2 """Quick-quick implementation of WikiWikiWeb in Python
5 # Copyright (C) 1999, 2000 Martin Pool <mbp@humbug.org.au>
6 # This version includes additional changes by Gerardo Poggiali (2002)
7 # This version includes additional changes by Bernardo Innocenti (2007)
9 # This program is free software: you can redistribute it and/or modify
10 # it under the terms of the GNU General Public License as published by
11 # the Free Software Foundation, either version 3 of the License, or
12 # (at your option) any later version.
14 # This program is distributed in the hope that it will be useful, but
15 # WITHOUT ANY WARRANTY; without even the implied warranty of
16 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 # General Public License for more details.
19 # You should have received a copy of the GNU General Public License
20 # along with this program. If not, see <http://www.gnu.org/licenses/>.
22 __version__ = '$Revision: 1.63+gerry+bernie $'[11:-2]
24 import cgi, sys, string, os, re, errno, time, stat
25 from os import path, environ
27 # Regular expression defining a WikiWord
28 # (but this definition is also assumed in other places)
29 file_re = re.compile(r"^\b([A-Za-z0-9_\.\-]+)\b$")
30 word_re = re.compile(r"^\b([A-Z][a-z]+){2,}\b$")
31 img_re = re.compile(r"^.*\.(png|gif|jpg|jpeg)$", re.IGNORECASE)
32 url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$")
37 # CGI stuff ---------------------------------------------------------
40 return environ.get('SCRIPT_NAME', '')
42 def privileged_path():
43 return privileged_url or script_name()
46 user = environ.get('REMOTE_USER', '')
47 if user is None or user == '' or user == 'anonymous':
48 user = 'AnonymousCoward'
52 return environ.get('REMOTE_ADDR', '')
54 def get_hostname(addr):
56 from socket import gethostbyaddr
57 return gethostbyaddr(addr)[0] + ' (' + addr + ')'
61 # Formatting stuff --------------------------------------------------
63 def emit_header(type="text/html"):
64 print "Content-type: " + type + "; charset=utf-8"
67 def send_guru(msg, msg_type):
68 if msg is None or msg == '': return
69 print '<pre id="guru" onclick="this.style.display = \'none\'" class="' + msg_type + '">'
70 if msg_type == 'error':
71 print ' Software Failure. Press left mouse button to continue.\n'
73 if msg_type == 'error':
74 print ' Guru Meditation #DEADBEEF.ABADC0DE'
76 # FIXME: This simple JS snippet is harder to pass than ACID 3.0
78 <script language="JavaScript" type="text/javascript">
79 var guru = document.getElementById('guru');
80 // Firefox 2.0 doesn't take border-color, but returns border-top-color fine
81 var color = document.defaultView.getComputedStyle(guru,null).getPropertyValue('border-top-color');
84 guru.style.setProperty('border-color', color, '');
85 setTimeout('guruOff()', 1000);
88 guru.style.setProperty('border-color', '#000000', '');
89 setTimeout('guruOn()', 1000);
91 // Safari 2.0 returns this rgba crap
92 // Konqueror 3.5.6 doesn't seem to support computed properties
93 if (color && color != 'rgba(0, 0, 0, 0)') {
94 //window.alert("enabled! color='" + color + "'");
99 def send_title(name, text="Limbo", msg=None, msg_type='error'):
101 if title_done: return
105 print """<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
106 "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
107 <html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
109 print "<head><title>%s: %s</title>" % (site_name, text)
110 print ' <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8" />'
112 print ' <meta name="robots" content="noindex,nofollow" />'
114 print ' <link rel="stylesheet" type="text/css" href="%s" />' % css_url
118 if name and privileged_url is not None:
119 print '<body ondblclick="location.href=\'' + privileged_path() + '?edit=' + name + '\'">'
124 send_guru(msg, msg_type)
127 print '<div class="nav">'
128 print ' <b>' + site_name + ': ',
130 print link_tag('?fullsearch=' + name, text) + '</b> '
133 print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink')
134 print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
135 print ' | <a href="' + history_url + '" class="navlink">Recent Changes</a>'
138 print ' | <a href="' + history_url + '?a=history;f=' + name + '" class="navlink">Page History</a>'
139 print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
