2 # -*- coding: utf-8 -*-
4 # Copyright 1999, 2000 Martin Pool <mbp@humbug.org.au>
5 # Copyright 2002 Gerardo Poggiali
6 # Copyright 2007 Bernardo 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, string, os, re, errno, stat
27 from os import path, environ
29 # Regular expression defining a WikiWord
30 # (but this definition is also assumed in other places)
31 file_re = re.compile(r"^\b([A-Za-z0-9_\.\-]+)\b$")
32 word_re = re.compile(r"^\b([A-Z][a-z]+){2,}\b$")
33 img_re = re.compile(r"^.*\.(png|gif|jpg|jpeg)$", re.IGNORECASE)
34 url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$")
39 # CGI stuff ---------------------------------------------------------
42 return environ.get('SCRIPT_NAME', '')
44 def privileged_path():
45 return privileged_url or script_name()
48 user = environ.get('REMOTE_USER', '')
49 if user is None or user == '' or user == 'anonymous':
50 user = 'AnonymousCoward'
54 return environ.get('REMOTE_ADDR', '')
56 def get_hostname(addr):
58 from socket import gethostbyaddr
59 return gethostbyaddr(addr)[0] + ' (' + addr + ')'
63 # Formatting stuff --------------------------------------------------
65 def emit_header(type="text/html"):
66 print "Content-type: " + type + "; charset=utf-8"
69 def send_guru(msg, msg_type):
70 if msg is None or msg == '': return
71 print '<pre id="guru" onclick="this.style.display = \'none\'" class="' + msg_type + '">'
72 if msg_type == 'error':
73 print ' Software Failure. Press left mouse button to continue.\n'
75 if msg_type == 'error':
76 print ' Guru Meditation #DEADBEEF.ABADC0DE'
78 # FIXME: This simple JS snippet is harder to pass than ACID 3.0
80 <script language="JavaScript" type="text/javascript">
81 var guru = document.getElementById('guru');
82 // Firefox 2.0 doesn't take border-color, but returns border-top-color fine
83 var color = document.defaultView.getComputedStyle(guru,null).getPropertyValue('border-top-color');
86 guru.style.setProperty('border-color', color, '');
87 setTimeout('guruOff()', 1000);
90 guru.style.setProperty('border-color', '#000000', '');
91 setTimeout('guruOn()', 1000);
93 // Safari 2.0 returns this rgba crap
94 // Konqueror 3.5.6 doesn't seem to support computed properties
95 if (color && color != 'rgba(0, 0, 0, 0)') {
96 //window.alert("enabled! color='" + color + "'");
101 def send_title(name, text="Limbo", msg=None, msg_type='error'):
103 if title_done: return
107 print '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'
108 print ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">'
109 print '<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">'
111 site_name = globals().get('site_name', 'Unconfigured Site')
112 print "<head><title>%s: %s</title>" % (site_name, text)
113 print ' <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8" />'
115 print ' <meta name="robots" content="noindex,nofollow" />'
117 print ' <link rel="stylesheet" type="text/css" href="%s" />' % css
121 if name and privileged_url is not None:
122 print '<body ondblclick="location.href=\'' + privileged_path() + '?edit=' + name + '\'">'
127 send_guru(msg, msg_type)
130 print '<div class="nav">'
131 print ' <b>' + site_name + ': ',
133 print link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
136 print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink')
137 print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
138 if globals().has_key('history_url'):
139 print ' | <a href="' + history_url + '" class="navlink">Recent Changes</a>'
141 print ' | <a href="' + history_url + '?a=history;f=' + name + '" class="navlink">Page History</a>'
144 print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
145 if privileged_url is not None:
146 print ' | ' + link_tag('?edit=' + name, 'Edit Page', 'navlink', authentication=True)
148 print ' | <i>Immutable Page</i>'
151 if user != 'AnonymousCoward':
152 print ' | <span class="login"><i>logged in as <b>' + cgi.escape(user) + '</b></i></span>'
156 def link_tag(params, text=None, ss_class=None, authentication=False):
