2 # -*- coding: utf-8 -*-
4 # Copyright 1999, 2000 Martin Pool <mbp@humbug.org.au>
5 # Copyright 2002 Gerardo Poggiali
6 # Copyright 2007, 2008 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$")
32 word_re = re.compile(r"^\b((([A-Z][a-z]+){2,}/)*([A-Z][a-z]+){2,})\b$")
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 def relative_url(path, privileged=False):
64 if not (url_re.match(path) or path.startswith('/')):
66 url = privileged_path()
69 path = url + '/' + path
72 # Formatting stuff --------------------------------------------------
74 def emit_header(type="text/html"):
75 print "Content-type: " + type + "; charset=utf-8"
78 def send_guru(msg, msg_type):
79 if msg is None or msg == '': return
80 print '<pre id="guru" onclick="this.style.display = \'none\'" class="' + msg_type + '">'
81 if msg_type == 'error':
82 print ' Software Failure. Press left mouse button to continue.\n'
84 if msg_type == 'error':
85 print ' Guru Meditation #DEADBEEF.ABADC0DE'
87 # FIXME: This little JS snippet is harder to pass than ACID 3.0
89 <script language="JavaScript" type="text/javascript">
90 var guru = document.getElementById('guru');
91 // Firefox 2.0 doesn't take border-color, but returns border-top-color fine
92 var color = document.defaultView.getComputedStyle(guru,null).getPropertyValue('border-top-color');
95 guru.style.setProperty('border-color', color, '');
96 setTimeout('guruOff()', 1000);
99 guru.style.setProperty('border-color', '#000000', '');
100 setTimeout('guruOn()', 1000);
102 // Safari 2.0 returns this rgba crap
103 // Konqueror 3.5.6 doesn't seem to support computed properties
104 if (color && color != 'rgba(0, 0, 0, 0)') {
105 //window.alert("enabled! color='" + color + "'");
110 def send_title(name, text="Limbo", msg=None, msg_type='error'):
112 if title_done: return
116 print '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'
117 print ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">'
118 print '<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">'
120 site_name = globals().get('site_name', 'Unconfigured Site')
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 print ' <link rel="stylesheet" type="text/css" href="%s" />' % relative_url(css)
130 if name and privileged_url is not None:
131 print '<body ondblclick="location.href=\'' + privileged_path() + '?edit=' + name + '\'">'
136 send_guru(msg, msg_type)
139 print '<div class="nav">'
140 print ' <b>' + site_name + ': ',
142 print link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
145 print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink')
146 print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
147 if globals().has_key('history_url'):
148 print ' | <a href="' + history_url + '" class="navlink">Recent Changes</a>'
150 print ' | <a href="' + history_url + '?a=history;f=' + name + '" class="navlink">Page History</a>'
153 print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
154 if privileged_url is not None:
155 print ' | ' + link_tag('?edit=' + name, 'Edit Page', 'navlink', privileged=True)
157 print ' | <i>Immutable Page</i>'
160 if user != 'AnonymousCoward':
161 print ' | <span class="login"><i>logged in as <b>' + cgi.escape(user) + '</b></i></span>'
165 def link_tag(params, text=None, ss_class=None, privileged=False):
167 text = params # default
170 classattr += 'class="%s" ' % ss_class
171 # Prevent crawlers from following links potentially added by spammers or to generated pages
172 if ss_class == 'external' or ss_class == 'navlink':
173 classattr += 'rel="nofollow" '
174 elif url_re.match(params):
175 classattr += 'rel="nofollow" '
176 return '<a %shref="%s">%s</a>' % (classattr, relative_url(params, privileged=privileged), text)
