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("(?:\[\[|{{)([^\s\|]+)(?:\s*\|\s*([^\]]+)|)(?:\]\]|}})")
40 # CGI stuff ---------------------------------------------------------
42 return os.environ.get('SCRIPT_NAME', '')
45 path_info = os.environ.get('PATH_INFO', '')
46 if len(path_info) and path_info[0] == '/':
47 return path_info[1:] or 'FrontPage'
49 return os.environ.get('QUERY_STRING', '') or 'FrontPage'
51 def privileged_path():
52 return privileged_url or script_name()
55 user = os.environ.get('REMOTE_USER', '')
56 if user is None or user == '' or user == 'anonymous':
57 user = 'AnonymousCoward'
61 return os.environ.get('REMOTE_ADDR', '')
63 def get_hostname(addr):
65 from socket import gethostbyaddr
66 return gethostbyaddr(addr)[0] + ' (' + addr + ')'
70 def is_external_url(pathname):
71 return (url_re.match(pathname) or pathname.startswith('/'))
73 def relative_url(pathname, privileged=False):
74 if not is_external_url(pathname):
76 url = privileged_path()
79 pathname = url + '/' + pathname
83 return re.sub(' ', '-', re.sub('[^a-z0-9_ ]', '', s.lower()).strip())
85 # Formatting stuff --------------------------------------------------
86 def emit_header(mime_type="text/html"):
87 print "Content-type: " + mime_type + "; charset=utf-8\n"
89 def sendfile(dest_file, src_file):
90 """Efficiently copy file data between file descriptors"""
92 data = src_file.read(65536)
96 def send_guru(msg_text, msg_type):
97 if not msg_text: return
98 print '<pre id="guru" onclick="this.style.display = \'none\'" class="' + msg_type + '">'
99 if msg_type == 'error':
100 print ' Software Failure. Press left mouse button to continue.\n'
102 if msg_type == 'error':
103 print '\n Guru Meditation #DEADBEEF.ABADC0DE'
106 sendfile(sys.stdout, open('gurumeditation.js', 'rb'))
110 def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=False):
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 print "<head><title>%s: %s</title>" % (site_name, text)
121 print ' <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8" />'
123 print ' <meta name="robots" content="noindex,nofollow" />'
125 for meta in meta_urls:
126 http_equiv, content = meta
127 print ' <meta http-equiv="%s" content="%s" />' % (http_equiv, relative_url(content))
129 for link in link_urls:
131 print ' <link rel="%s" href="%s" />' % (rel, relative_url(href))
133 if name and writable and privileged_url is not None:
134 print ' <link rel="alternate" type="application/x-wiki" title="Edit this page" href="%s" />' \
135 % (privileged_path() + '?edit=' + name)
137 if history_url is not None:
138 print ' <link rel="alternate" type="application/rss+xml" title="RSS" href="%s" />' \
139 % relative_url(history_url + '?a=rss')
144 if name and writable and privileged_url is not None:
145 print '<body ondblclick="location.href=\'' + privileged_path() + '?edit=' + name + '\'">'
150 send_guru(msg_text, msg_type)
153 print '<div class="nav">'
154 print link_tag('FrontPage', relative_url(site_icon or 'Home'), 'navlink')
156 print ' <b>' + link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
158 print ' <b>' + text + '</b> '
159 print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
160 if 'history_url' in globals():
161 print ' | <a href="' + relative_url(history_url) + '" class="navlink">Recent Changes</a>'
163 print ' | <a href="' + relative_url(history_url + '?a=history;f=' + name) + '" class="navlink">Page History</a>'
166 print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
167 if privileged_url is not None:
169 print ' | ' + link_tag('?edit=' + name, 'Edit', 'navlink', privileged=True)
171 print ' | ' + link_tag(name, 'Login', 'navlink', privileged=True)
174 print ' | <i>Immutable Page</i>'
177 if user != 'AnonymousCoward':
178 print ' | <span class="login"><i><b>' + link_tag('User/' + user, user) + '</b></i></span>'
182 def send_httperror(status="403 Not Found", query=""):
183 print "Status: %s" % status
184 send_title(None, msg_text=("%s: on query '%s'" % (status, query)))
