X-Git-Url: https://codewiz.org/gitweb?a=blobdiff_plain;f=geekigeeki.py;h=a1c99a02742b3a862e2590af9b64addcd40fe6a3;hb=4b551793f87baab929ddac359aa48b3eac34ac54;hp=e110d8dafe256395a021980b7d8813eb0fa5d58c;hpb=1549a288ad7fa65066b0b52739618c011898d2a3;p=geekigeeki.git

diff --git a/geekigeeki.py b/geekigeeki.py
index e110d8d..a1c99a0 100755
--- a/geekigeeki.py
+++ b/geekigeeki.py
@@ -1,8 +1,9 @@
-#! /usr/bin/env python
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
 #
 # Copyright 1999, 2000 Martin Pool <mbp@humbug.org.au>
 # Copyright 2002 Gerardo Poggiali
-# Copyright 2007 Bernardo Innocenti <bernie@codewiz.org>
+# Copyright 2007, 2008 Bernardo Innocenti <bernie@codewiz.org>
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -27,8 +28,8 @@ from os import path, environ
 
 # Regular expression defining a WikiWord
 # (but this definition is also assumed in other places)
-file_re = re.compile(r"^\b([A-Za-z0-9_\.\-]+)\b$")
-word_re = re.compile(r"^\b([A-Z][a-z]+){2,}\b$")
+file_re = re.compile(r"^\b([A-Za-z0-9_\.\-/]+)\b$")
+word_re = re.compile(r"^\b((([A-Z][a-z]+){2,}/)*([A-Z][a-z]+){2,})\b$")
 img_re = re.compile(r"^.*\.(png|gif|jpg|jpeg)$", re.IGNORECASE)
 url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$")
 
@@ -59,6 +60,15 @@ def get_hostname(addr):
     except:
         return addr
 
+def relative_url(path, privileged=False):
+    if not (url_re.match(path) or path.startswith('/')):
+        if privileged:
+            url = privileged_path()
+        else:
+            url = script_name()
+        path = url + '/' + path
+    return path
+
 # Formatting stuff --------------------------------------------------
 
 def emit_header(type="text/html"):
@@ -74,7 +84,7 @@ def send_guru(msg, msg_type):
     if msg_type == 'error':
         print '      Guru Meditation #DEADBEEF.ABADC0DE'
     print '</pre>'
-    # FIXME: This simple JS snippet is harder to pass than ACID 3.0 
+    # FIXME: This little JS snippet is harder to pass than ACID 3.0 
     print """
     <script language="JavaScript" type="text/javascript">
         var guru = document.getElementById('guru');
@@ -112,8 +122,8 @@ def send_title(name, text="Limbo", msg=None, msg_type='error'):
     print ' <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8" />'
     if not name:
         print ' <meta name="robots" content="noindex,nofollow" />'
-    if globals().has_key('css_url'):
-        print ' <link rel="stylesheet" type="text/css" href="%s" />' % css_url
+    for css in css_url:
+        print ' <link rel="stylesheet" type="text/css" href="%s" />' % relative_url(css)
     print '</head>'
 
     # Body
@@ -129,7 +139,7 @@ def send_title(name, text="Limbo", msg=None, msg_type='error'):
     print '<div class="nav">'
     print '  <b>' + site_name + ': ',
     if name:
-        print link_tag('?fullsearch=' + name, text) + '</b> '
+        print link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
     else:
         print text + '</b> '
     print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink')
@@ -142,7 +152,7 @@ def send_title(name, text="Limbo", msg=None, msg_type='error'):
     if name:
         print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
         if privileged_url is not None:
-            print ' | ' + link_tag('?edit=' + name, 'Edit Page', 'navlink', authentication=True)
+            print ' | ' + link_tag('?edit=' + name, 'Edit Page', 'navlink', privileged=True)
     else:
         print ' | <i>Immutable Page</i>'
 
@@ -152,7 +162,7 @@ def send_title(name, text="Limbo", msg=None, msg_type='error'):
 
     print '<hr /></div>'
 
-def link_tag(params, text=None, ss_class=None, authentication=False):
+def link_tag(params, text=None, ss_class=None, privileged=False):
     if text is None:
         text = params # default
     classattr = ''
@@ -161,11 +171,9 @@ def link_tag(params, text=None, ss_class=None, authentication=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" '
-    if authentication:
-        path = privileged_path()
-    else:
-        path = script_name()
-    return '<a %shref="%s/%s">%s</a>' % (classattr, path, params, text)
+    elif url_re.match(params):
+        classattr += 'rel="nofollow" '
+    return '<a %shref="%s">%s</a>' % (classattr, relative_url(params, privileged=privileged), text)
 
 # Search ---------------------------------------------------
 
@@ -214,14 +222,10 @@ def do_titlesearch(needle):
 def print_search_stats(hits, searched):
     print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
 
