2 # -*- coding: utf-8 -*-
4 # Copyright 1999, 2000 Martin Pool <mbp@humbug.org.au>
5 # Copyright 2002 Gerardo Poggiali
6 # Copyright 2007, 2008 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)$", re.IGNORECASE)
34 url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$")
35 link_re = re.compile("(?:\[\[|{{)([^\s\|]+)(?:\s*\|\s*([^\]]+)|)(?:\]\]|}})")
40 # CGI stuff ---------------------------------------------------------
43 return os.environ.get('SCRIPT_NAME', '')
45 def privileged_path():
46 return privileged_url or script_name()
49 user = os.environ.get('REMOTE_USER', '')
50 if user is None or user == '' or user == 'anonymous':
51 user = 'AnonymousCoward'
55 return os.environ.get('REMOTE_ADDR', '')
57 def get_hostname(addr):
59 from socket import gethostbyaddr
60 return gethostbyaddr(addr)[0] + ' (' + addr + ')'
64 def relative_url(pathname, privileged=False):
65 if not (url_re.match(pathname) or pathname.startswith('/')):
67 url = privileged_path()
70 pathname = url + '/' + pathname
74 return re.sub(' ', '-', re.sub('[^a-z0-9_ ]', '', s.lower()).strip())
76 # Formatting stuff --------------------------------------------------
77 def emit_header(mime_type="text/html"):
78 print "Content-type: " + mime_type + "; charset=utf-8\n"
80 def send_guru(msg_text, msg_type):
81 if not msg_text: return
82 print '<pre id="guru" onclick="this.style.display = \'none\'" class="' + msg_type + '">'
83 if msg_type == 'error':
84 print ' Software Failure. Press left mouse button to continue.\n'
86 if msg_type == 'error':
87 print '\n Guru Meditation #DEADBEEF.ABADC0DE'
89 # FIXME: This little JS snippet is harder to pass than ACID 3.0
91 <script language="JavaScript" type="text/javascript">
92 var guru = document.getElementById('guru');
93 // Firefox 2.0 doesn't take border-color, but returns border-top-color fine
94 var color = document.defaultView.getComputedStyle(guru,null).getPropertyValue('border-top-color');
97 guru.style.setProperty('border-color', color, '');
98 setTimeout('guruOff()', 1000);
101 guru.style.setProperty('border-color', '#000000', '');
102 setTimeout('guruOn()', 1000);
104 // Safari 2.0 returns this rgba crap
105 // Konqueror 3.5.6 doesn't seem to support computed properties
106 if (color && color != 'rgba(0, 0, 0, 0)') {
107 //window.alert("enabled! color='" + color + "'");
112 def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=False):
114 if title_done: return
118 print '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'
119 print ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">'
120 print '<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">'
122 print "<head><title>%s: %s</title>" % (site_name, text)
123 print ' <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8" />'
125 print ' <meta name="robots" content="noindex,nofollow" />'
127 for meta in meta_urls:
128 http_equiv, content = meta
129 print ' <meta http-equiv="%s" content="%s" />' % (http_equiv, relative_url(content))
131 for link in link_urls:
133 print ' <link rel="%s" href="%s" />' % (rel, relative_url(href))
135 if name and writable and privileged_url is not None:
136 print ' <link rel="alternate" type="application/x-wiki" title="Edit this page" href="%s" />' \
137 % (privileged_path() + '?edit=' + name)
139 if history_url is not None:
140 print ' <link rel="alternate" type="application/rss+xml" title="RSS" href="%s" />' \
141 % relative_url(history_url + '?a=rss')
146 if name and writable and privileged_url is not None:
147 print '<body ondblclick="location.href=\'' + privileged_path() + '?edit=' + name + '\'">'
152 send_guru(msg_text, msg_type)
155 print '<div class="nav">'
157 print ' <b>' + link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
159 print ' <b>' + text + '</b> '
160 print ' | ' + link_tag('FrontPage', 'Home', 'navlink')
161 print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
162 if 'history_url' in globals():
163 print ' | <a href="' + relative_url(history_url) + '" class="navlink">Recent Changes</a>'
165 print ' | <a href="' + relative_url(history_url + '?a=history;f=' + name) + '" class="navlink">Page History</a>'