158 text = params # default
161 classattr += 'class="%s" ' % ss_class
162 # Prevent crawlers from following links potentially added by spammers or to generated pages
163 if ss_class == 'external' or ss_class == 'navlink':
164 classattr += 'rel="nofollow" '
166 path = privileged_path()
169 return '<a %shref="%s/%s">%s</a>' % (classattr, path, params, text)
171 # Search ---------------------------------------------------
173 def do_fullsearch(needle):
174 send_title(None, 'Full text search for "%s"' % (needle))
176 needle_re = re.compile(needle, re.IGNORECASE)
178 all_pages = page_list()
179 for page_name in all_pages:
180 body = Page(page_name).get_raw_body()
181 count = len(needle_re.findall(body))
183 hits.append((count, page_name))
185 # The default comparison for tuples compares elements in order,
186 # so this sorts by number of hits
191 for (count, page_name) in hits:
192 print '<li><p>' + Page(page_name).link_to()
193 print ' . . . . ' + `count`
194 print ['match', 'matches'][count != 1]
198 print_search_stats(len(hits), len(all_pages))
200 def do_titlesearch(needle):
201 # TODO: check needle is legal -- but probably we can just accept any RE
202 send_title(None, "Title search for \"" + needle + '"')
204 needle_re = re.compile(needle, re.IGNORECASE)
205 all_pages = page_list()
206 hits = filter(needle_re.search, all_pages)
209 for filename in hits:
210 print '<li><p>' + Page(filename).link_to() + "</p></li>"
213 print_search_stats(len(hits), len(all_pages))
215 def print_search_stats(hits, searched):
216 print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
218 #TODO: merge into do_savepage()
219 def do_edit(pagename):
220 Page(pagename).send_editor()
222 def do_raw(pagename):
223 Page(pagename).send_raw()
225 def do_savepage(pagename):
228 if 'preview' in form:
229 pg.send_editor(form['savetext'].value)
231 pg.save_text(form['savetext'].value)
233 elif 'cancel' in form:
234 pg.msg = 'Editing canceled'
235 pg.msg_type = 'notice'
238 raise 'What did you press?'
240 def make_index_key():
242 links = map(lambda ch: '<a href="#%s">%s</a>' % (ch, ch),
244 s = s + string.join(links, ' | ')
245 s = s + '</center></p>'
249 return filter(word_re.match, os.listdir(data_dir))
251 def send_footer(name, mod_string=None):
252 if globals().get('debug_cgi', False):
253 cgi.print_arguments()
254 cgi.print_form(cgi.FieldStorage())
257 print '<div id="footer"><hr />'
258 print ('<p class="copyright"><span class="benchmark">generated in %0.3fs</span>' +
259 ' by <a href="http://www.codewiz.org/wiki/GeekiGeeki">GeekiGeeki</a>' +
260 ' version %s</p>') % (clock() - start_time, __version__)
262 print '<p class="modified">last modified %s</p>' % mod_string
263 print '</div></body></html>'
265 # ----------------------------------------------------------
267 def _macro_TitleSearch(*vargs):
268 return _macro_search("titlesearch")
270 def _macro_FullSearch(*vargs):
271 return _macro_search("fullsearch")
273 def _macro_search(type):
274 if form.has_key('value'):
275 default = form["value"].value
278 return """<form method="get"><input name="%s" size="30" value="%s" /><input type="submit" value="Search" /></form>""" % (type, default)
280 def _macro_WordIndex(*vargs):
282 pages = list(page_list())
284 word_re = re.compile('[A-Z][a-z]+')
286 for word in word_re.findall(name):
288 map[word].append(name)
292 all_words = map.keys()
296 for word in all_words:
297 letter = string.lower(word[0])
298 if letter != last_letter:
299 s = s + '; <a name="%s"><h3>%s</h3></a>' % (letter, letter)
302 s = s + '<b>%s</b><ul>' % word
307 if name == last_page: continue
308 s = s + '<li>' + Page(name).link_to()
313 def _macro_TitleIndex(*vargs):
315 pages = list(page_list())
317 current_letter = None
319 letter = string.lower(name[0])
320 if letter != current_letter:
321 s = s + '<a name="%s"><h3>%s</h3></a>' % (letter, letter)
322 current_letter = letter
325 s = s + Page(name).link_to()
329 # ----------------------------------------------------------
331 """Object that turns Wiki markup into HTML.