187 def link_tag(params, text=None, link_class=None, privileged=False):
189 text = params # default
190 elif img_re.match(text):
191 text = '<img border="0" src="' + text + '" />'
194 if is_external_url(params):
195 link_class = 'external'
196 elif file_re.match(params) and Page(params).exists():
197 link_class = 'wikilink'
199 params = nonexist_pfx + params
200 link_class = 'nonexistent'
202 classattr = 'class="%s" ' % link_class
203 # Prevent crawlers from following links potentially added by spammers or to generated pages
204 if link_class == 'external' or link_class == 'navlink':
205 classattr += 'rel="nofollow"'
207 return '<a %shref="%s">%s</a>' % (classattr, relative_url(params, privileged=privileged), text)
209 def link_inline(name, descr=None, args=''):
210 if not descr: descr = name
211 if video_re.match(name):
212 return '<video src="%s">Your browser does not support the HTML5 video tag</video>' % name
214 return '<a href="%s"><img border="0" src="%s" alt="%s" /></a>' % (name, name + args, descr)
216 # Search ---------------------------------------------------
218 def handle_fullsearch(needle):
219 send_title(None, 'Full text search for "%s"' % (needle))
221 needle_re = re.compile(needle, re.IGNORECASE)
223 all_pages = page_list()
224 for page_name in all_pages:
225 body = Page(page_name).get_raw_body()
226 count = len(needle_re.findall(body))
228 hits.append((count, page_name))
230 # The default comparison for tuples compares elements in order,
231 # so this sorts by number of hits
236 for (count, page_name) in hits:
237 print '<li><p>' + link_tag(page_name)
238 print ' . . . . ' + `count`
239 print ['match', 'matches'][count != 1]
243 print_search_stats(len(hits), len(all_pages))
245 def handle_titlesearch(needle):
246 # TODO: check needle is legal -- but probably we can just accept any RE
247 send_title(None, "Title search for \"" + needle + '"')
249 needle_re = re.compile(needle, re.IGNORECASE)
250 all_pages = page_list()
251 hits = filter(needle_re.search, all_pages)
254 for filename in hits:
255 print '<li><p>' + link_tag(filename) + "</p></li>"
258 print_search_stats(len(hits), len(all_pages))
260 def print_search_stats(hits, searched):
261 print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
263 def handle_raw(pagename):
264 if not file_re.match(pagename):
265 send_httperror("403 Forbidden", pagename)
268 Page(pagename).send_raw()
270 def handle_edit(pagename):
271 if not file_re.match(pagename):
272 send_httperror("403 Forbidden", pagename)
277 if form['file'].value:
278 pg.save(form['file'].file.read(), form['changelog'].value)
280 pg.save(form['savetext'].value.replace('\r\n', '\n'), form['changelog'].value)
282 elif 'cancel' in form:
283 pg.msg_text = 'Editing canceled'
284 pg.msg_type = 'notice'
286 else: # preview or edit
288 if 'preview' in form:
289 text = form['savetext'].value
292 def make_index_key():
293 links = map(lambda ch: '<a href="#%s">%s</a>' % (ch, ch), 'abcdefghijklmnopqrstuvwxyz')
294 return '<p><center>'+ ' | '.join(links) + '</center></p>'
296 def page_list(dirname = None, re = word_re):
297 return sorted(filter(re.match, os.listdir(dirname or data_dir)))
299 def send_footer(mod_string=None):
300 if globals().get('debug_cgi', False):
301 cgi.print_arguments()
305 <div id="footer"><hr />
306 <p class="copyright">
307 <a rel="license" href="http://creativecommons.org/licenses/by-sa/3.0/"><img class="license" alt="Creative Commons License" src="%s" /></a>
308 <span class="benchmark">generated in %0.3fs</span> by <a href="http://www.codewiz.org/wiki/GeekiGeeki">GeekiGeeki</a> version %s
310 ''' % (relative_url('cc-by-sa.png'), clock() - start_time, __version__)
312 print '<p class="modified">last modified %s</p>' % mod_string
313 print '</div></body></html>'
316 """Object that turns Wiki markup into HTML.
318 All formatting commands can be parsed one line at a time, though
319 some state is carried over between lines.