168 print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
169 if privileged_url is not None:
171 print ' | ' + link_tag('?edit=' + name, 'Edit', 'navlink', privileged=True)
173 print ' | ' + link_tag(name, 'Login', 'navlink', privileged=True)
176 print ' | <i>Immutable Page</i>'
179 if user != 'AnonymousCoward':
180 print ' | <span class="login"><i>logged in as <b>' + cgi.escape(user) + '</b></i></span>'
184 def send_httperror(status="403 Not Found", query=""):
185 print "Status: %s" % status
186 send_title(None, msg_text=("%s: on query '%s'" % (status, query)))
189 def link_tag(params, text=None, ss_class=None, privileged=False):
191 text = params # default
194 classattr += 'class="%s" ' % ss_class
195 # Prevent crawlers from following links potentially added by spammers or to generated pages
196 if ss_class == 'external' or ss_class == 'navlink':
197 classattr += 'rel="nofollow" '
198 elif url_re.match(params):
199 classattr += 'rel="nofollow" '
200 return '<a %shref="%s">%s</a>' % (classattr, relative_url(params, privileged=privileged), text)
202 # Search ---------------------------------------------------
204 def handle_fullsearch(needle):
205 send_title(None, 'Full text search for "%s"' % (needle))
207 needle_re = re.compile(needle, re.IGNORECASE)
209 all_pages = page_list()
210 for page_name in all_pages:
211 body = Page(page_name).get_raw_body()
212 count = len(needle_re.findall(body))
214 hits.append((count, page_name))
216 # The default comparison for tuples compares elements in order,
217 # so this sorts by number of hits
222 for (count, page_name) in hits:
223 print '<li><p>' + Page(page_name).link_to()
224 print ' . . . . ' + `count`
225 print ['match', 'matches'][count != 1]
229 print_search_stats(len(hits), len(all_pages))
231 def handle_titlesearch(needle):
232 # TODO: check needle is legal -- but probably we can just accept any RE
233 send_title(None, "Title search for \"" + needle + '"')
235 needle_re = re.compile(needle, re.IGNORECASE)
236 all_pages = page_list()
237 hits = filter(needle_re.search, all_pages)
240 for filename in hits:
241 print '<li><p>' + Page(filename).link_to() + "</p></li>"
244 print_search_stats(len(hits), len(all_pages))
246 def print_search_stats(hits, searched):
247 print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
249 def handle_raw(pagename):
250 if not file_re.match(pagename):
251 send_httperror("403 Forbidden", pagename)
254 Page(pagename).send_raw()
256 def handle_edit(pagename):
257 if not file_re.match(pagename):
258 send_httperror("403 Forbidden", pagename)
263 if form['file'].value:
264 pg.save(form['file'].file.read(), form['changelog'].value)
266 pg.save(form['savetext'].value.replace('\r\n', '\n'), form['changelog'].value)
268 elif 'cancel' in form:
269 pg.msg_text = 'Editing canceled'
270 pg.msg_type = 'notice'
272 else: # preview or edit
274 if 'preview' in form:
275 text = form['savetext'].value
278 def make_index_key():
279 links = map(lambda ch: '<a href="#%s">%s</a>' % (ch, ch), 'abcdefghijklmnopqrstuvwxyz')
280 return '<p><center>'+ ' | '.join(links) + '</center></p>'
282 def page_list(dirname = None, re = word_re):
283 return sorted(filter(re.match, os.listdir(dirname or data_dir)))
285 def send_footer(mod_string=None):
286 if globals().get('debug_cgi', False):
287 cgi.print_arguments()
291 <div id="footer"><hr />
292 <p class="copyright">
293 <a rel="license" href="http://creativecommons.org/licenses/by-sa/3.0/"><img class="license" alt="Creative Commons License" src="http://i.creativecommons.org/l/by-sa/3.0/80x15.png" /></a>
294 <span class="benchmark">generated in %0.3fs</span> by <a href="http://www.codewiz.org/wiki/GeekiGeeki">GeekiGeeki</a> version %s
296 ''' % (clock() - start_time, __version__)
298 print '<p class="modified">last modified %s</p>' % mod_string
299 print '</div></body></html>'
302 """Object that turns Wiki markup into HTML.
304 All formatting commands can be parsed one line at a time, though
305 some state is carried over between lines.
