' - print ' ' + site_name + ': ', if name: - print link_tag('?fullsearch=' + name, text, 'navlink') + ' ' + print ' ' + link_tag('?fullsearch=' + name, text, 'navlink') + ' ' else: - print text + ' ' + print ' ' + text + ' ' print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink') print ' | ' + link_tag('FindPage', 'Find Page', 'navlink') - if globals().has_key('history_url'): + if 'history_url' in globals(): print ' | Recent Changes' if name: print ' | Page History' @@ -171,11 +172,13 @@ def link_tag(params, text=None, ss_class=None, privileged=False): # Prevent crawlers from following links potentially added by spammers or to generated pages if ss_class == 'external' or ss_class == 'navlink': classattr += 'rel="nofollow" ' + elif url_re.match(params): + classattr += 'rel="nofollow" ' return '%s' % (classattr, relative_url(params, privileged=privileged), text) # Search --------------------------------------------------- -def do_fullsearch(needle): +def handle_fullsearch(needle): send_title(None, 'Full text search for "%s"' % (needle)) needle_re = re.compile(needle, re.IGNORECASE) @@ -202,7 +205,7 @@ def do_fullsearch(needle): print_search_stats(len(hits), len(all_pages)) -def do_titlesearch(needle): +def handle_titlesearch(needle): # TODO: check needle is legal -- but probably we can just accept any RE send_title(None, "Title search for \"" + needle + '"') @@ -220,45 +223,39 @@ def do_titlesearch(needle): def print_search_stats(hits, searched): print "

%d hits out of %d pages searched.

" % (hits, searched) -#TODO: merge into do_savepage() -def do_edit(pagename): - Page(pagename).send_editor() - -def do_raw(pagename): +def handle_raw(pagename): Page(pagename).send_raw() -def do_savepage(pagename): - global form +def handle_edit(pagename): pg = Page(pagename) - if 'preview' in form: - pg.send_editor(form['savetext'].value) - elif 'save' in form: - pg.save_text(form['savetext'].value) - pg.send_page() + if 'save' in form: + if form['file'].value: + pg.save(form['file'].file.read()) + else: + pg.save(form['savetext'].value.replace('\r\n', '\n')) + pg.format() elif 'cancel' in form: - pg.msg = 'Editing canceled' + pg.msg_text = 'Editing canceled' pg.msg_type = 'notice' - pg.send_page() - else: - raise 'What did you press?' + pg.format() + else: # preview or edit + text = None + if 'preview' in form: + text = form['savetext'].value + pg.send_editor(text) def make_index_key(): - s = '

' - links = map(lambda ch: '%s' % (ch, ch), - string.lowercase) - s = s + string.join(links, ' | ') - s = s + '

' - return s + links = map(lambda ch: '%s' % (ch, ch), 'abcdefghijklmnopqrstuvwxyz') + return '

'+ ' | '.join(links) + '

' -def page_list(): - return filter(word_re.match, os.listdir(data_dir)) +def page_list(dir = None, re = word_re): + return filter(re.match, os.listdir(dir or data_dir)) def send_footer(name, mod_string=None): if globals().get('debug_cgi', False): cgi.print_arguments() - cgi.print_form(cgi.FieldStorage()) + cgi.print_form(form) cgi.print_environ() - global __version__ print '' -# ---------------------------------------------------------- -# Macros -def _macro_TitleSearch(*vargs): - return _macro_search("titlesearch") - -def _macro_FullSearch(*vargs): - return _macro_search("fullsearch") - -def _macro_search(type): - if form.has_key('value'): - default = form["value"].value - else: - default = '' - return """""" % (type, default) - -def _macro_WordIndex(*vargs): - s = make_index_key() - pages = list(page_list()) - map = {} - word_re = re.compile('[A-Z][a-z]+') - for name in pages: - for word in word_re.findall(name): - try: - map[word].append(name) - except KeyError: - map[word] = [name] - - all_words = map.keys() - all_words.sort() - last_letter = None - # set title - for word in all_words: - letter = string.lower(word[0]) - if letter != last_letter: - s = s + ';

%s

' % (letter, letter) - last_letter = letter - - s = s + '%s

' + Page(name).link_to() - s = s + '