-#TODO: merge into do_savepage()
-def do_edit(pagename):
-    Page(pagename).send_editor()
-
 def do_raw(pagename):
     Page(pagename).send_raw()
 
-def do_savepage(pagename):
+def do_edit(pagename):
     global form
     pg = Page(pagename)
     if 'preview' in form:
@@ -234,7 +238,7 @@ def do_savepage(pagename):
         pg.msg_type = 'notice'
         pg.send_page()
     else:
-        raise 'What did you press?'
+        pg.send_editor()
 
 def make_index_key():
     s = '<p><center>'
@@ -245,7 +249,7 @@ def make_index_key():
     return s
 
 def page_list():
-    return filter(word_re.match, os.listdir(text_dir))
+    return filter(word_re.match, os.listdir(data_dir))
 
 def send_footer(name, mod_string=None):
     if globals().get('debug_cgi', False):
@@ -317,11 +321,11 @@ def _macro_TitleIndex(*vargs):
     for name in pages:
         letter = string.lower(name[0])
         if letter != current_letter:
-            s = s + '<a name="%s"><h3>%s</h3></a>' % (letter, letter)
+            s += '<a name="%s"><h3>%s</h3></a>' % (letter, letter)
             current_letter = letter
         else:
-            s = s + '<br />'
-        s = s + Page(name).link_to()
+            s += '<br />'
+        s += Page(name).link_to()
     return s
 
 
@@ -334,23 +338,28 @@ class PageFormatter:
     """
     def __init__(self, raw):
         self.raw = raw
-        self.is_em = self.is_b = 0
         self.h_level = 0
-        self.h_count = 0
-        self.list_indents = []
-        self.in_pre = False
-        self.in_table = False
-        self.tr_cnt = 0
-        self.in_var = False
+        self.in_pre = self.in_table = False
         self.in_header = True
+        self.list_indents = []
+        self.tr_cnt = self.h_cnt = 0
+        self.styles = {
+            #wiki   html   enabled?
+            "//":  ["em",  False],
+            "''":  ["em",  False],
+            "**":  ["b",   False],
+            "'''": ["b",   False],
+            "##":  ["tt",  False],
+            "``":  ["tt",  False],
+            "__":  ["u",   False],
+            "^^":  ["sup", False],
+            ",,":  ["sub", False]
+        }
 
-    def _emph_repl(self, word):
-        if len(word) == 3:
-            self.is_b = not self.is_b
-            return ['</strong>', '<strong>'][self.is_b]
-        else:
-            self.is_em = not self.is_em
-            return ['</em>', '<em>'][self.is_em]
+    def _b_repl(self, word):
+        style = self.styles[word]
+        style[1] = not style[1]
+        return ['</','<'][style[1]] + style[0] + '>'
 
     def _tit_repl(self, word):
         if self.h_level:
@@ -358,10 +367,14 @@ class PageFormatter:
             self.h_level = 0
         else:
             self.h_level = len(word) - 1
-            self.h_count += 1
-            result = '<h%d id="%d"><a class="heading" href="#%d">*</a> ' % (self.h_level, self.h_count, self.h_count)
+            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)
         return result
 
+    def _br_repl(self, word):
+        return '<br />'
+
     def _rule_repl(self, word):
         return self._undent() + '\n<hr size="%d" noshade="noshade" />\n' % (len(word) - 2)
 
@@ -369,7 +382,7 @@ class PageFormatter:
         return Page(word).link_to()
 
     def _img_repl(self, word):
-        path = script_name() + '/' + word;
+        path = relative_url(word)
         return '<a href="%s"><img border="0" src="%s" /></a>' % (path, path)
 
     def _url_repl(self, word):
@@ -379,21 +392,22 @@ class PageFormatter:
             return '<a href="%s" rel="nofollow" class="external">%s</a>' % (word, word)
 
     def _hurl_repl(self, word):
-        m = re.compile("\[\[(\S+)([^\]]*)\]\]").match(word)
+        m = re.compile("\[\[([^ \t\n\r\f\v\|]+)(?:\s*\|\s*([^\]]+)|)\]\]").match(word)
         name = m.group(1)
-        descr = m.group(2).strip() or name
+        descr = m.group(2) or name
 
         macro = globals().get('_macro_' + name)
         if macro:
             return apply(macro, (name, descr))
         elif img_re.match(name):
-            return '<a href="%s"><img border="0" src="%s" alt="%s" /></a>' % (name, name, descr)
-        elif url_re.match(name):
-            return '<a href="%s" rel="nofollow" class="external">%s</a>' % (name, descr)
-        elif name.startswith('/'):
-            return '<a href="%s">%s</a>' % (name, descr)
+            name = relative_url(name)
+            # The "extthumb" nonsense works around a limitation of the HTML block model
+            return '<div class="extthumb"><div class="thumb"><a href="%s"><img border="0" src="%s" alt="%s" /></a><div class="caption">%s</div></div></div>' % (name, name, descr, descr)
         else:
-            return link_tag(name, descr)
+            if img_re.match(descr):
+                descr = '<img border="0" src="' + descr + '" />'
+
+            return link_tag(name, descr, 'wikilink')
 