333 All formatting commands can be parsed one line at a time, though
334 some state is carried over between lines.
336 def __init__(self, raw):
339 self.in_pre = self.in_table = self.in_var = self.in_em = self.in_b = False
340 self.in_header = True
341 self.list_indents = []
342 self.tr_cnt = self.h_cnt = 0
344 def _b_repl(self, word):
345 self.in_b = not self.in_b
346 return ['</strong>', '<strong>'][self.in_b]
348 def _em_repl(self, word):
349 self.in_em = not self.in_em
350 return ['</em>', '<em>'][self.in_em]
352 def _tit_repl(self, word):
354 result = '</h%d>' % self.h_level
357 self.h_level = len(word) - 1
359 #abridged = re.sub('[^a-z_]', '', word.lower().replace(' ', '_'))
360 result = '<h%d id="%d"><a class="heading" href="#%d">¶</a> ' % (self.h_level, self.h_cnt, self.h_cnt)
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 _word_repl(self, word):
370 return Page(word).link_to()
372 def _img_repl(self, word):
373 path = script_name() + '/' + word;
374 return '<a href="%s"><img border="0" src="%s" /></a>' % (path, path)
376 def _url_repl(self, word):
377 if img_re.match(word):
378 return '<a href="%s"><img border="0" src="%s" /></a>' % (word, word)
380 return '<a href="%s" rel="nofollow" class="external">%s</a>' % (word, word)
382 def _hurl_repl(self, word):
383 m = re.compile("\[\[(\S+)(?:\s*\|\s*([^\]]*)|)\]\]").match(word)
385 descr = m.group(2) or name
387 macro = globals().get('_macro_' + name)
389 return apply(macro, (name, descr))
390 elif img_re.match(name):
391 return '<a href="%s"><img border="0" src="%s" alt="%s" /></a>' % (name, name, descr)
392 elif url_re.match(name):
393 return '<a href="%s" rel="nofollow" class="external">%s</a>' % (name, descr)
394 elif name.startswith('/'):
395 return '<a href="%s">%s</a>' % (name, descr)
397 return link_tag(name, descr, 'wikilink')
399 def _email_repl(self, word):
400 return '<a href="mailto:%s">%s</a>' % (word, word)
402 def _html_repl(self, word):
403 return word; # Pass through
405 def _ent_repl(self, s):
406 return {'&': '&',
410 def _li_repl(self, match):
413 def _pre_repl(self, word):
414 if word == '{{{' and not self.in_pre:
422 def _hi_repl(self, word):
429 return '<strong class="highlight ' + cl + '">' + word + '</strong>'
431 def _var_repl(self, word):
432 if word == '{{' and not self.in_var:
440 def _tr_repl(self, word):
442 if not self.in_table:
445 out = '</p><table><tbody>\n'
447 out = out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '">'
448 return out + ['<td>', '<th>'][word.strip() == '||=']
450 def _td_repl(self, word):
452 return ['</td><td>', '</th><th>'][word.strip() == '||=']
455 def _tre_repl(self, word):
457 return ['</td></tr>', '</th></tr>'][word.strip() == '||=']
460 def _indent_level(self):
461 return len(self.list_indents) and self.list_indents[-1]
463 def _indent_to(self, new_level):
464 if self._indent_level() == new_level:
467 while self._indent_level() > new_level:
468 del(self.list_indents[-1])
470 while self._indent_level() < new_level:
471 self.list_indents.append(new_level)
478 res += '</ul>' * len(self.list_indents)
480 self.list_indents = []
483 def replace(self, match):
484 for type, hit in match.groupdict().items():
486 return apply(getattr(self, '_' + type + '_repl'), (hit,))
488 raise "Can't handle match " + `match`