178 # Search ---------------------------------------------------
180 def do_fullsearch(needle):
181 send_title(None, 'Full text search for "%s"' % (needle))
183 needle_re = re.compile(needle, re.IGNORECASE)
185 all_pages = page_list()
186 for page_name in all_pages:
187 body = Page(page_name).get_raw_body()
188 count = len(needle_re.findall(body))
190 hits.append((count, page_name))
192 # The default comparison for tuples compares elements in order,
193 # so this sorts by number of hits
198 for (count, page_name) in hits:
199 print '<li><p>' + Page(page_name).link_to()
200 print ' . . . . ' + `count`
201 print ['match', 'matches'][count != 1]
205 print_search_stats(len(hits), len(all_pages))
207 def do_titlesearch(needle):
208 # TODO: check needle is legal -- but probably we can just accept any RE
209 send_title(None, "Title search for \"" + needle + '"')
211 needle_re = re.compile(needle, re.IGNORECASE)
212 all_pages = page_list()
213 hits = filter(needle_re.search, all_pages)
216 for filename in hits:
217 print '<li><p>' + Page(filename).link_to() + "</p></li>"
220 print_search_stats(len(hits), len(all_pages))
222 def print_search_stats(hits, searched):
223 print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
225 def do_raw(pagename):
226 Page(pagename).send_raw()
228 def do_edit(pagename):
231 if 'preview' in form:
232 pg.send_editor(form['savetext'].value)
234 pg.save_text(form['savetext'].value)
236 elif 'cancel' in form:
237 pg.msg = 'Editing canceled'
238 pg.msg_type = 'notice'
243 def make_index_key():
245 links = map(lambda ch: '<a href="#%s">%s</a>' % (ch, ch),
247 s = s + string.join(links, ' | ')
248 s = s + '</center></p>'
252 return filter(word_re.match, os.listdir(data_dir))
254 def send_footer(name, mod_string=None):
255 if globals().get('debug_cgi', False):
256 cgi.print_arguments()
257 cgi.print_form(cgi.FieldStorage())
260 print '<div id="footer"><hr />'
261 print ('<p class="copyright"><span class="benchmark">generated in %0.3fs</span>' +
262 ' by <a href="http://www.codewiz.org/wiki/GeekiGeeki">GeekiGeeki</a>' +
263 ' version %s</p>') % (clock() - start_time, __version__)
265 print '<p class="modified">last modified %s</p>' % mod_string
266 print '</div></body></html>'
268 # ----------------------------------------------------------
270 def _macro_TitleSearch(*vargs):
271 return _macro_search("titlesearch")
273 def _macro_FullSearch(*vargs):
274 return _macro_search("fullsearch")
276 def _macro_search(type):
277 if form.has_key('value'):
278 default = form["value"].value
281 return """<form method="get"><input name="%s" size="30" value="%s" /><input type="submit" value="Search" /></form>""" % (type, default)
283 def _macro_WordIndex(*vargs):
285 pages = list(page_list())
287 word_re = re.compile('[A-Z][a-z]+')
289 for word in word_re.findall(name):
291 map[word].append(name)
295 all_words = map.keys()
299 for word in all_words:
300 letter = string.lower(word[0])
301 if letter != last_letter:
302 s = s + '; <a name="%s"><h3>%s</h3></a>' % (letter, letter)
305 s = s + '<b>%s</b><ul>' % word
310 if name == last_page: continue
311 s = s + '<li>' + Page(name).link_to()
316 def _macro_TitleIndex(*vargs):
318 pages = list(page_list())
320 current_letter = None
322 letter = string.lower(name[0])
323 if letter != current_letter:
324 s += '<a name="%s"><h3>%s</h3></a>' % (letter, letter)
325 current_letter = letter
328 s += Page(name).link_to()
332 # ----------------------------------------------------------
334 """Object that turns Wiki markup into HTML.
336 All formatting commands can be parsed one line at a time, though
337 some state is carried over between lines.
