history_url + '?a=rss')
145 if name and writable and privileged_url is not None:
146 print '<body ondblclick="location.href=\'' + privileged_path() + '?edit=' + name + '\'">'
151 send_guru(msg_text, msg_type)
154 print '<div class="nav">'
155 print link_tag('FrontPage', relative_url(site_icon or 'Home'), 'navlink')
157 print ' <b>' + link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
159 print ' <b>' + text + '</b> '
160 print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
161 if 'history_url' in globals():
162 print ' | <a href="' + relative_url(history_url) + '" class="navlink">Recent Changes</a>'
164 print ' | <a href="' + relative_url(history_url + '?a=history;f=' + name) + '" class="navlink">Page History</a>'
167 print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
168 if privileged_url is not None:
170 print ' | ' + link_tag('?edit=' + name, 'Edit', 'navlink', privileged=True)
172 print ' | ' + link_tag(name, 'Login', 'navlink', privileged=True)
175 print ' | <i>Immutable Page</i>'
178 if user != 'AnonymousCoward':
179 print ' | <span class="login"><i><b>' + link_tag('User/' + user, user) + '</b></i></span>'
183 def send_httperror(status="403 Not Found", query=""):
184 print "Status: %s" % status
185 send_title(None, msg_text=("%s: on query '%s'" % (status, query)))
188 def link_tag(params, text=None, link_class=None, privileged=False):
190 text = params # default
191 elif img_re.match(text):
192 text = '<img border="0" src="' + text + '" />'
195 if is_external_url(params):
196 link_class = 'external'
197 elif file_re.match(params) and Page(params).exists():
198 link_class = 'wikilink'
200 params = nonexist_pfx + params
201 link_class = 'nonexistent'
203 classattr = 'class="%s" ' % link_class
204 # Prevent crawlers from following links potentially added by spammers or to generated pages
205 if link_class == 'external' or link_class == 'navlink':
206 classattr += 'rel="nofollow"'
208 return '<a %shref="%s">%s</a>' % (classattr, relative_url(params, privileged=privileged), text)
210 def link_inline(name, descr=None, args=''):
211 if not descr: descr = name
212 url = relative_url(name)
213 if video_re.match(name):
214 return '<video src="%s">Your browser does not support the HTML5 video tag</video>' % url
215 elif img_re.match(name):
216 return '<a href="%s"><img border="0" src="%s" alt="%s" /></a>' % (url, url + args, descr)
217 elif file_re.match(name) and not ext_re.search(name): # FIXME: this guesses a wiki page
218 return Page(name).send_naked()
220 return '<iframe width="100%%" scrolling="auto" frameborder="0" src="%s"><a href="%s">%s</a></iframe>' \
223 # Search ---------------------------------------------------
225 def handle_fullsearch(needle):
226 send_title(None, 'Full text search for "%s"' % (needle))
228 needle_re = re.compile(needle, re.IGNORECASE)
230 all_pages = page_list()
231 for page_name in all_pages:
232 body = Page(page_name).get_raw_body()
233 count = len(needle_re.findall(body))
235 hits.append((count, page_name))
237 # The default comparison for tuples compares elements in order,
238 # so this sorts by number of hits
243 for (count, page_name) in hits:
244 print '<li><p>' + link_tag(page_name)
245 print ' . . . . ' + `count`
246 print ['match', 'matches'][count != 1]
250 print_search_stats(len(hits), len(all_pages))
252 def handle_titlesearch(needle):
253 # TODO: check needle is legal -- but probably we can just accept any RE
254 send_title(None, "Title search for \"" + needle + '"')
256 needle_re = re.compile(needle, re.IGNORECASE)
257 all_pages = page_list()
258 hits = filter(needle_re.search, all_pages)
261 for filename in hits:
262 print '<li><p>' + link_tag(filename) + "</p></li>"
265 print_search_stats(len(hits), len(all_pages))
267 def print_search_stats(hits, searched):
268 print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
270 def handle_raw(pagename):
271 if not file_re.match(pagename):
272 send_httperror("403 Forbidden", pagename)
275 Page(pagename).send_raw()
277 def handle_edit(pagename):
278 if not file_re.match(pagename):
279 send_httperror("403 Forbidden", pagename)
284 if form['file'].value:
285 pg.save(form['file'].file.read(), form['changelog'].value)
287 pg.save(form['savetext'].value.replace('\r\n', '\n'), form['changelog'].value)
289 elif 'cancel' in form:
290 pg.msg_text = 'Editing canceled'
291 pg.msg_type = 'notice'
293 else: # preview or edit
295 if 'preview' in form:
296 text = form['savetext'].value
299 def make_index_key():
300 links = map(lambda ch: '<a href="#%s">%s</a>' % (ch, ch), 'abcdefghijklmnopqrstuvwxyz')
301 return '<p><center>'+ ' | '.join(links) + '</center></p>'
303 def page_list(dirname = None, re = word_re):
304 return sorted(filter(re.match, os.listdir(dirname or data_dir)))
306 def send_footer(mod_string=None):
307 if globals().get('debug_cgi', False):
308 cgi.print_arguments()
312 <div id="footer"><hr />
313 <p class="copyright">
314 <a rel="license" href="http://creativecommons.org/licenses/by-sa/3.0/"><img class="license" alt="Creative Commons License" src="%s" /></a>
315 <span class="benchmark">generated in %0.3fs</span> by <a href="http://www.codewiz.org/wiki/GeekiGeeki">GeekiGeeki</a> version %s
317 ''' % (relative_url('cc-by-sa.png'), clock() - start_time, __version__)
319 print '<p class="modified">last modified %s</p>' % mod_string
320 print '</div></body></html>'
323 """Object that turns Wiki markup into HTML.