' - return s - - -def _macro_TitleIndex(*vargs): - s = make_index_key() - pages = list(page_list()) - pages.sort() - current_letter = None - for name in pages: - letter = string.lower(name[0]) - if letter != current_letter: - s = s + '

%s

' % (letter, letter) - current_letter = letter - else: - s = s + '
' - s = s + Page(name).link_to() - return s - - -# ---------------------------------------------------------- -class PageFormatter: +class WikiFormatter: """Object that turns Wiki markup into HTML. All formatting commands can be parsed one line at a time, though @@ -341,10 +273,10 @@ class PageFormatter: def __init__(self, raw): self.raw = raw self.h_level = 0 - self.in_pre = self.in_table = False + self.in_pre = self.in_html = self.in_table = self.in_li = False self.in_header = True self.list_indents = [] - self.tr_cnt = self.h_cnt = 0 + self.tr_cnt = 0 self.styles = { #wiki html enabled? "//": ["em", False], @@ -361,17 +293,16 @@ class PageFormatter: def _b_repl(self, word): style = self.styles[word] style[1] = not style[1] - return ['' + return ['' def _tit_repl(self, word): if self.h_level: - result = '' % self.h_level + result = '

\n' % self.h_level self.h_level = 0 else: self.h_level = len(word) - 1 - self.h_cnt += 1 - #abridged = re.sub('[^a-z_]', '', word.lower().replace(' ', '_')) - result = 'Â¶ ' % (self.h_level, self.h_cnt, self.h_cnt) + link = permalink(self.line) + result = '\n

Â¶ ' % (self.h_level, link, link) return result def _br_repl(self, word): @@ -393,41 +324,59 @@ class PageFormatter: else: return '%s' % (word, word) - def _hurl_repl(self, word): - m = re.compile("\[\[([^ \t\n\r\f\v\|]+)(?:\s*\|\s*([^\]]+)|)\]\]").match(word) + def _macro_repl(self, word): + m = re.compile("\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>").match(word) name = m.group(1) - descr = m.group(2) or name + argv = [name] + if m.group(2): + argv.extend(m.group(2).split('|')) + argv = map(str.strip, argv) macro = globals().get('_macro_' + name) + if not macro: + execfile("macros/" + name + ".py", globals()) + macro = globals().get('_macro_' + name) if macro: - return apply(macro, (name, descr)) - elif img_re.match(name): + return macro(argv) + else: + return '<<' + '|'.join(argv) + '>>' + + def _hurl_repl(self, word): + m = re.compile("\[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\]").match(word) + name = m.group(1) + descr = m.group(2) or name + + if img_re.match(name): + name = relative_url(name) # The "extthumb" nonsense works around a limitation of the HTML block model return '

' % (name, name, descr, descr) else: if img_re.match(descr): descr = '

' - if url_re.match(name): - return '%s' % (name, descr) - elif name.startswith('/'): - return '%s' % (name, descr) - else: - return link_tag(name, descr, 'wikilink') + return link_tag(name, descr, 'wikilink') def _email_repl(self, word): return '%s' % (word, word) def _html_repl(self, word): + self.in_html += 1 return word; # Pass through def _ent_repl(self, s): + if self.in_html and s == '>': + self.in_html -= 1 + return '>' return {'&': '&', '<': '<', '>': '>'}[s] def _li_repl(self, match): - return '

' + if self.in_li: + return '

' + else: + self.in_li = True + return '

' def _pre_repl(self, word): if word == '{{{' and not self.in_pre: @@ -439,13 +388,7 @@ class PageFormatter: return '' def _hi_repl(self, word): - if word == 'FIXME': - cl = 'error' - elif word == 'DONE': - cl = 'success' - elif word == 'TODO': - cl = 'notice' - return '' + word + '' + return '' + word + '' def _tr_repl(self, word): out = '' @@ -476,6 +419,9 @@ class PageFormatter: s = '

' while self._indent_level() > new_level: del(self.list_indents[-1]) + if self.in_li: + s += '