140 if privileged_url is not None:
141 print ' | ' + link_tag('?edit=' + name, 'Edit Page', 'navlink', authentication=True)
143 print ' | <i>Immutable Page</i>'
146 if user != 'AnonymousCoward':
147 print ' | <span class="login"><i>logged in as <b>' + cgi.escape(user) + '</b></i></span>'
151 def link_tag(params, text=None, ss_class=None, authentication=False):
153 text = params # default
156 classattr += 'class="%s" ' % ss_class
157 # Prevent crawlers from following links to generated pages
158 # and links added by potential spammers
159 if ss_class == 'external' or ss_class == 'navlink':
160 classattr += 'rel="nofollow" '
162 path = privileged_path()
165 return '<a %shref="%s/%s">%s</a>' % (classattr, path, params, text)
167 # Search ---------------------------------------------------
169 def do_fullsearch(needle):
170 send_title(None, 'Full text search for "%s"' % (needle))
172 needle_re = re.compile(needle, re.IGNORECASE)
174 all_pages = page_list()
175 for page_name in all_pages:
176 body = Page(page_name).get_raw_body()
177 count = len(needle_re.findall(body))
179 hits.append((count, page_name))
181 # The default comparison for tuples compares elements in order,
182 # so this sorts by number of hits
187 for (count, page_name) in hits:
188 print '<li><p>' + Page(page_name).link_to()
189 print ' . . . . ' + `count`
190 print ['match', 'matches'][count <> 1]
194 print_search_stats(len(hits), len(all_pages))
197 def do_titlesearch(needle):
198 # TODO: check needle is legal -- but probably we can just accept any RE
199 send_title(None, "Title search for \"" + needle + '"')
201 needle_re = re.compile(needle, re.IGNORECASE)
202 all_pages = page_list()
203 hits = filter(needle_re.search, all_pages)
206 for filename in hits:
207 print '<li><p>' + Page(filename).link_to() + "</p></li>"
210 print_search_stats(len(hits), len(all_pages))
213 def print_search_stats(hits, searched):
214 print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
216 def do_edit(pagename):
217 Page(pagename).send_editor()
219 def do_raw(pagename):
220 Page(pagename).send_raw()
222 def do_savepage(pagename):
223 if privileged_url is None:
224 raise 'editing disallowed for ' + 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 cancelled'
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>'
250 return filter(word_re.match, os.listdir(text_dir))
253 def send_footer(name, mod_string=None):
255 cgi.print_arguments()
256 cgi.print_form(cgi.FieldStorage())
258 print '<div id="footer"><hr />'
259 print '<p class="copyright">Powered by <a href="http://www.codewiz.org/wiki/GeekiGeeki">GeekiGeeki</a></p>'
261 print '<p class="modified">last modified %s</p>' % mod_string
262 print '</div></body></html>'
265 # ----------------------------------------------------------
267 def _macro_TitleSearch():
268 return _macro_search("titlesearch")
270 def _macro_FullSearch():
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="Go" /></form>""" % (type, default)
280 def _macro_WordIndex():
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():
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):
338 self.is_em = self.is_b = 0
340 self.list_indents = []
342 self.in_table = False
345 self.in_header = True
347 def _emph_repl(self, word):
349 self.is_b = not self.is_b
350 return ['</strong>', '<strong>'][self.is_b]
352 self.is_em = not self.is_em
353 return ['</em>', '<em>'][self.is_em]
355 def _tit_repl(self, word):
357 result = "</h%d>" % self.h_level
360 self.h_level = len(word) - 1
361 result = "<h%d>" % self.h_level
364 def _rule_repl(self, word):
367 s = s + "\n<hr size='1' noshade=\"noshade\" />\n"
369 s = s + "\n<hr size='%d' noshade=\"noshade\" />\n" % (len(word) - 2 )
372 def _word_repl(self, word):
373 return Page(word).link_to()
375 def _img_repl(self, word):
376 return '<img border="0" src="%s/%s" />' % (script_name(), word)
378 def _url_repl(self, word):
379 if img_re.match(word):
380 return '<img border="0" src="%s" />' % word
382 return '<a href="%s" rel="nofollow" class="external">%s</a>' % (word, word)
384 def _hurl_repl(self, word):
385 m = re.compile("\[\[(\S+)\ (.+)\]\]").match(word)
388 if img_re.match(anchor):
389 return '<img border="0" src="%s" alt="%s" />' % (anchor, descr)
390 elif url_re.match(anchor):
391 return '<a href="%s" rel="nofollow" class="external">%s</a>' % (anchor, descr)