490 def print_html(self):
491 print '<div class="wiki"><p>'
493 # For each line, we scan through looking for magic
494 # strings, outputting verbatim any intervening text
495 # TODO: highlight search words (look at referer)
496 scan_re = re.compile(
501 + r"|(?P<tit>\={2,6})"
503 + r"|(?P<rule>^-{3,})"
504 + r"|(?P<html><(/|)(div|span|iframe)[^<>]*>)"
506 + r"|(?P<hi>\b(FIXME|TODO|DONE)\b)"
509 + r"|(?P<img>\b[a-zA-Z0-9_-]+\.(png|gif|jpg|jpeg|bmp))"
510 + r"|(?P<word>\b(?:[A-Z][a-z]+){2,}\b)"
511 + r"|(?P<hurl>\[\[(\S+)(?:\s*\|\s*([^\]]*)|)\]\])"
512 + r"|(?P<url>(http|https|ftp|mailto)\:[^\s'\"]+\S)"
513 + r"|(?P<email>[-\w._+]+\@[\w.-]+)"
516 + r"|(?P<li>^\s+[\*#] +)"
517 + r"|(?P<pre>\{\{\{|\s*\}\}\})"
518 + r"|(?P<var>\{\{|\}\})"
521 + r"|(?P<tr>^\s*\|\|(=|)\s*)"
522 + r"|(?P<tre>\s*\|\|(=|)\s*$)"
523 + r"|(?P<td>\s*\|\|(=|)\s*)"
527 + r"(?P<pre>\s*\}\}\})"
529 blank_re = re.compile(r"^\s*$")
530 indent_re = re.compile(r"^\s*")
531 tr_re = re.compile(r"^\s*\|\|")
532 eol_re = re.compile(r"\r?\n")
533 raw = string.expandtabs(self.raw)
534 for line in eol_re.split(raw):
537 if line.startswith('#'):
539 self.in_header = False
542 print re.sub(pre_re, self.replace, line)
544 if self.in_table and not tr_re.match(line):
545 self.in_table = False
546 print '</tbody></table><p>'
548 if blank_re.match(line):
551 indent = indent_re.match(line)
552 print self._indent_to(len(indent.group(0)))
553 print re.sub(scan_re, self.replace, line)
555 if self.in_pre: print '</pre>'
556 if self.in_table: print '</tbody></table><p>'
560 # ----------------------------------------------------------
562 def __init__(self, page_name):
563 self.page_name = page_name
565 self.msg_type = 'error'
567 def split_title(self):
568 # look for the end of words and the start of a new word,
569 # and insert a space there
570 return re.sub('([a-z])([A-Z])', r'\1 \2', self.page_name)
572 def _text_filename(self):
573 return path.join(data_dir, self.page_name)
575 def _tmp_filename(self):
576 return path.join(data_dir, ('#' + self.page_name + '.' + `os.getpid()` + '#'))
580 os.stat(self._text_filename())
583 if er.errno == errno.ENOENT:
589 word = self.page_name
591 return link_tag(word, word, 'wikilink')
593 return link_tag(word, nonexist_pfx + word, 'nonexistent')
595 def get_raw_body(self):
597 return open(self._text_filename(), 'rt').read()
599 if er.errno == errno.ENOENT:
600 return '' # just doesn't exist, use default
604 if self.__dict__.has_key('attrs'):
608 file = open(self._text_filename(), 'rt')
609 attr_re = re.compile(r"^#(\S*)(.*)$")
611 m = attr_re.match(line)
614 self.attrs[m.group(1)] = m.group(2).strip()
615 #print "bernie: attrs[" + m.group(1) + "] = " + m.group(2) + "<br>\n"
617 if er.errno != errno.ENOENT:
621 def get_attr(self, name, default):
622 if self.get_attrs().has_key(name):
623 return self.get_attrs()[name]
627 def can(self, action, default=True):
629 #acl SomeUser:read,write All:read
630 acl = self.get_attr("acl", None)
631 for rule in acl.split():
632 (user,perms) = rule.split(':')
633 if user == remote_user() or user == "All":
634 if action in perms.split(','):
639 except Exception, er:
644 return self.can("write", True)
647 return self.can("read", True)
652 page_name = self.page_name
654 #FIXME: are there security implications?