339 def __init__(self, raw):
342 self.in_pre = self.in_table = False
343 self.in_header = True
344 self.list_indents = []
345 self.tr_cnt = self.h_cnt = 0
355 "^^": ["sup", False],
359 def _b_repl(self, word):
360 style = self.styles[word]
361 style[1] = not style[1]
362 return ['</','<'][style[1]] + style[0] + '>'
364 def _tit_repl(self, word):
366 result = '</h%d>' % self.h_level
369 self.h_level = len(word) - 1
371 #abridged = re.sub('[^a-z_]', '', word.lower().replace(' ', '_'))
372 result = '<h%d id="%d"><a class="heading" href="#%d">¶</a> ' % (self.h_level, self.h_cnt, self.h_cnt)
375 def _br_repl(self, word):
378 def _rule_repl(self, word):
379 return self._undent() + '\n<hr size="%d" noshade="noshade" />\n' % (len(word) - 2)
381 def _word_repl(self, word):
382 return Page(word).link_to()
384 def _img_repl(self, word):
385 path = relative_url(word)
386 return '<a href="%s"><img border="0" src="%s" /></a>' % (path, path)
388 def _url_repl(self, word):
389 if img_re.match(word):
390 return '<a href="%s"><img border="0" src="%s" /></a>' % (word, word)
392 return '<a href="%s" rel="nofollow" class="external">%s</a>' % (word, word)
394 def _hurl_repl(self, word):
395 m = re.compile("\[\[([^ \t\n\r\f\v\|]+)(?:\s*\|\s*([^\]]+)|)\]\]").match(word)
397 descr = m.group(2) or name
399 macro = globals().get('_macro_' + name)
401 return apply(macro, (name, descr))
402 elif img_re.match(name):
403 name = relative_url(name)
404 # The "extthumb" nonsense works around a limitation of the HTML block model
405 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>' % (name, name, descr, descr)
407 if img_re.match(descr):
408 descr = '<img border="0" src="' + descr + '" />'
410 return link_tag(name, descr, 'wikilink')
412 def _email_repl(self, word):
413 return '<a href="mailto:%s">%s</a>' % (word, word)
415 def _html_repl(self, word):
416 return word; # Pass through
418 def _ent_repl(self, s):
419 return {'&': '&',
423 def _li_repl(self, match):
426 def _pre_repl(self, word):
427 if word == '{{{' and not self.in_pre:
435 def _hi_repl(self, word):
436 return '<strong class="highlight ' + word + '">' + word + '</strong>'
438 def _tr_repl(self, word):
440 if not self.in_table:
443 out = '</p><table><tbody>\n'
445 out = out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '">'
446 return out + ['<td>', '<th>'][word.strip() == '||=']
448 def _td_repl(self, word):
450 return ['</td><td>', '</th><th>'][word.strip() == '||=']
453 def _tre_repl(self, word):
455 return ['</td></tr>', '</th></tr>'][word.strip() == '||=']
458 def _indent_level(self):
459 return len(self.list_indents) and self.list_indents[-1]
461 def _indent_to(self, new_level):
462 if self._indent_level() == new_level:
465 while self._indent_level() > new_level:
466 del(self.list_indents[-1])
468 while self._indent_level() < new_level:
469 self.list_indents.append(new_level)
476 res += '</ul>' * len(self.list_indents)
478 self.list_indents = []
481 def replace(self, match):
482 for type, hit in match.groupdict().items():
484 return apply(getattr(self, '_' + type + '_repl'), (hit,))
486 raise "Can't handle match " + `match`
488 def print_html(self):
489 print '<div class="wiki"><p>'
491 # For each line, we scan through looking for magic
492 # strings, outputting verbatim any intervening text
493 # TODO: highlight search words (look at referrer)