392 elif anchor.startswith('/'):
393 return '<a href="%s">%s</a>' % (anchor, descr)
395 return link_tag(anchor, descr)
397 def _email_repl(self, word):
398 return '<a href="mailto:%s">%s</a>' % (word, word)
401 def _ent_repl(self, s):
402 return {'&': '&',
407 def _li_repl(self, match):
411 def _pre_repl(self, word):
412 if word == '{{{' and not self.in_pre:
420 def _hilight_repl(self, word):
421 return '<strong class="hilight">' + word + '</strong>'
423 def _var_repl(self, word):
424 if word == '{{' and not self.in_var:
432 def _tr_repl(self, word):
434 if not self.in_table:
437 out = '</p><table><tbody>\n'
439 return out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '"><td>'
441 def _tre_repl(self, word):
446 def _td_repl(self, word):
451 def _macro_repl(self, word):
452 macro_name = word[2:-2]
453 # TODO: Somehow get the default value into the search field
454 return apply(globals()['_macro_' + macro_name], ())
456 def _indent_level(self):
457 return len(self.list_indents) and self.list_indents[-1]
459 def _indent_to(self, new_level):
460 if self._indent_level() == new_level:
463 while self._indent_level() > new_level:
464 del(self.list_indents[-1])
466 while self._indent_level() < new_level:
467 self.list_indents.append(new_level)
474 res += '</ul>' * len(self.list_indents)
476 self.list_indents = []
479 def replace(self, match):
480 for type, hit in match.groupdict().items():
482 return apply(getattr(self, '_' + type + '_repl'), (hit,))
484 raise "Can't handle match " + `match`
486 def print_html(self):
487 print "<div class='wiki'><p>"
489 # For each line, we scan through looking for magic
490 # strings, outputting verbatim any intervening text
491 # TODO: highlight search words (look at referer)
492 scan_re = re.compile(
495 + r"(?P<emph>'{2,3})"
496 + r"|(?P<tit>\={2,6})"
497 + r"|(?P<rule>^-{3,})"
499 + r"|(?P<hilight>\b(FIXME|TODO)\b)"
502 + r"|(?P<img>\b[a-zA-Z0-9_-]+\.(png|gif|jpg|jpeg|bmp))"
503 + r"|(?P<word>\b(?:[A-Z][a-z]+){2,}\b)"
504 + r"|(?P<hurl>\[\[\S+\s+.+\]\])"
505 + r"|(?P<url>(http|ftp|nntp|news|mailto)\:[^\s'\"]+\S)"
506 + r"|(?P<email>[-\w._+]+\@[\w.-]+)"
510 + r"|(?P<pre>(\{\{\{|\s*\}\}\}))"
511 + r"|(?P<var>(\{\{|\}\}))"
514 + r"|(?P<tr>^\s*\|\|\s*)"
515 + r"|(?P<tre>\s*\|\|\s*$)"
516 + r"|(?P<td>\s*\|\|\s*)"
519 + r"|(?P<macro>\[\[(TitleSearch|FullSearch|WordIndex|TitleIndex)\]\])"
523 + r"(?P<pre>\s*\}\}\})"
525 blank_re = re.compile(r"^\s*$")
526 indent_re = re.compile(r"^\s*")
527 tr_re = re.compile(r"^\s*\|\|")
528 eol_re = re.compile(r"\r?\n")
529 raw = string.expandtabs(self.raw)
530 for line in eol_re.split(raw):
533 if line.startswith('#'):
535 self.in_header = False
538 print re.sub(pre_re, self.replace, line)
540 if self.in_table and not tr_re.match(line):
541 self.in_table = False
542 print '</tbody></table><p>'
544 if blank_re.match(line):
547 indent = indent_re.match(line)
548 print self._indent_to(len(indent.group(0)))
549 print re.sub(scan_re, self.replace, line)
551 if self.in_pre: print '</pre>'
552 if self.in_table: print '</tbody></table><p>'
556 # ----------------------------------------------------------
558 def __init__(self, page_name):
559 self.page_name = page_name
561 self.msg_type = 'error'
564 def split_title(self):
565 # look for the end of words and the start of a new word,
566 # and insert a space there
567 return re.sub('([a-z])([A-Z])', r'\1 \2', self.page_name)
569 def _text_filename(self):
570 return path.join(text_dir, self.page_name)
572 def _tmp_filename(self):
573 return path.join(text_dir, ('#' + self.page_name + '.' + `os.getpid()` + '#'))
577 os.stat(self._text_filename())
580 if er.errno == errno.ENOENT:
586 word = self.page_name
588 return link_tag(word, word, 'wikilink')
591 return link_tag(word, '?', 'nonexistent') + word
593 return link_tag(word, word, 'nonexistent')
596 def get_raw_body(self):
598 return open(self._text_filename(), 'rt').read()
600 if er.errno == errno.ENOENT:
601 # just doesn't exist, use default
602 return 'Describe %s here.' % self.page_name
610 file = open(self._text_filename(), 'rt')
611 attr_re = re.compile(r"^#(\S*)(.*)$")
613 m = attr_re.match(line)
616 self.attrs[m.group(1)] = m.group(2).strip()
617 #print "bernie: attrs[" + m.group(1) + "] = " + m.group(2) + "<br>\n"