657 css_url = css_url + self.get_attr("css", "").split()
659 send_title(page_name, self.split_title(), msg=self.msg, msg_type=self.msg_type)
661 PageFormatter(self.get_raw_body()).print_html()
663 send_guru("Read access denied by ACLs", "notice")
664 send_footer(page_name, self._last_modified())
666 def _last_modified(self):
668 from time import localtime, strftime
669 modtime = localtime(os.stat(self._text_filename())[stat.ST_MTIME])
671 if er.errno != errno.ENOENT:
674 return strftime(datetime_fmt, modtime)
676 def send_editor(self, preview=None):
677 send_title(None, 'Edit ' + self.split_title(), msg=self.msg, msg_type=self.msg_type)
678 if not self.can_write():
679 send_guru("Write access denied by ACLs", "error")
682 print ('<p><b>Editing ' + self.page_name
683 + ' for ' + cgi.escape(remote_user())
684 + ' from ' + cgi.escape(get_hostname(remote_host()))
686 print '<div class="editor"><form method="post" action="%s/%s">' % (script_name(), self.page_name)
687 print '<input type="hidden" name="savepage" value="%s">' % (self.page_name)
688 print """<textarea wrap="off" spellcheck="true" id="editor" name="savetext" rows="17" cols="100">%s</textarea>""" % (preview or self.get_raw_body())
691 <input type="submit" name="save" value="Save" />
692 <input type="submit" name="preview" value="Preview" />
693 <input type="reset" value="Reset" />
694 <input type="submit" name="cancel" value="Cancel" />
697 print "<p>" + Page('EditingTips').link_to() + "</p>"
699 print "<div class='preview'>"
700 PageFormatter(preview).print_html()
702 send_footer(self.page_name)
705 if not self.can_read():
706 send_title(None, msg='Read access denied by ACLs', msg_type='notice')
708 emit_header("text/plain")
709 print self.get_raw_body()
711 def _write_file(self, text):
712 tmp_filename = self._tmp_filename()
713 open(tmp_filename, 'wt').write(text.replace('\r\n', '\n'))
714 text = self._text_filename()
716 # Bad Bill! POSIX rename ought to replace. :-(
720 if er.errno != errno.ENOENT: raise er
721 os.rename(tmp_filename, text)
723 def save_text(self, newtext):
724 if not self.can_write():
725 self.msg = 'Write access denied by ACLs'
726 self.msg_type = 'error'
729 self._write_file(newtext)
732 # FIXME: what's the std way to perform shell quoting in python?
733 cmd = ( post_edit_hook
734 + " '" + data_dir + '/' + self.page_name
735 + "' '" + remote_user()
736 + "' '" + remote_host() + "'"
742 self.msg += "Post-editing hook returned %d.\n" % rc
743 self.msg += 'Command was: ' + cmd + '\n'
745 self.msg += 'Output follows:\n' + msg
747 self.msg = 'Thank you for your contribution. Your attention to detail is appreciated.'
748 self.msg_type = 'success'
750 #TODO: merge into send_raw()
751 def send_verbatim(filename, mime_type='application/octet-stream'):
752 pathname = path.join(data_dir, filename)
753 data = open(pathname, 'rb').read()
754 emit_header(mime_type)
755 sys.stdout.write(data)
757 # Main ---------------------------------------------------------------
759 execfile("geekigeeki.conf.py")
761 form = cgi.FieldStorage()
763 handlers = { 'fullsearch': do_fullsearch,
764 'titlesearch': do_titlesearch,
767 'savepage': do_savepage }
769 for cmd in handlers.keys():
770 if form.has_key(cmd):
771 apply(handlers[cmd], (form[cmd].value,))
774 path_info = environ.get('PATH_INFO', '')
775 if len(path_info) and path_info[0] == '/':
776 query = path_info[1:] or 'FrontPage'
778 query = environ.get('QUERY_STRING', '') or 'FrontPage'
780 if file_re.match(query):
781 if word_re.match(query):
782 Page(query).send_page()
783 elif img_re.match(query):
784 #FIXME: use correct mime type
785 send_verbatim(query, 'image/jpeg')
789 print "Status: 404 Not Found"
790 send_title(None, msg='Can\'t work out query: ' + query)
793 msg=traceback.format_exc()
795 send_guru(msg, "error")
797 send_title(None, msg=msg)