494 scan_re = re.compile(
497 + r"(?P<b>\*\*|'''|//|''|##|``|__|\^\^|,,)"
498 + r"|(?P<tit>\={2,6})"
500 + r"|(?P<rule>^-{3,})"
501 + r"|(?P<html><(/|)(div|span|iframe)[^<>]*>)"
503 + r"|(?P<hi>\b(FIXME|TODO|DONE)\b)"
506 + r"|(?P<img>\b[a-zA-Z0-9_-]+\.(png|gif|jpg|jpeg|bmp))"
507 + r"|(?P<word>\b(?:[A-Z][a-z]+){2,}\b)"
508 + r"|(?P<hurl>\[\[([^ \t\n\r\f\v\|]+)(?:\s*\|\s*([^\]]+)|)\]\])"
509 + r"|(?P<url>(http|https|ftp|mailto)\:[^\s'\"]+\S)"
510 + r"|(?P<email>[-\w._+]+\@[\w.-]+)"
513 + r"|(?P<li>^\s+[\*#] +)"
514 + r"|(?P<pre>\{\{\{|\s*\}\}\})"
517 + r"|(?P<tr>^\s*\|\|(=|)\s*)"
518 + r"|(?P<tre>\s*\|\|(=|)\s*$)"
519 + r"|(?P<td>\s*\|\|(=|)\s*)"
523 + r"(?P<pre>\s*\}\}\})"
526 blank_re = re.compile(r"^\s*$")
527 indent_re = re.compile(r"^\s*")
528 tr_re = re.compile(r"^\s*\|\|")
529 eol_re = re.compile(r"\r?\n")
530 raw = string.expandtabs(self.raw)
531 for line in eol_re.split(raw):
534 if line.startswith('#'):
536 self.in_header = False
539 print re.sub(pre_re, self.replace, line)
541 if self.in_table and not tr_re.match(line):
542 self.in_table = False
543 print '</tbody></table><p>'
545 if blank_re.match(line):
548 indent = indent_re.match(line)
549 print self._indent_to(len(indent.group(0)))
550 print re.sub(scan_re, self.replace, line)
552 if self.in_pre: print '</pre>'
553 if self.in_table: print '</tbody></table><p>'
557 # ----------------------------------------------------------
559 def __init__(self, page_name):
560 self.page_name = page_name
562 self.msg_type = 'error'
564 def split_title(self):
565 # look for the end of words and the start of a new word and insert a space there
566 return re.sub('([a-z])([A-Z])', r'\1 \2', self.page_name)
568 def _text_filename(self):
569 return path.join(data_dir, self.page_name)
571 def _tmp_filename(self):
572 return path.join(data_dir, ('#' + self.page_name.replace('/','_') + '.' + `os.getpid()` + '#'))
576 os.stat(self._text_filename())
579 if er.errno == errno.ENOENT:
584 word = self.page_name
586 return link_tag(word, word, 'wikilink')
588 return link_tag(word, nonexist_pfx + word, 'nonexistent')
590 def get_raw_body(self):
592 return open(self._text_filename(), 'rb').read()
594 if er.errno == errno.ENOENT:
595 return '' # just doesn't exist, use default
596 if er.errno == errno.EISDIR:
601 if self.__dict__.has_key('attrs'):
605 file = open(self._text_filename(), 'rt')
606 attr_re = re.compile(r"^#(\S*)(.*)$")
608 m = attr_re.match(line)
611 self.attrs[m.group(1)] = m.group(2).strip()
612 #print "bernie: attrs[" + m.group(1) + "] = " + m.group(2) + "<br>\n"
614 if er.errno != errno.ENOENT:
618 def get_attr(self, name, default):
619 return self.get_attrs().get(name, default)
621 def can(self, action, default=True):
623 #acl SomeUser:read,write All:read
624 acl = self.get_attr("acl", None)
625 for rule in acl.split():
626 (user,perms) = rule.split(':')
627 if user == remote_user() or user == "All":
628 return action in perms.split(',')
630 except Exception, er:
635 return self.can("write", True)
638 return self.can("read", True)
643 page_name = self.page_name
647 css_url = css_url + self.get_attr("css", "").split()
649 send_title(page_name, self.split_title(), msg=self.msg, msg_type=self.msg_type)
651 PageFormatter(self.get_raw_body()).print_html()
653 send_guru("Read access denied by ACLs", "notice")
654 send_footer(page_name, self._last_modified())
