Flag legacy wiki markup for styles as such
[geekigeeki.git] / geekigeeki.py
index 651d98db20e984506af833ebf7c810e20a8ae635..38870c8539c1bb39fc29b61be63878cefe0f635d 100755 (executable)
@@ -37,12 +37,17 @@ link_re = re.compile("(?:\[\[|{{)([^\s\|]+)(?:\s*\|\s*([^\]]+)|)(?:\]\]|}})")
 
 title_done = False
 
-
 # CGI stuff ---------------------------------------------------------
-
 def script_name():
     return os.environ.get('SCRIPT_NAME', '')
 
+def query_string():
+    path_info = os.environ.get('PATH_INFO', '')
+    if len(path_info) and path_info[0] == '/':
+        return path_info[1:] or 'FrontPage'
+    else:
+        return os.environ.get('QUERY_STRING', '') or 'FrontPage'
+
 def privileged_path():
     return privileged_url or script_name()
 
@@ -316,14 +321,14 @@ class WikiFormatter:
         self.styles = {
             #wiki   html   enabled?
             "//":  ["em",  False],
-            "''":  ["em",  False],
             "**":  ["b",   False],
-            "'''": ["b",   False],
             "##":  ["tt",  False],
-            "``":  ["tt",  False],
             "__":  ["u",   False],
             "^^":  ["sup", False],
-            ",,":  ["sub", False]
+            ",,":  ["sub", False],
+            "''":  ["em",  False], # OBSOLETE
+            "'''": ["b",   False], # OBSOLETE
+            "``":  ["tt",  False], # OBSOLETE
         }
 
     def _b_repl(self, word):
@@ -365,7 +370,10 @@ class WikiFormatter:
         if macro:
             return macro(argv)
         else:
-            return '<strong class="error">&lt;&lt;' + '|'.join(argv) + '&gt;&gt;</strong>'
+            msg = '&lt;&lt;' + '|'.join(argv) + '&gt;&gt;'
+            if not self.in_html:
+                msg = '<strong class="error">' + msg + '</strong>'
+            return msg
 
     def _hurl_repl(self, word):
         m = link_re.match(word)
@@ -407,10 +415,13 @@ class WikiFormatter:
         self.in_html += 1
         return word; # Pass through
 
+    def _htmle_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 '>'
+        if self.in_html:
+            return s; # Pass through
         return {'&': '&amp;',
                 '<': '&lt;',
                 '>': '&gt;'}[s]
@@ -490,51 +501,51 @@ class WikiFormatter:
     def print_html(self):
         print '<div class="wiki"><p>'
 
-        # For each line, we scan through looking for magic
-        # strings, outputting verbatim any intervening text
-        # TODO: highlight search words (look at referrer)
-        scan_re = re.compile(
-            r"(?:"
-            # Formatting
-            + r"(?P<b>\*\*|'''|//|''|##|``|__|\^\^|,,)"
-            + r"|(?P<tit>\={2,6})"
-            + r"|(?P<br>\\\\)"
-            + r"|(?P<rule>^-{3,})"
-            + r"|(?P<hi>\b(FIXME|TODO|DONE)\b)"
+        scan_re = re.compile(r"""(?:
+            # Styles and formatting
+              (?P<b>     \*\*|'''|//|''|\#\#|``|__|\^\^|,,)
+            | (?P<tit>   \={2,6})
+            | (?P<br>    \\\\)
+            | (?P<rule>  ^-{3,})
+            | (?P<hi>    \b( FIXME | TODO | DONE )\b )
 
             # Links
-            + r"|(?P<macro>\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)"
-            + r"|(?P<hurl>\[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])"
+            | (?P<macro> \<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)
+            | (?P<hurl>  \[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])
 
             # Inline HTML
-            + r"|(?P<html><(/|)(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])[^>]*>)"
-            + r"|(?P<ent>[<>&])"
+            | (?P<html>  <(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])\b )
+            | (?P<htmle> ( /\s*> | </(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])> ) )
+            | (?P<ent>   [<>&] )
 
             # Auto links (LEGACY)
-            + r"|(?P<img>\b[a-zA-Z0-9_/-]+\.(png|gif|jpg|jpeg|bmp|ico|ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt))"
-            + r"|(?P<word>\b(?:[A-Z][a-z]+){2,}\b)"
-            + r"|(?P<url>(http|https|ftp|mailto)\:[^\s'\"]+\S)"
-            + r"|(?P<email>[-\w._+]+\@[\w.-]+)"
+            | (?P<img>   \b[a-zA-Z0-9_/-]+\.(png|gif|jpg|jpeg|bmp|ico|ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt))
+            | (?P<word>  \b(?:[A-Z][a-z]+){2,}\b)
+            | (?P<url>   (http|https|ftp|mailto)\:[^\s'\"]+\S)
+            | (?P<email> [-\w._+]+\@[\w.-]+)
 
             # Lists, divs, spans
-            + r"|(?P<li>^\s+[\*#] +)"
-            + r"|(?P<pre>\{\{\{|\s*\}\}\})"
-            + r"|(?P<inl>\{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})"
+            | (?P<li>    ^\s+[\*\#]\s+)
+            | (?P<pre>   \{\{\{|\s*\}\}\})
+            | (?P<inl>   \{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})
 
             # Tables
-            + r"|(?P<tr>^\s*\|\|(=|)\s*)"
-            + r"|(?P<tre>\s*\|\|(=|)\s*$)"
-            + r"|(?P<td>\s*\|\|(=|)\s*)"
-            + r")")
-        pre_re = re.compile(
-            r"(?:"
-            + r"(?P<pre>\s*\}\}\})"
-            + r"|(?P<ent>[<>&])"
-            + r")")
+            | (?P<tr>    ^\s*\|\|(=|)\s*)
+            | (?P<tre>   \s*\|\|(=|)\s*$)
+            | (?P<td>    \s*\|\|(=|)\s*)
+
+            # TODO: highlight search words (look at referrer)
+          )""", re.VERBOSE)
+        pre_re = re.compile("""(?:
+              (?P<pre>\s*\}\}\})
+            | (?P<ent>[<>&])"
+            )""", re.VERBOSE)
         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")
+
+        # For each line, we scan through looking for magic strings, outputting verbatim any intervening text
         for self.line in eol_re.split(self.raw.expandtabs()):
             # Skip pragmas
             if self.in_header:
@@ -787,12 +798,7 @@ def main():
             handler(form[cmd].value)
             break
     else:
-        path_info = os.environ.get('PATH_INFO', '')
-        if len(path_info) and path_info[0] == '/':
-            query = path_info[1:] or 'FrontPage'
-        else:
-            query = os.environ.get('QUERY_STRING', '') or 'FrontPage'
-
+        query = query_string()
         if file_re.match(query):
             if word_re.match(query):
                 Page(query).format()