325 All formatting commands can be parsed one line at a time, though
326 some state is carried over between lines.
328 def __init__(self, raw):
331 self.in_pre = self.in_html = self.in_table = self.in_li = False
332 self.in_header = True
333 self.list_indents = []
341 "^^": ["sup", False],
342 ",,": ["sub", False],
343 "''": ["em", False], # LEGACY
344 "'''": ["b", False], # LEGACY
345 "``": ["tt", False], # LEGACY
348 def _b_repl(self, word):
349 style = self.styles[word]
350 style[1] = not style[1]
351 return ['</', '<'][style[1]] + style[0] + '>'
353 def _tit_repl(self, word):
355 result = '</h%d><p>\n' % self.h_level
358 self.h_level = len(word) - 1
359 link = permalink(self.line)
360 result = '\n</p><h%d id="%s"><a class="heading" href="#%s">¶</a> ' % (self.h_level, link, link)
363 def _br_repl(self, word):
366 def _rule_repl(self, word):
367 return self._undent() + '\n<hr size="%d" noshade="noshade" />\n' % (len(word) - 2)
369 def _macro_repl(self, word):
370 m = re.compile("\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>").match(word)
374 argv.extend(m.group(2).split('|'))
375 argv = map(str.strip, argv)
377 macro = globals().get('_macro_' + name)
380 execfile("macros/" + name + ".py", globals())
382 if err.errno == errno.ENOENT: pass
383 macro = globals().get('_macro_' + name)
387 msg = '<<' + '|'.join(argv) + '>>'
389 msg = '<strong class="error">' + msg + '</strong>'
392 def _hurl_repl(self, word):
393 m = link_re.match(word)
394 return link_tag(m.group(1), m.group(2))
396 def _inl_repl(self, word):
397 (name, descr) = link_re.match(word).groups()
400 argv = descr.split('|')
404 args = '?' + '&'.join(argv)
406 # The "extthumb" nonsense works around a limitation of the HTML block model
407 return '<div class="extthumb"><div class="thumb">' \
408 + link_inline(name, descr, args) \
409 + '<div class="caption">' + descr + '</div></div></div>'
411 return link_inline(name, name)
413 def _html_repl(self, word):
415 return word; # Pass through
417 def _htmle_repl(self, word):
419 return word; # Pass through
421 def _ent_repl(self, s):
423 return s; # Pass through
424 return {'&': '&',
428 def _img_repl(self, word): # LEGACY
429 return self._inl_repl('{{' + word + '}}')
431 def _word_repl(self, word): # LEGACY
432 if self.in_html: return word # pass through
433 return link_tag(word)
435 def _url_repl(self, word): # LEGACY
436 if self.in_html: return word # pass through
437 return link_tag(word)
439 def _email_repl(self, word): # LEGACY
440 if self.in_html: return word # pass through
441 return '<a href="mailto:%s">%s</a>' % (word, word)
443 def _li_repl(self, match):
450 def _pre_repl(self, word):
451 if word == '{{{' and not self.in_pre:
459 def _hi_repl(self, word):
460 return '<strong class="highlight ' + word + '">' + word + '</strong>'
462 def _tr_repl(self, word):
464 if not self.in_table:
467 out = '</p><table><tbody>\n'
469 out = out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '">'
470 return out + ['<td>', '<th>'][word.strip() == '||=']
472 def _td_repl(self, word):
474 return ['</td><td>', '</th><th>'][word.strip() == '||=']
477 def _tre_repl(self, word):
479 return ['</td></tr>', '</th></tr>'][word.strip() == '||=']
482 def _indent_level(self):
483 return len(self.list_indents) and self.list_indents[-1]
485 def _indent_to(self, new_level):
486 if self._indent_level() == new_level:
489 while self._indent_level() > new_level:
490 del(self.list_indents[-1])
493 self.in_li = False # FIXME
495 while self._indent_level() < new_level:
496 self.list_indents.append(new_level)
503 res += '</ul>' * len(self.list_indents)
505 self.list_indents = []
508 def replace(self, match):
509 for rule, hit in match.groupdict().items():
511 return getattr(self, '_' + rule + '_repl')(hit)
513 raise "Can't handle match " + repr(match)
515 def print_html(self):
516 print '<div class="wiki"><p>'
518 scan_re = re.compile(r"""(?:
519 # Styles and formatting