307 def __init__(self, raw):
310 self.in_pre = self.in_html = self.in_table = self.in_li = False
311 self.in_header = True
312 self.list_indents = []
323 "^^": ["sup", False],
327 def _b_repl(self, word):
328 style = self.styles[word]
329 style[1] = not style[1]
330 return ['</', '<'][style[1]] + style[0] + '>'
332 def _tit_repl(self, word):
334 result = '</h%d><p>\n' % self.h_level
337 self.h_level = len(word) - 1
338 link = permalink(self.line)
339 result = '\n</p><h%d id="%s"><a class="heading" href="#%s">¶</a> ' % (self.h_level, link, link)
342 def _br_repl(self, word):
345 def _rule_repl(self, word):
346 return self._undent() + '\n<hr size="%d" noshade="noshade" />\n' % (len(word) - 2)
348 def _word_repl(self, word):
349 return Page(word).link_to()
351 def _img_repl(self, word):
352 pathname = relative_url(word)
353 return '<a href="%s"><img border="0" src="%s" /></a>' % (pathname, pathname)
355 def _url_repl(self, word):
356 if img_re.match(word):
357 return '<a href="%s"><img border="0" src="%s" /></a>' % (word, word)
359 return '<a href="%s" rel="nofollow" class="external">%s</a>' % (word, word)
361 def _macro_repl(self, word):
362 m = re.compile("\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>").match(word)
366 argv.extend(m.group(2).split('|'))
367 argv = map(str.strip, argv)
369 macro = globals().get('_macro_' + name)
372 execfile("macros/" + name + ".py", globals())
374 if err.errno == errno.ENOENT: pass
375 macro = globals().get('_macro_' + name)
379 return '<strong class="error"><<' + '|'.join(argv) + '>></strong>'
381 def _hurl_repl(self, word):
382 m = link_re.match(word)
387 elif img_re.match(m.group(2)):
388 descr = '<img border="0" src="' + descr + '" />'
390 return link_tag(name, descr, 'wikilink')
392 def _inl_repl(self, word):
393 m = link_re.match(word)
395 descr = m.group(2) or name
396 name = relative_url(name)
397 argv = descr.split('|')
401 args = '?' + '&'.join(argv)
406 # The "extthumb" nonsense works around a limitation of the HTML block model
407 return '<div class="extthumb"><div class="thumb"><a href="%s"><img border="0" src="%s" alt="%s" /></a><div class="caption">%s</div></div></div>' \
408 % (name, name + args, descr, descr)
410 return '<a href="%s"><img border="0" src="%s" /></a>' % (name, name + args)
412 def _email_repl(self, word):
413 return '<a href="mailto:%s">%s</a>' % (word, word)
415 def _html_repl(self, word):
417 return word; # Pass through
419 def _ent_repl(self, s):
420 if self.in_html and s == '>':
423 return {'&': '&',
427 def _li_repl(self, match):
434 def _pre_repl(self, word):
435 if word == '{{{' and not self.in_pre:
443 def _hi_repl(self, word):
444 return '<strong class="highlight ' + word + '">' + word + '</strong>'
446 def _tr_repl(self, word):
448 if not self.in_table:
451 out = '</p><table><tbody>\n'
453 out = out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '">'
454 return out + ['<td>', '<th>'][word.strip() == '||=']
456 def _td_repl(self, word):
458 return ['</td><td>', '</th><th>'][word.strip() == '||=']
461 def _tre_repl(self, word):
463 return ['</td></tr>', '</th></tr>'][word.strip() == '||=']
466 def _indent_level(self):
467 return len(self.list_indents) and self.list_indents[-1]
469 def _indent_to(self, new_level):
470 if self._indent_level() == new_level:
473 while self._indent_level() > new_level:
474 del(self.list_indents[-1])
477 self.in_li = False # FIXME
479 while self._indent_level() < new_level:
480 self.list_indents.append(new_level)
487 res += '</ul>' * len(self.list_indents)
489 self.list_indents = []
492 def replace(self, match):
493 for rule, hit in match.groupdict().items():
495 return getattr(self, '_' + rule + '_repl')(hit)
497 raise "Can't handle match " + repr(match)
499 def print_html(self):
500 print '<div class="wiki"><p>'
502 # For each line, we scan through looking for magic
503 # strings, outputting verbatim any intervening text
504 # TODO: highlight search words (look at referrer)
505 scan_re = re.compile(