619 if er.errno != errno.ENOENT:
624 attrs = self.get_attrs()
626 # SomeUser:read,write All:read
628 for rule in acl.split():
629 (user,perms) = acl.split(':')
630 if user == remote_user() or user == "All":
631 if 'write' in perms.split(','):
641 page_name = self.page_name
642 send_title(page_name, self.split_title(), msg=self.msg, msg_type=self.msg_type)
643 PageFormatter(self.get_raw_body()).print_html()
644 send_footer(page_name, self._last_modified())
646 def _last_modified(self):
647 if not self.exists():
649 from time import localtime, strftime
650 modtime = localtime(os.stat(self._text_filename())[stat.ST_MTIME])
651 return strftime(datetime_fmt, modtime)
653 def send_editor(self, preview=None):
654 send_title(None, 'Edit ' + self.split_title(), msg=self.msg, msg_type=self.msg_type)
656 print ('<p><b>Editing ' + self.page_name
657 + ' for ' + cgi.escape(remote_user())
658 + ' from ' + cgi.escape(get_hostname(remote_host()))
660 print '<div class="editor"><form method="post" action="%s/%s">' % (script_name(), self.page_name)
661 print '<input type="hidden" name="savepage" value="%s">' % (self.page_name)
662 print """<textarea wrap="virtual" id="editor" name="savetext" rows="17" cols="80">%s</textarea>""" % (preview or self.get_raw_body())
665 <input type="submit" name="save" value="Save" />
666 <input type="submit" name="preview" value="Preview" />
667 <input type="reset" value="Reset" />
668 <input type="submit" name="cancel" value="Cancel" />
671 print "<p>" + Page('EditingTips').link_to() + "</p>"
673 print "<div class='preview'>"
674 PageFormatter(preview).print_html()
676 send_footer(self.page_name)
679 emit_header("text/plain")
680 print self.get_raw_body()
682 def _write_file(self, text):
683 tmp_filename = self._tmp_filename()
684 open(tmp_filename, 'wt').write(text.replace('\r\n', '\n'))
685 text = self._text_filename()
687 # Bad Bill! POSIX rename ought to replace. :-(
691 if er.errno <> errno.ENOENT: raise er
692 os.rename(tmp_filename, text)
694 def save_text(self, newtext):
695 self._write_file(newtext)
698 # FIXME: what's the std way to perform shell quoting in python?
699 cmd = ( post_edit_hook
700 + " '" + text_dir + '/' + self.page_name
701 + "' '" + remote_user()
702 + "' '" + remote_host() + "'"
708 self.msg += "Post-editing hook returned %d.\n" % rc
709 self.msg += 'Command was: ' + cmd + '\n'
711 self.msg += 'Output follows:\n' + msg
713 self.msg = 'Thankyou for your contribution. Your attention to detail is appreciated.'
714 self.msg_type = 'success'
716 def send_verbatim(filename, mime_type='application/octet-stream'):
717 pathname = path.join(text_dir, filename)
718 data = open(pathname, 'rb').read()
719 emit_header(mime_type)
720 sys.stdout.write(data)
722 # Main ---------------------------------------------------------------
724 # Configuration values
725 site_name = 'Codewiz'
727 # set to None for read-only sites
728 # leave empty ('') to allow anonymous edits
729 # otherwise, set to a URL that requires authentication
730 privileged_url = 'https://www.codewiz.org/~bernie/wiki'
732 data_dir = '/home/bernie/public_html/wiki'
733 text_dir = path.join(data_dir, 'text')
734 css_url = '../wikidata/geekigeeki.css' # optional stylesheet link
735 history_url = '../wikigit/wiki.git'
736 post_edit_hook = './post_edit_hook.sh'
737 datetime_fmt = '%a %d %b %Y %I:%M %p'
738 allow_edit = True # Is it possible to edit pages?
739 show_hosts = True # show hostnames?
740 nonexist_qm = False # show '?' for nonexistent?
741 debug_cgi = False # Set to True for CGI var dump
743 form = cgi.FieldStorage()
745 handlers = { 'fullsearch': do_fullsearch,
746 'titlesearch': do_titlesearch,
749 'savepage': do_savepage }
751 for cmd in handlers.keys():
752 if form.has_key(cmd):
753 apply(handlers[cmd], (form[cmd].value,))
756 path_info = environ.get('PATH_INFO', '')
757 if len(path_info) and path_info[0] == '/':
758 query = path_info[1:] or 'FrontPage'
760 query = environ.get('QUERY_STRING', '') or 'FrontPage'
762 if file_re.match(query):
763 if word_re.match(query):
764 Page(query).send_page()
765 elif img_re.match(query):
766 #FIXME: use correct mime type
767 send_verbatim(query, 'image/jpeg')
772 send_title(None, msg='Can\'t work out query: ' + query)
775 msg=traceback.format_exc()
777 send_guru(msg, "error")
779 send_title(None, msg=msg)