' + self.in_li = False # FIXME s += '\n' while self._indent_level() < new_level: self.list_indents.append(new_level) @@ -493,7 +439,7 @@ class PageFormatter: def replace(self, match): for type, hit in match.groupdict().items(): if hit: - return apply(getattr(self, '_' + type + '_repl'), (hit,)) + return getattr(self, '_' + type + '_repl')(hit) else: raise "Can't handle match " + `match` @@ -502,7 +448,7 @@ class PageFormatter: # For each line, we scan through looking for magic # strings, outputting verbatim any intervening text - # TODO: highlight search words (look at referer) + # TODO: highlight search words (look at referrer) scan_re = re.compile( r"(?:" # Formatting @@ -510,20 +456,26 @@ class PageFormatter: + r"|(?P\={2,6})" + r"|(?P
\\\\)" + r"|(?P^-{3,})" - + r"|(?P<(/|)(div|span|iframe)[^<>]*>)" - + r"|(?P[<>&])" + r"|(?P\b(FIXME|TODO|DONE)\b)" # Links - + r"|(?P\b[a-zA-Z0-9_-]+\.(png|gif|jpg|jpeg|bmp))" - + r"|(?P\b(?:[A-Z][a-z]+){2,}\b)" - + r"|(?P\[\[([^ \t\n\r\f\v\|]+)(?:\s*\|\s*([^\]]+)|)\]\])" - + r"|(?P(http|https|ftp|mailto)\:[^\s'\"]+\S)" - + r"|(?P[-\w._+]+\@[\w.-]+)" + + r"|(?P\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)" + + r"|(?P\[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])" + + # Inline HTML + + r"|(?P<(/|)(br|hr|div|form|iframe|input|span))" + + r"|(?P[<>&])" + + # Auto links + + r"|(?P\b[a-zA-Z0-9_/-]+\.(png|gif|jpg|jpeg|bmp))" # LEGACY + + r"|(?P\b(?:[A-Z][a-z]+){2,}\b)" # LEGACY + + r"|(?P(http|https|ftp|mailto)\:[^\s'\"]+\S)" # LEGACY + + r"|(?P[-\w._+]+\@[\w.-]+)" # LEGACY # Lists, divs, spans + r"|(?P

^\s+[\*#] +)" + r"|(?P

\{\{\{|\s*\}\}\})"
+            + r"|(?P\{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})" #TODO
 
             # Tables
             + r"|(?P^\s*\|\|(=|)\s*)"
@@ -533,65 +485,62 @@ class PageFormatter:
         pre_re = re.compile(
             r"(?:"
             + r"(?P\s*\}\}\})"
+            + r"|(?P[<>&])"
             + r")")
         blank_re = re.compile(r"^\s*$")
         indent_re = re.compile(r"^\s*")
         tr_re = re.compile(r"^\s*\|\|")
         eol_re = re.compile(r"\r?\n")
-        raw = string.expandtabs(self.raw)
-        for line in eol_re.split(raw):
+        for self.line in eol_re.split(self.raw.expandtabs()):
             # Skip ACLs
             if self.in_header:
-                if line.startswith('#'):
-                   continue
+                if self.line.startswith('#'):
+                    continue
                 self.in_header = False
 
             if self.in_pre:
-                print re.sub(pre_re, self.replace, line)
+                print re.sub(pre_re, self.replace, self.line)
             else:
-                if self.in_table and not tr_re.match(line):
+                if self.in_table and not tr_re.match(self.line):
                     self.in_table = False
                     print ''
 
-                if blank_re.match(line):
+                if blank_re.match(self.line):
                     print '
'
                 else:
-                    indent = indent_re.match(line)
+                    indent = indent_re.match(self.line)
                     print self._indent_to(len(indent.group(0)))
-                    print re.sub(scan_re, self.replace, line)
+                    print re.sub(scan_re, self.replace, self.line)
 
         if self.in_pre: print ''
         if self.in_table: print ''
         print self._undent()
         print '

Editing ' + self.page_name + ' for ' + cgi.escape(remote_user()) + ' from ' + cgi.escape(get_hostname(remote_host())) + '

' % relative_url(self.page_name) - print '' % (self.page_name) - print """""" % (preview or self.get_raw_body()) + print '