508 + r"(?P<b>\*\*|'''|//|''|##|``|__|\^\^|,,)"
509 + r"|(?P<tit>\={2,6})"
511 + r"|(?P<rule>^-{3,})"
512 + r"|(?P<hi>\b(FIXME|TODO|DONE)\b)"
515 + r"|(?P<macro>\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)"
516 + r"|(?P<hurl>\[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])"
519 + r"|(?P<html><(/|)(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])[^>]*>)"
522 # Auto links (LEGACY)
523 + r"|(?P<img>\b[a-zA-Z0-9_/-]+\.(png|gif|jpg|jpeg|bmp|ico))"
524 + r"|(?P<word>\b(?:[A-Z][a-z]+){2,}\b)"
525 + r"|(?P<url>(http|https|ftp|mailto)\:[^\s'\"]+\S)"
526 + r"|(?P<email>[-\w._+]+\@[\w.-]+)"
529 + r"|(?P<li>^\s+[\*#] +)"
530 + r"|(?P<pre>\{\{\{|\s*\}\}\})"
531 + r"|(?P<inl>\{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})"
534 + r"|(?P<tr>^\s*\|\|(=|)\s*)"
535 + r"|(?P<tre>\s*\|\|(=|)\s*$)"
536 + r"|(?P<td>\s*\|\|(=|)\s*)"
540 + r"(?P<pre>\s*\}\}\})"
543 blank_re = re.compile(r"^\s*$")
544 indent_re = re.compile(r"^\s*")
545 tr_re = re.compile(r"^\s*\|\|")
546 eol_re = re.compile(r"\r?\n")
547 for self.line in eol_re.split(self.raw.expandtabs()):
550 if self.line.startswith('#'):
552 self.in_header = False
555 print re.sub(pre_re, self.replace, self.line)
557 if self.in_table and not tr_re.match(self.line):
558 self.in_table = False
559 print '</tbody></table><p>'
561 if blank_re.match(self.line):
564 indent = indent_re.match(self.line)
565 print self._indent_to(len(indent.group(0))) ,
566 print re.sub(scan_re, self.replace, self.line)
568 if self.in_pre: print '</pre>'
569 if self.in_table: print '</tbody></table><p>'
574 def __init__(self, page_name):
575 self.page_name = page_name
577 self.msg_type = 'error'
579 def split_title(self):
580 # look for the end of words and the start of a new word and insert a space there
581 return re.sub('([a-z])([A-Z])', r'\1 \2', self.page_name)
584 return os.path.join(data_dir, self.page_name)
586 def _tmp_filename(self):
587 return os.path.join(data_dir, ('#' + self.page_name.replace('/','_') + '.' + `os.getpid()` + '#'))
591 os.stat(self._filename())
594 if err.errno == errno.ENOENT:
599 word = self.page_name
601 return link_tag(word, word, 'wikilink')
603 return link_tag(word, nonexist_pfx + word, 'nonexistent')
605 def get_raw_body(self):
607 return open(self._filename(), 'rb').read()
609 if err.errno == errno.ENOENT:
610 return '' # just doesn't exist, use default
611 if err.errno == errno.EISDIR:
612 return self.format_dir()
615 def format_dir(self):
618 for dirname in self.page_name.split('/'):
619 pathname = (pathname + '/' + dirname) if pathname else dirname
620 out += '[[' + pathname + '|' + dirname + ']]/'
623 for filename in page_list(self._filename(), file_re):
624 if img_re.match(filename):
626 maxwidth_arg = '|maxwidth=' + str(image_maxwidth)
627 out += '{{' + self.page_name + '/' + filename + '|' + filename + maxwidth_arg + '}}\n'
629 out += ' * [[' + self.page_name + '/' + filename + ']]\n'
633 if not '_pragmas' in self.__dict__:
636 f = open(self._filename(), 'rt')
637 attr_re = re.compile(r"^#(\S*)(.*)$")
639 m = attr_re.match(line)
642 self._pragmas[m.group(1)] = m.group(2).strip()
643 #print "bernie: _pragmas[" + m.group(1) + "] = " + m.group(2) + "<br>\n"
645 if err.errno != errno.ENOENT and err.errno != errno.EISDIR:
649 def pragma(self, name, default):
650 return self.pragmas().get(name, default)
652 def can(self, action, default=True):
655 #acl SomeUser:read,write All:read
656 acl = self.pragma("acl", None)
657 for rule in acl.split():
658 (user, perms) = rule.split(':')
659 if user == remote_user() or user == "All":
660 return action in perms.split(',')
664 self.msg_text = 'Illegal acl line: ' + acl
668 return self.can("write", True)
671 return self.can("read", True)
673 def send_naked(self):
675 WikiFormatter(self.get_raw_body()).print_html()
677 send_guru("Read access denied by ACLs", "notice")
681 value = self.pragma("css", None)
684 link_urls += [ [ "stylesheet", value ] ]
686 send_title(self.page_name, self.split_title(),