321 def __init__(self, raw):
324 self.in_pre = self.in_html = self.in_table = self.in_li = False
325 self.in_header = True
326 self.list_indents = []
334 "^^": ["sup", False],
335 ",,": ["sub", False],
336 "''": ["em", False], # LEGACY
337 "'''": ["b", False], # LEGACY
338 "``": ["tt", False], # LEGACY
341 def _b_repl(self, word):
342 style = self.styles[word]
343 style[1] = not style[1]
344 return ['</', '<'][style[1]] + style[0] + '>'
346 def _tit_repl(self, word):
348 result = '</h%d><p>\n' % self.h_level
351 self.h_level = len(word) - 1
352 link = permalink(self.line)
353 result = '\n</p><h%d id="%s"><a class="heading" href="#%s">¶</a> ' % (self.h_level, link, link)
356 def _br_repl(self, word):
359 def _rule_repl(self, word):
360 return self._undent() + '\n<hr size="%d" noshade="noshade" />\n' % (len(word) - 2)
362 def _macro_repl(self, word):
363 m = re.compile("\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>").match(word)
367 argv.extend(m.group(2).split('|'))
368 argv = map(str.strip, argv)
370 macro = globals().get('_macro_' + name)
373 execfile("macros/" + name + ".py", globals())
375 if err.errno == errno.ENOENT: pass
376 macro = globals().get('_macro_' + name)
380 msg = '<<' + '|'.join(argv) + '>>'
382 msg = '<strong class="error">' + msg + '</strong>'
385 def _hurl_repl(self, word):
386 m = link_re.match(word)
387 return link_tag(m.group(1), m.group(2))
389 def _inl_repl(self, word):
390 m = link_re.match(word)
391 name = relative_url(m.group(1))
395 argv = descr.split('|')
399 args = '?' + '&'.join(argv)
401 # The "extthumb" nonsense works around a limitation of the HTML block model
402 return '<div class="extthumb"><div class="thumb">' \
403 + link_inline(name, descr, args) \
404 + '<div class="caption">' + descr + '</div></div></div>'
406 return link_inline(name, name)
408 def _html_repl(self, word):
410 return word; # Pass through
412 def _htmle_repl(self, word):
414 return word; # Pass through
416 def _ent_repl(self, s):
418 return s; # Pass through
419 return {'&': '&',
423 def _img_repl(self, word): # LEGACY
424 return self._inl_repl('{{' + word + '}}')
426 def _word_repl(self, word): # LEGACY
427 if self.in_html: return word # pass through
428 return link_tag(word)
430 def _url_repl(self, word): # LEGACY
431 if self.in_html: return word # pass through
432 return link_tag(word)
434 def _email_repl(self, word): # LEGACY
435 if self.in_html: return word # pass through
436 return '<a href="mailto:%s">%s</a>' % (word, word)
438 def _li_repl(self, match):
445 def _pre_repl(self, word):
446 if word == '{{{' and not self.in_pre:
454 def _hi_repl(self, word):
455 return '<strong class="highlight ' + word + '">' + word + '</strong>'
457 def _tr_repl(self, word):
459 if not self.in_table:
462 out = '</p><table><tbody>\n'
464 out = out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '">'
465 return out + ['<td>', '<th>'][word.strip() == '||=']
467 def _td_repl(self, word):
469 return ['</td><td>', '</th><th>'][word.strip() == '||=']
472 def _tre_repl(self, word):
474 return ['</td></tr>', '</th></tr>'][word.strip() == '||=']
477 def _indent_level(self):
478 return len(self.list_indents) and self.list_indents[-1]
480 def _indent_to(self, new_level):
481 if self._indent_level() == new_level:
484 while self._indent_level() > new_level:
485 del(self.list_indents[-1])
488 self.in_li = False # FIXME
490 while self._indent_level() < new_level:
491 self.list_indents.append(new_level)
498 res += '</ul>' * len(self.list_indents)
500 self.list_indents = []
503 def replace(self, match):
504 for rule, hit in match.groupdict().items():
506 return getattr(self, '_' + rule + '_repl')(hit)
508 raise "Can't handle match " + repr(match)
510 def print_html(self):
511 print '<div class="wiki"><p>'
513 scan_re = re.compile(r"""(?:
514 # Styles and formatting