520 (?P<b> \*\*|'''|//|''|\#\#|``|__|\^\^|,,)
524 | (?P<hi> \b( FIXME | TODO | DONE )\b )
527 | (?P<macro> \<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)
528 | (?P<hurl> \[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])
531 | (?P<html> <(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])\b )
532 | (?P<htmle> ( /\s*> | </(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])> ) )
535 # Auto links (LEGACY)
536 | (?P<img> \b[a-zA-Z0-9_/-]+\.(png|gif|jpg|jpeg|bmp|ico|ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt))
537 | (?P<word> \b(?:[A-Z][a-z]+){2,}\b)
538 | (?P<url> (http|https|ftp|mailto)\:[^\s'\"]+\S)
539 | (?P<email> [-\w._+]+\@[\w.-]+)
542 | (?P<li> ^\s+[\*\#]\s+)
543 | (?P<pre> \{\{\{|\s*\}\}\})
544 | (?P<inl> \{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})
547 | (?P<tr> ^\s*\|\|(=|)\s*)
548 | (?P<tre> \s*\|\|(=|)\s*$)
549 | (?P<td> \s*\|\|(=|)\s*)
551 # TODO: highlight search words (look at referrer)
553 pre_re = re.compile("""(?:
557 blank_re = re.compile(r"^\s*$")
558 indent_re = re.compile(r"^\s*")
559 tr_re = re.compile(r"^\s*\|\|")
560 eol_re = re.compile(r"\r?\n")
562 # For each line, we scan through looking for magic strings, outputting verbatim any intervening text
563 for self.line in eol_re.split(self.raw.expandtabs()):
566 if self.line.startswith('#'):
568 self.in_header = False
571 print re.sub(pre_re, self.replace, self.line)
573 if self.in_table and not tr_re.match(self.line):
574 self.in_table = False
575 print '</tbody></table><p>'
577 if blank_re.match(self.line):
580 indent = indent_re.match(self.line)
581 print self._indent_to(len(indent.group(0))) ,
582 print re.sub(scan_re, self.replace, self.line)
584 if self.in_pre: print '</pre>'
585 if self.in_table: print '</tbody></table><p>'
590 def __init__(self, page_name):
591 self.page_name = page_name
593 self.msg_type = 'error'
595 def split_title(self):
596 # look for the end of words and the start of a new word and insert a space there
597 return re.sub('([a-z])([A-Z])', r'\1 \2', self.page_name)
600 return os.path.join(data_dir, self.page_name)
602 def _tmp_filename(self):
603 return os.path.join(data_dir, ('#' + self.page_name.replace('/','_') + '.' + `os.getpid()` + '#'))
607 os.stat(self._filename())
610 if err.errno == errno.ENOENT:
614 def get_raw_body(self, default=None):
616 return open(self._filename(), 'rb').read()
618 if err.errno == errno.ENOENT:
620 default = '//[[?edit=%s|Describe %s]]//' % (self.page_name, self.page_name)
622 if err.errno == errno.EISDIR:
623 return self.format_dir()
626 def format_dir(self):
629 for dirname in self.page_name.split('/'):
630 pathname = (pathname + '/' + dirname) if pathname else dirname
631 out += '[[' + pathname + '|' + dirname + ']]/'
634 for filename in page_list(self._filename(), file_re):
635 if img_re.match(filename):
637 maxwidth_arg = '|maxwidth=' + str(image_maxwidth)
638 out += '{{' + self.page_name + '/' + filename + '|' + filename + maxwidth_arg + '}}\n'
640 out += ' * [[' + self.page_name + '/' + filename + ']]\n'
644 if not '_pragmas' in self.__dict__:
647 f = open(self._filename(), 'rt')
648 attr_re = re.compile(r"^#(\S*)(.*)$")
650 m = attr_re.match(line)
653 self._pragmas[m.group(1)] = m.group(2).strip()
654 #print "bernie: _pragmas[" + m.group(1) + "] = " + m.group(2) + "<br>\n"
656 if err.errno != errno.ENOENT and err.errno != errno.EISDIR:
660 def pragma(self, name, default):
661 return self.pragmas().get(name, default)
663 def can(self, action, default=True):
666 #acl SomeUser:read,write All:read
667 acl = self.pragma("acl", None)
668 for rule in acl.split():
669 (user, perms) = rule.split(':')
670 if user == remote_user() or user == "All":
671 return action in perms.split(',')
675 self.msg_text = 'Illegal acl line: ' + acl
679 return self.can("write", True)
682 return self.can("read", True)
684 def send_naked(self):
686 WikiFormatter(self.get_raw_body()).print_html()
