Make permalinks more search engine friendly

[geekigeeki.git] / geekigeeki.py

diff --git a/geekigeeki.py b/geekigeeki.py
index a6158a3122b41b184930bf0255664a428851858d..3b66445ab80b4665304db00481b344b31b4376e1 100755 (executable)
--- a/geekigeeki.py
+++ b/geekigeeki.py
@@ -70,8 +70,10 @@ def relative_url(pathname, privileged=False):
         pathname = url + '/' + pathname
     return pathname
 
-# Formatting stuff --------------------------------------------------
+def permalink(s):
+    return re.sub(' ', '-', re.sub('[^a-z0-9_ ]', '', s.lower()).strip())
 
+# Formatting stuff --------------------------------------------------
 def emit_header(mime_type="text/html"):
     print "Content-type: " + mime_type + "; charset=utf-8\n"
 
@@ -137,11 +139,10 @@ def send_title(name, text="Limbo", msg_text=None, msg_type='error'):
 
     # Navbar
     print '<div class="nav">'
-    print '  <b>' + site_name + ': ',
     if name:
-        print link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
+        print '  <b>' + link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
     else:
-        print text + '</b> '
+        print '  <b>' + text + '</b> '
     print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink')
     print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
     if 'history_url' in globals():
@@ -263,8 +264,7 @@ def send_footer(name, mod_string=None):
         print '<p class="modified">last modified %s</p>' % mod_string
     print '</div></body></html>'
 
-# ----------------------------------------------------------
-# Macros
+# Macros ------------------------------------------------------------
 def _macro_TitleSearch(*vargs):
     return _macro_search("titlesearch")
 
@@ -325,8 +325,6 @@ def _macro_TitleIndex(*vargs):
         s += Page(name).link_to()
     return s
 
-
-# ----------------------------------------------------------
 class PageFormatter:
     """Object that turns Wiki markup into HTML.
 
@@ -336,10 +334,10 @@ 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_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],
@@ -360,13 +358,12 @@ class PageFormatter:
 
     def _tit_repl(self, word):
         if self.h_level:
-            result = '</h%d>' % self.h_level
+            result = '</h%d><p>\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 = '<h%d id="%d"><a class="heading" href="#%d">¶</a> ' % (self.h_level, self.h_cnt, self.h_cnt)
+            link = permalink(self.line)
+            result = '\n</p><h%d id="%s"><a class="heading" href="#%s">¶</a> ' % (self.h_level, link, link)
         return result
 
     def _br_repl(self, word):
@@ -418,7 +415,11 @@ class PageFormatter:
                 '>': '&gt;'}[s]
 
     def _li_repl(self, match):
-        return '<li>'
+        if self.in_li:
+            return '</li><li>'
+        else:
+            self.in_li = True
+            return '<li>'
 
     def _pre_repl(self, word):
         if word == '{{{' and not self.in_pre:
@@ -461,6 +462,9 @@ class PageFormatter:
         s = '</p>'
         while self._indent_level() > new_level:
             del(self.list_indents[-1])
+            if self.in_li:
+                s += '</li>'
+                self.in_li = False # FIXME
             s += '</ul>\n'
         while self._indent_level() < new_level:
             self.list_indents.append(new_level)
@@ -524,33 +528,32 @@ class PageFormatter:
         indent_re = re.compile(r"^\s*")
         tr_re = re.compile(r"^\s*\|\|")
         eol_re = re.compile(r"\r?\n")
-        for line in eol_re.split(self.raw.expandtabs()):
+        for self.line in eol_re.split(self.raw.expandtabs()):
             # Skip ACLs
             if self.in_header:
-                if line.startswith('#'):
+                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 '</tbody></table><p>'
 
-                if blank_re.match(line):
+                if blank_re.match(self.line):
                     print '</p><p>'
                 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 '</pre>'
         if self.in_table: print '</tbody></table><p>'
         print self._undent()
         print '</p></div>'
 
-# ----------------------------------------------------------
 class Page:
     def __init__(self, page_name):
         self.page_name = page_name
@@ -615,6 +618,7 @@ class Page:
         return self.get_attrs().get(name, default)
 
     def can(self, action, default=True):
+        acl = None
         try:
             #acl SomeUser:read,write All:read
             acl = self.get_attr("acl", None)