     def _email_repl(self, word):
         return '<a href="mailto:%s">%s</a>' % (word, word)
@@ -419,22 +433,7 @@ class PageFormatter:
         return ''
 
     def _hi_repl(self, word):
-        if word == 'FIXME':
-            cl = 'error'
-        elif word == 'DONE':
-            cl = 'success'
-        elif word == 'TODO':
-            cl = 'notice'
-        return '<strong class="highlight ' + cl + '">' + word + '</strong>'
-
-    def _var_repl(self, word):
-        if word == '{{' and not self.in_var:
-            self.in_var = True
-            return '<code>'
-        elif self.in_var:
-            self.in_var = False
-            return '</code>'
-        return ''
+        return '<strong class="highlight ' + word + '">' + word + '</strong>'
 
     def _tr_repl(self, word):
         out = ''
@@ -443,16 +442,17 @@ class PageFormatter:
             self.tr_cnt = 0
             out = '</p><table><tbody>\n'
         self.tr_cnt += 1
-        return out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '"><td>'
+        out = out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '">'
+        return out + ['<td>', '<th>'][word.strip() == '||=']
 
-    def _tre_repl(self, word):
+    def _td_repl(self, word):
         if self.in_table:
-            return '</td></tr>'
+            return ['</td><td>', '</th><th>'][word.strip() == '||=']
         return ''
 
-    def _td_repl(self, word):
+    def _tre_repl(self, word):
         if self.in_table:
-            return '</td><td>'
+            return ['</td></tr>', '</th></tr>'][word.strip() == '||=']
         return ''
 
     def _indent_level(self):
@@ -490,12 +490,13 @@ 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
-            + r"(?P<emph>'{2,3})"
+            + r"(?P<b>\*\*|'''|//|''|##|``|__|\^\^|,,)"
             + r"|(?P<tit>\={2,6})"
+            + r"|(?P<br>\\\\)"
             + r"|(?P<rule>^-{3,})"
             + r"|(?P<html><(/|)(div|span|iframe)[^<>]*>)"
             + r"|(?P<ent>[<>&])"
@@ -504,23 +505,23 @@ class PageFormatter:
             # Links
             + r"|(?P<img>\b[a-zA-Z0-9_-]+\.(png|gif|jpg|jpeg|bmp))"
             + r"|(?P<word>\b(?:[A-Z][a-z]+){2,}\b)"
-            + r"|(?P<hurl>\[\[\S+[^\]]*\]\])"
+            + r"|(?P<hurl>\[\[([^ \t\n\r\f\v\|]+)(?:\s*\|\s*([^\]]+)|)\]\])"
             + r"|(?P<url>(http|https|ftp|mailto)\:[^\s'\"]+\S)"
             + r"|(?P<email>[-\w._+]+\@[\w.-]+)"
 
             # Lists, divs, spans
-            + r"|(?P<li>^\s+\*)"
-            + r"|(?P<pre>(\{\{\{|\s*\}\}\}))"
-            + r"|(?P<var>(\{\{|\}\}))"
+            + r"|(?P<li>^\s+[\*#] +)"
+            + r"|(?P<pre>\{\{\{|\s*\}\}\})"
 
             # Tables
-            + r"|(?P<tr>^\s*\|\|\s*)"
-            + r"|(?P<tre>\s*\|\|\s*$)"
-            + r"|(?P<td>\s*\|\|\s*)"
+            + r"|(?P<tr>^\s*\|\|(=|)\s*)"
+            + r"|(?P<tre>\s*\|\|(=|)\s*$)"
+            + r"|(?P<td>\s*\|\|(=|)\s*)"
             + r")")
         pre_re = re.compile(
             r"(?:"
             + r"(?P<pre>\s*\}\}\})"
+            + r"|(?P<ent>[<>&])"
             + r")")
         blank_re = re.compile(r"^\s*$")
         indent_re = re.compile(r"^\s*")
@@ -559,18 +560,16 @@ class Page:
         self.page_name = page_name
         self.msg = ''
         self.msg_type = 'error'
-        self.attrs = {}
 
     def split_title(self):
-        # look for the end of words and the start of a new word,
-        # and insert a space there
+        # look for the end of words and the start of a new word and insert a space there
         return re.sub('([a-z])([A-Z])', r'\1 \2', self.page_name)
 
     def _text_filename(self):
-        return path.join(text_dir, self.page_name)
+        return path.join(data_dir, self.page_name)
 