688 send_guru("Read access denied by ACLs", "notice")
692 value = self.pragma("css", None)
695 link_urls += [ [ "stylesheet", value ] ]
697 send_title(self.page_name, self.split_title(),
698 msg_text=self.msg_text, msg_type=self.msg_type, writable=self.can_write())
700 send_footer(self._last_modified())
702 def _last_modified(self):
704 from time import localtime, strftime
705 modtime = localtime(os.stat(self._filename())[stat.ST_MTIME])
707 if err.errno != errno.ENOENT:
710 return strftime(datetime_fmt, modtime)
712 def send_editor(self, preview=None):
713 send_title(None, 'Edit ' + self.split_title(), msg_text=self.msg_text, msg_type=self.msg_type)
714 if not self.can_write():
715 send_guru("Write access denied by ACLs", "error")
720 filename = form['file'].value
722 print ('<p><b>Editing ' + self.page_name
723 + ' for ' + cgi.escape(remote_user())
724 + ' from ' + cgi.escape(get_hostname(remote_host()))
726 print '<div class="editor"><form name="editform" method="post" enctype="multipart/form-data" action="%s">' % relative_url(self.page_name)
727 print '<input type="hidden" name="edit" value="%s">' % (self.page_name)
728 print '<input type="input" id="editor" name="changelog" value="Edit page %s" accesskey="c" /><br />' % (self.page_name)
729 print '<textarea wrap="off" spellcheck="true" id="editor" name="savetext" rows="17" cols="100" accesskey="e">%s</textarea>' \
730 % cgi.escape(preview or self.get_raw_body(default=''))
731 print '<label for="file" accesskey="u">Or Upload a file:</label> <input type="file" name="file" value="%s" />' % filename
734 <input type="submit" name="save" value="Save" accesskey="s">
735 <input type="submit" name="preview" value="Preview" accesskey="p" />
736 <input type="reset" value="Reset" />
737 <input type="submit" name="cancel" value="Cancel" />
740 <script language="javascript">
742 document.editform.savetext.focus()
746 print "<p>" + link_tag('EditingTips') + "</p>"
748 print "<div class='preview'>"
749 WikiFormatter(preview).print_html()
753 def send_raw(self, mimetype='text/plain', args=[]):
754 if not self.can_read():
755 send_title(None, msg_text='Read access denied by ACLs', msg_type='notice')
758 if 'maxwidth' in args:
760 emit_header(mimetype)
762 subprocess.check_call(['gm', 'convert', self._filename(),
763 '-scale', args['maxwidth'].value + ' >', '-'])
765 body = self.get_raw_body()
766 emit_header(mimetype)
769 def _write_file(self, data):
770 tmp_filename = self._tmp_filename()
771 open(tmp_filename, 'wb').write(data)
772 name = self._filename()
774 # Bad Bill! POSIX rename ought to replace. :-(
778 if err.errno != errno.ENOENT: raise err
779 os.rename(tmp_filename, name)
781 def save(self, newdata, changelog):
782 if not self.can_write():
783 self.msg_text = 'Write access denied by ACLs'
784 self.msg_type = 'error'
787 self._write_file(newdata)
791 cmd = [ post_edit_hook, data_dir + '/' + self.page_name, remote_user(), remote_host(), changelog]
792 child = subprocess.Popen(cmd, stdout=subprocess.PIPE, close_fds=True)
793 output = child.stdout.read()
796 self.msg_text += "Post-editing hook returned %d. Command was:\n'%s'\n" % (rc, "' '".join(cmd))
798 self.msg_text += 'Output follows:\n' + output
800 self.msg_text = 'Thank you for your contribution. Your attention to detail is appreciated.'
801 self.msg_type = 'success'
805 handler = globals().get('handle_' + cmd)
807 handler(form[cmd].value)
810 query = query_string()
811 if file_re.match(query):
812 # FIMXE: this is all bullshit, MimeTypes bases its guess on the extension!
813 from mimetypes import MimeTypes
814 mimetype, encoding = MimeTypes().guess_type(query)
816 Page(query).send_raw(mimetype=mimetype, args=form)
820 send_httperror("403 Forbidden", query)
823 execfile("geekigeeki.conf.py")
824 form = cgi.FieldStorage()
828 msg_text = traceback.format_exc()
830 send_guru(msg_text, "error")
832 send_title(None, msg_text=msg_text)