515 (?P<b> \*\*|'''|//|''|\#\#|``|__|\^\^|,,)
519 | (?P<hi> \b( FIXME | TODO | DONE )\b )
522 | (?P<macro> \<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)
523 | (?P<hurl> \[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])
526 | (?P<html> <(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])\b )
527 | (?P<htmle> ( /\s*> | </(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])> ) )
530 # Auto links (LEGACY)
531 | (?P<img> \b[a-zA-Z0-9_/-]+\.(png|gif|jpg|jpeg|bmp|ico|ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt))
532 | (?P<word> \b(?:[A-Z][a-z]+){2,}\b)
533 | (?P<url> (http|https|ftp|mailto)\:[^\s'\"]+\S)
534 | (?P<email> [-\w._+]+\@[\w.-]+)
537 | (?P<li> ^\s+[\*\#]\s+)
538 | (?P<pre> \{\{\{|\s*\}\}\})
539 | (?P<inl> \{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})
542 | (?P<tr> ^\s*\|\|(=|)\s*)
543 | (?P<tre> \s*\|\|(=|)\s*$)
544 | (?P<td> \s*\|\|(=|)\s*)
546 # TODO: highlight search words (look at referrer)
548 pre_re = re.compile("""(?:
552 blank_re = re.compile(r"^\s*$")
553 indent_re = re.compile(r"^\s*")
554 tr_re = re.compile(r"^\s*\|\|")
555 eol_re = re.compile(r"\r?\n")
557 # For each line, we scan through looking for magic strings, outputting verbatim any intervening text
558 for self.line in eol_re.split(self.raw.expandtabs()):
561 if self.line.startswith('#'):
563 self.in_header = False
566 print re.sub(pre_re, self.replace, self.line)
568 if self.in_table and not tr_re.match(self.line):
569 self.in_table = False
570 print '</tbody></table><p>'
572 if blank_re.match(self.line):
575 indent = indent_re.match(self.line)
576 print self._indent_to(len(indent.group(0))) ,
577 print re.sub(scan_re, self.replace, self.line)
579 if self.in_pre: print '</pre>'
580 if self.in_table: print '</tbody></table><p>'
585 def __init__(self, page_name):
586 self.page_name = page_name
588 self.msg_type = 'error'
590 def split_title(self):
591 # look for the end of words and the start of a new word and insert a space there
592 return re.sub('([a-z])([A-Z])', r'\1 \2', self.page_name)
595 return os.path.join(data_dir, self.page_name)
597 def _tmp_filename(self):
598 return os.path.join(data_dir, ('#' + self.page_name.replace('/','_') + '.' + `os.getpid()` + '#'))
602 os.stat(self._filename())
605 if err.errno == errno.ENOENT:
609 def get_raw_body(self):
611 return open(self._filename(), 'rb').read()
613 if err.errno == errno.ENOENT:
614 return '' # just doesn't exist, use default
615 if err.errno == errno.EISDIR:
616 return self.format_dir()
619 def format_dir(self):
622 for dirname in self.page_name.split('/'):
623 pathname = (pathname + '/' + dirname) if pathname else dirname
624 out += '[[' + pathname + '|' + dirname + ']]/'
627 for filename in page_list(self._filename(), file_re):
628 if img_re.match(filename):
630 maxwidth_arg = '|maxwidth=' + str(image_maxwidth)
631 out += '{{' + self.page_name + '/' + filename + '|' + filename + maxwidth_arg + '}}\n'
633 out += ' * [[' + self.page_name + '/' + filename + ']]\n'
637 if not '_pragmas' in self.__dict__:
640 f = open(self._filename(), 'rt')
641 attr_re = re.compile(r"^#(\S*)(.*)$")
643 m = attr_re.match(line)
646 self._pragmas[m.group(1)] = m.group(2).strip()
647 #print "bernie: _pragmas[" + m.group(1) + "] = " + m.group(2) + "<br>\n"
649 if err.errno != errno.ENOENT and err.errno != errno.EISDIR:
653 def pragma(self, name, default):
654 return self.pragmas().get(name, default)
656 def can(self, action, default=True):
659 #acl SomeUser:read,write All:read
660 acl = self.pragma("acl", None)
661 for rule in acl.split():
662 (user, perms) = rule.split(':')
663 if user == remote_user() or user == "All":
664 return action in perms.split(',')
668 self.msg_text = 'Illegal acl line: ' + acl
672 return self.can("write", True)
675 return self.can("read", True)
677 def send_naked(self):