     def _tmp_filename(self):
-        return path.join(text_dir, ('#' + self.page_name + '.' + `os.getpid()` + '#'))
+        return path.join(data_dir, ('#' + self.page_name.replace('/','_') + '.' + `os.getpid()` + '#'))
 
     def exists(self):
         try:
@@ -579,8 +578,7 @@ class Page:
         except OSError, er:
             if er.errno == errno.ENOENT:
                 return False
-            else:
-                raise er
+            raise er
 
     def link_to(self):
         word = self.page_name
@@ -591,15 +589,16 @@ class Page:
 
     def get_raw_body(self):
         try:
-            return open(self._text_filename(), 'rt').read()
+            return open(self._text_filename(), 'rb').read()
         except IOError, er:
             if er.errno == errno.ENOENT:
                 return '' # just doesn't exist, use default
             raise er
 
     def get_attrs(self):
-        if self.attrs:
+        if self.__dict__.has_key('attrs'):
             return self.attrs
+        self.attrs = {}
         try:
             file = open(self._text_filename(), 'rt')
             attr_re = re.compile(r"^#(\S*)(.*)$")
@@ -614,18 +613,17 @@ class Page:
                 raise er
         return self.attrs
 
+    def get_attr(self, name, default):
+        return self.get_attrs().get(name, default)
+
     def can(self, action, default=True):
-        attrs = self.get_attrs()
         try:
-            # SomeUser:read,write All:read
-            acl = attrs["acl"]
+            #acl SomeUser:read,write All:read
+            acl = self.get_attr("acl", None)
             for rule in acl.split():
                 (user,perms) = rule.split(':')
                 if user == remote_user() or user == "All":
-                    if action in perms.split(','):
-                        return True
-                    else:
-                        return False
+                    return action in perms.split(',')
             return False
         except Exception, er:
             pass
@@ -641,6 +639,11 @@ class Page:
         page_name = None
         if self.can_write():
             page_name = self.page_name
+
+        #css foo.css bar.css
+        global css_url
+        css_url = css_url + self.get_attr("css", "").split()
+
         send_title(page_name, self.split_title(), msg=self.msg, msg_type=self.msg_type)
         if self.can_read():
             PageFormatter(self.get_raw_body()).print_html()
@@ -668,9 +671,9 @@ class Page:
             + ' for ' + cgi.escape(remote_user())
             + ' from ' + cgi.escape(get_hostname(remote_host()))
             + '</b></p>')
-        print '<div class="editor"><form method="post" action="%s/%s">' % (script_name(), self.page_name)
-        print '<input type="hidden" name="savepage" value="%s">' % (self.page_name)
-        print """<textarea wrap="virtual" id="editor" name="savetext" rows="17" cols="80">%s</textarea>""" % (preview or self.get_raw_body())
+        print '<div class="editor"><form method="post" action="%s">' % relative_url(self.page_name)
+        print '<input type="hidden" name="edit" value="%s">' % (self.page_name)
+        print """<textarea wrap="off" spellcheck="true" id="editor" name="savetext" rows="17" cols="100">%s</textarea>""" % (preview or self.get_raw_body())
         print """
             <br />
             <input type="submit" name="save" value="Save" />
@@ -686,12 +689,13 @@ class Page:
             print "</div>"
         send_footer(self.page_name)
 
-    def send_raw(self):
+    def send_raw(self, mimetype='text/plain'):
         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()
+        body = self.get_raw_body()
+        emit_header(mimetype)
+        print body
 
     def _write_file(self, text):
         tmp_filename = self._tmp_filename()
@@ -716,7 +720,7 @@ class Page:
         if post_edit_hook:
             # FIXME: what's the std way to perform shell quoting in python?
             cmd = ( post_edit_hook
-                + " '" + text_dir + '/' + self.page_name
+                + " '" + data_dir + '/' + self.page_name
                 + "' '" + remote_user()
                 + "' '" + remote_host() + "'"
             )
@@ -732,24 +736,16 @@ class Page:
             self.msg = '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(text_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 }
+    handlers = {
+        'fullsearch':  do_fullsearch,
+        'titlesearch': do_titlesearch,
+        'edit':        do_edit,
+        'raw':         do_raw,
+    }
 
     for cmd in handlers.keys():
         if form.has_key(cmd):
@@ -767,9 +763,9 @@ try:
                 Page(query).send_page()
             elif img_re.match(query):
                 #FIXME: use correct mime type
-                send_verbatim(query, 'image/jpeg')
+                Page(query).send_raw(mimetype='image/jpeg')
             else:
-                send_verbatim(query)
+                Page(query).send_raw(mimetype='application/octet-stream')
         else:
             print "Status: 404 Not Found"
             send_title(None, msg='Can\'t work out query: ' + query)