' % relative_url(self.page_name) + print '' % (self.page_name) + print '' % (preview or self.get_raw_body()) + print 'Or upload a file: ' % file print """
@@ -705,36 +667,37 @@ class Page: print "

" + Page('EditingTips').link_to() + "

" if preview: print "

" - PageFormatter(preview).print_html() + WikiFormatter(preview).print_html() print "

" send_footer(self.page_name) - def send_raw(self): - if not self.can_read(): - send_title(None, msg='Read access denied by ACLs', msg_type='notice') - return - emit_header("text/plain") - print self.get_raw_body() + def send_raw(self, mimetype='text/plain'): + if self.can_read(): + body = self.get_raw_body() + emit_header(mimetype) + print body + else: + send_title(None, msg_text='Read access denied by ACLs', msg_type='notice') - def _write_file(self, text): + def _write_file(self, data): tmp_filename = self._tmp_filename() - open(tmp_filename, 'wt').write(text.replace('\r\n', '\n')) - text = self._text_filename() + open(tmp_filename, 'wb').write(data) + name = self._filename() if os.name == 'nt': # Bad Bill! POSIX rename ought to replace. :-( try: - os.remove(text) + os.remove(name) except OSError, er: if er.errno != errno.ENOENT: raise er - os.rename(tmp_filename, text) + os.rename(tmp_filename, name) - def save_text(self, newtext): + def save(self, newdata): if not self.can_write(): - self.msg = 'Write access denied by ACLs' + self.msg_text = 'Write access denied by ACLs' self.msg_type = 'error' return - self._write_file(newtext) + self._write_file(newdata) rc = 0 if post_edit_hook: # FIXME: what's the std way to perform shell quoting in python? @@ -744,39 +707,26 @@ class Page: + "' '" + remote_host() + "'" ) out = os.popen(cmd) - msg = out.read() + output = out.read() rc = out.close() if rc: - self.msg += "Post-editing hook returned %d.\n" % rc - self.msg += 'Command was: ' + cmd + '\n' - if msg: - self.msg += 'Output follows:\n' + msg + self.msg_text += "Post-editing hook returned %d.\n" % rc + self.msg_text += 'Command was: ' + cmd + '\n' + if output: + self.msg_text += 'Output follows:\n' + output else: - self.msg = 'Thank you for your contribution. Your attention to detail is appreciated.' + self.msg_text = 'Thank you for your contribution. Your attention to detail is appreciated.' self.msg_type = 'success' -#TODO: merge into send_raw() -def send_verbatim(filename, mime_type='application/octet-stream'): - pathname = path.join(data_dir, filename) - data = open(pathname, 'rb').read() - emit_header(mime_type) - sys.stdout.write(data) - # Main --------------------------------------------------------------- try: execfile("geekigeeki.conf.py") - form = cgi.FieldStorage() - handlers = { 'fullsearch': do_fullsearch, - 'titlesearch': do_titlesearch, - 'edit': do_edit, - 'raw': do_raw, - 'savepage': do_savepage } - - for cmd in handlers.keys(): - if form.has_key(cmd): - apply(handlers[cmd], (form[cmd].value,)) + for cmd in form: + handler = globals().get('handle_' + cmd) + if handler: + handler(form[cmd].value) break else: path_info = environ.get('PATH_INFO', '') @@ -787,22 +737,23 @@ try: if file_re.match(query): if word_re.match(query): - Page(query).send_page() - elif img_re.match(query): - #FIXME: use correct mime type - send_verbatim(query, 'image/jpeg') + Page(query).format() else: - send_verbatim(query) + from mimetypes import MimeTypes + type, encoding = MimeTypes().guess_type(query) + type = type or 'text/plain' + Page(query).send_raw(mimetype=type) else: print "Status: 404 Not Found" - send_title(None, msg='Can\'t work out query: ' + query) -except: + send_title(None, msg_text='Can\'t work out query: ' + query) + send_footer(None) +except Exception: import traceback - msg=traceback.format_exc() + msg_text = traceback.format_exc() if title_done: - send_guru(msg, "error") + send_guru(msg_text, "error") else: - send_title(None, msg=msg) + send_title(None, msg_text=msg_text) send_footer(None) sys.stdout.flush()