679 WikiFormatter(self.get_raw_body()).print_html()
681 send_guru("Read access denied by ACLs", "notice")
685 value = self.pragma("css", None)
688 link_urls += [ [ "stylesheet", value ] ]
690 send_title(self.page_name, self.split_title(),
691 msg_text=self.msg_text, msg_type=self.msg_type, writable=self.can_write())
693 send_footer(self._last_modified())
695 def _last_modified(self):
697 from time import localtime, strftime
698 modtime = localtime(os.stat(self._filename())[stat.ST_MTIME])
700 if err.errno != errno.ENOENT:
703 return strftime(datetime_fmt, modtime)
705 def send_editor(self, preview=None):
706 send_title(None, 'Edit ' + self.split_title(), msg_text=self.msg_text, msg_type=self.msg_type)
707 if not self.can_write():
708 send_guru("Write access denied by ACLs", "error")
713 filename = form['file'].value
715 print ('<p><b>Editing ' + self.page_name
716 + ' for ' + cgi.escape(remote_user())
717 + ' from ' + cgi.escape(get_hostname(remote_host()))
719 print '<div class="editor"><form name="editform" method="post" enctype="multipart/form-data" action="%s">' % relative_url(self.page_name)
720 print '<input type="hidden" name="edit" value="%s">' % (self.page_name)
721 print '<input type="input" id="editor" name="changelog" value="Edit page %s" accesskey="c" /><br />' % (self.page_name)
722 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())
723 print '<label for="file" accesskey="u">Or Upload a file:</label> <input type="file" name="file" value="%s" />' % filename
726 <input type="submit" name="save" value="Save" accesskey="s">
727 <input type="submit" name="preview" value="Preview" accesskey="p" />
728 <input type="reset" value="Reset" />
729 <input type="submit" name="cancel" value="Cancel" />
732 <script language="javascript">
734 document.editform.savetext.focus()
738 print "<p>" + link_tag('EditingTips') + "</p>"
740 print "<div class='preview'>"
741 WikiFormatter(preview).print_html()
745 def send_raw(self, mimetype='text/plain', args=[]):
746 if not self.can_read():
747 send_title(None, msg_text='Read access denied by ACLs', msg_type='notice')
750 if 'maxwidth' in args:
752 emit_header(mimetype)
754 subprocess.check_call(['gm', 'convert', self._filename(),
755 '-scale', args['maxwidth'].value + ' >', '-'])
757 body = self.get_raw_body()
758 emit_header(mimetype)
761 def _write_file(self, data):
762 tmp_filename = self._tmp_filename()
763 open(tmp_filename, 'wb').write(data)
764 name = self._filename()
766 # Bad Bill! POSIX rename ought to replace. :-(
770 if err.errno != errno.ENOENT: raise err
771 os.rename(tmp_filename, name)
773 def save(self, newdata, changelog):
774 if not self.can_write():
775 self.msg_text = 'Write access denied by ACLs'
776 self.msg_type = 'error'
779 self._write_file(newdata)
783 cmd = [ post_edit_hook, data_dir + '/' + self.page_name, remote_user(), remote_host(), changelog]
784 child = subprocess.Popen(cmd, stdout=subprocess.PIPE, close_fds=True)
785 output = child.stdout.read()
788 self.msg_text += "Post-editing hook returned %d.\n" % rc
789 self.msg_text += 'Command was: ' + ' '.join(cmd) + '\n'
791 self.msg_text += 'Output follows:\n' + output
793 self.msg_text = 'Thank you for your contribution. Your attention to detail is appreciated.'
794 self.msg_type = 'success'
798 handler = globals().get('handle_' + cmd)
800 handler(form[cmd].value)
803 query = query_string()
804 if file_re.match(query):
805 # FIMXE: this is all bullshit, MimeTypes bases its guess on the extension!
806 from mimetypes import MimeTypes
807 mimetype, encoding = MimeTypes().guess_type(query)
809 Page(query).send_raw(mimetype=mimetype, args=form)
813 send_httperror("403 Forbidden", query)
816 execfile("geekigeeki.conf.py")
817 form = cgi.FieldStorage()
821 msg_text = traceback.format_exc()
823 send_guru(msg_text, "error")
825 send_title(None, msg_text=msg_text)