687 msg_text=self.msg_text, msg_type=self.msg_type, writable=self.can_write())
689 send_footer(self._last_modified())
691 def _last_modified(self):
693 from time import localtime, strftime
694 modtime = localtime(os.stat(self._filename())[stat.ST_MTIME])
696 if err.errno != errno.ENOENT:
699 return strftime(datetime_fmt, modtime)
701 def send_editor(self, preview=None):
702 send_title(None, 'Edit ' + self.split_title(), msg_text=self.msg_text, msg_type=self.msg_type)
703 if not self.can_write():
704 send_guru("Write access denied by ACLs", "error")
709 filename = form['file'].value
711 print ('<p><b>Editing ' + self.page_name
712 + ' for ' + cgi.escape(remote_user())
713 + ' from ' + cgi.escape(get_hostname(remote_host()))
715 print '<div class="editor"><form name="editform" method="post" enctype="multipart/form-data" action="%s">' % relative_url(self.page_name)
716 print '<input type="hidden" name="edit" value="%s">' % (self.page_name)
717 print '<input type="input" id="editor" name="changelog" value="Edit page %s" accesskey="c" /><br />' % (self.page_name)
718 print '<textarea wrap="off" spellcheck="true" id="editor" name="savetext" rows="17" cols="100" accesskey="e">%s</textarea>' % cgi.escape(preview or self.get_raw_body())
719 print '<label for="file" accesskey="u">Or Upload a file:</label> <input type="file" name="file" value="%s" />' % filename
722 <input type="submit" name="save" value="Save" accesskey="s">
723 <input type="submit" name="preview" value="Preview" accesskey="p" />
724 <input type="reset" value="Reset" />
725 <input type="submit" name="cancel" value="Cancel" />
728 <script language="javascript">
730 document.editform.savetext.focus()
734 print "<p>" + Page('EditingTips').link_to() + "</p>"
736 print "<div class='preview'>"
737 WikiFormatter(preview).print_html()
741 def send_raw(self, mimetype='text/plain'):
743 body = self.get_raw_body()
744 emit_header(mimetype)
747 send_title(None, msg_text='Read access denied by ACLs', msg_type='notice')
749 def send_image(self, mimetype, args=[]):
750 if 'maxwidth' in args:
752 emit_header(mimetype)
754 subprocess.check_call(['gm', 'convert', self._filename(),
755 '-scale', args['maxwidth'].value + ' >', '-'])
757 self.send_raw(mimetype)
759 def _write_file(self, data):
760 tmp_filename = self._tmp_filename()
761 open(tmp_filename, 'wb').write(data)
762 name = self._filename()
764 # Bad Bill! POSIX rename ought to replace. :-(
768 if err.errno != errno.ENOENT: raise err
769 os.rename(tmp_filename, name)
771 def save(self, newdata, changelog):
772 if not self.can_write():
773 self.msg_text = 'Write access denied by ACLs'
774 self.msg_type = 'error'
777 self._write_file(newdata)
780 # FIXME: what's the std way to perform shell quoting in python?
781 cmd = ( post_edit_hook
782 + " '" + data_dir + '/' + self.page_name
783 + "' '" + remote_user()
784 + "' '" + remote_host()
785 + "' '" + changelog + "'"
791 self.msg_text += "Post-editing hook returned %d.\n" % rc
792 self.msg_text += 'Command was: ' + cmd + '\n'
794 self.msg_text += 'Output follows:\n' + output
796 self.msg_text = 'Thank you for your contribution. Your attention to detail is appreciated.'
797 self.msg_type = 'success'
801 handler = globals().get('handle_' + cmd)
803 handler(form[cmd].value)
806 path_info = os.environ.get('PATH_INFO', '')
807 if len(path_info) and path_info[0] == '/':
808 query = path_info[1:] or 'FrontPage'
810 query = os.environ.get('QUERY_STRING', '') or 'FrontPage'
812 if file_re.match(query):
813 if word_re.match(query):
816 from mimetypes import MimeTypes
817 mimetype, encoding = MimeTypes().guess_type(query)
819 if mimetype.startswith('image/'):
820 Page(query).send_image(mimetype=mimetype, args=form)
822 Page(query).send_raw(mimetype=mimetype)
826 send_httperror("403 Forbidden", query)
829 execfile("geekigeeki.conf.py")
830 form = cgi.FieldStorage()
834 msg_text = traceback.format_exc()
836 send_guru(msg_text, "error")
838 send_title(None, msg_text=msg_text)