656 def _last_modified(self):
658 from time import localtime, strftime
659 modtime = localtime(os.stat(self._text_filename())[stat.ST_MTIME])
661 if er.errno != errno.ENOENT:
664 return strftime(datetime_fmt, modtime)
666 def send_editor(self, preview=None):
667 send_title(None, 'Edit ' + self.split_title(), msg=self.msg, msg_type=self.msg_type)
668 if not self.can_write():
669 send_guru("Write access denied by ACLs", "error")
672 print ('<p><b>Editing ' + self.page_name
673 + ' for ' + cgi.escape(remote_user())
674 + ' from ' + cgi.escape(get_hostname(remote_host()))
676 print '<div class="editor"><form method="post" action="%s">' % relative_url(self.page_name)
677 print '<input type="hidden" name="edit" value="%s">' % (self.page_name)
678 print """<textarea wrap="off" spellcheck="true" id="editor" name="savetext" rows="17" cols="100">%s</textarea>""" % (preview or self.get_raw_body())
681 <input type="submit" name="save" value="Save" />
682 <input type="submit" name="preview" value="Preview" />
683 <input type="reset" value="Reset" />
684 <input type="submit" name="cancel" value="Cancel" />
687 print "<p>" + Page('EditingTips').link_to() + "</p>"
689 print "<div class='preview'>"
690 PageFormatter(preview).print_html()
692 send_footer(self.page_name)
694 def send_raw(self, mimetype='text/plain'):
696 body = self.get_raw_body()
697 emit_header(mimetype)
700 send_title(None, msg='Read access denied by ACLs', msg_type='notice')
702 def _write_file(self, text):
703 tmp_filename = self._tmp_filename()
704 open(tmp_filename, 'wt').write(text.replace('\r\n', '\n'))
705 text = self._text_filename()
707 # Bad Bill! POSIX rename ought to replace. :-(
711 if er.errno != errno.ENOENT: raise er
712 os.rename(tmp_filename, text)
714 def save_text(self, newtext):
715 if not self.can_write():
716 self.msg = 'Write access denied by ACLs'
717 self.msg_type = 'error'
720 self._write_file(newtext)
723 # FIXME: what's the std way to perform shell quoting in python?
724 cmd = ( post_edit_hook
725 + " '" + data_dir + '/' + self.page_name
726 + "' '" + remote_user()
727 + "' '" + remote_host() + "'"
733 self.msg += "Post-editing hook returned %d.\n" % rc
734 self.msg += 'Command was: ' + cmd + '\n'
736 self.msg += 'Output follows:\n' + msg
738 self.msg = 'Thank you for your contribution. Your attention to detail is appreciated.'
739 self.msg_type = 'success'
741 # Main ---------------------------------------------------------------
743 execfile("geekigeeki.conf.py")
744 form = cgi.FieldStorage()
746 'fullsearch': do_fullsearch,
747 'titlesearch': do_titlesearch,
752 for cmd in handlers.keys():
753 if form.has_key(cmd):
754 apply(handlers[cmd], (form[cmd].value,))
757 path_info = environ.get('PATH_INFO', '')
758 if len(path_info) and path_info[0] == '/':
759 query = path_info[1:] or 'FrontPage'
761 query = environ.get('QUERY_STRING', '') or 'FrontPage'
763 if file_re.match(query):
764 if word_re.match(query):
765 Page(query).send_page()
766 elif img_re.match(query):
767 #FIXME: use correct mime type
768 Page(query).send_raw(mimetype='image/jpeg')
770 Page(query).send_raw(mimetype='application/octet-stream')
772 print "Status: 404 Not Found"
773 send_title(None, msg='Can\'t work out query: ' + query)
776 msg=traceback.format_exc()
778 send_guru(msg, "error")
780 send_title(None, msg=msg)