-#! /usr/bin/env python
-"""Quick-quick implementation of WikiWikiWeb in Python
-"""
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
#
-# Copyright (C) 1999, 2000 Martin Pool <mbp@humbug.org.au>
-# This version includes additional changes by Gerardo Poggiali (2002)
-# This version includes additional changes by Bernardo Innocenti (2007)
+# Copyright 1999, 2000 Martin Pool <mbp@humbug.org.au>
+# Copyright 2002 Gerardo Poggiali
+# Copyright 2007, 2008, 2009 Bernie 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
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
-__version__ = '$Revision: 1.63+gerry+bernie $'[11:-2];
+__version__ = '$Id$'[4:12]
-import cgi, sys, string, os, re, errno, time, stat
-from os import path, environ
+from time import clock
+start_time = clock()
+
+import cgi, sys, os, re, errno, stat
# Regular expression defining a WikiWord
# (but this definition is also assumed in other places)
-file_re = re.compile(r"^\b([A-Za-z0-9_\.\-]+)\b$")
-word_re = re.compile(r"^\b([A-Z][a-z]+){2,}\b$")
-img_re = re.compile(r"^.*\.(png|gif|jpg|jpeg)$", re.IGNORECASE)
+word_re = re.compile(r"^\b((([A-Z][a-z0-9]+){2,}/)*([A-Z][a-z0-9]+){2,})\b$")
+# FIXME: we accept stuff like foo/../bar and we shouldn't
+file_re = re.compile(r"^\b([A-Za-z0-9_\-][A-Za-z0-9_\.\-/]*)\b$")
+img_re = re.compile(r"^.*\.(png|gif|jpg|jpeg|bmp|ico|ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt)$", re.IGNORECASE)
+video_re = re.compile(r"^.*\.(ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt)$", re.IGNORECASE)
url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$")
+link_re = re.compile(r"(?:\[\[|{{)([^\s\|]+)(?:\s*\|\s*([^\]]+)|)(?:\]\]|}})")
+ext_re = re.compile(r"\.([^\./]+)$")
title_done = False
-
# CGI stuff ---------------------------------------------------------
-
def script_name():
- return environ.get('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()
def remote_user():
- return environ.get('REMOTE_USER', 'AnonymousCoward')
+ user = os.environ.get('REMOTE_USER', '')
+ if user is None or user == '' or user == 'anonymous':
+ user = 'AnonymousCoward'
+ return user
def remote_host():
- return environ.get('REMOTE_ADDR', '')
+ return os.environ.get('REMOTE_ADDR', '')
def get_hostname(addr):
try:
from socket import gethostbyaddr
return gethostbyaddr(addr)[0] + ' (' + addr + ')'
- except:
- return addr;
+ except Exception:
+ return addr
-# Formatting stuff --------------------------------------------------
+def is_external_url(pathname):
+ return (url_re.match(pathname) or pathname.startswith('/'))
-def emit_header(type="text/html"):
- print "Content-type: " + type + "; charset=utf-8"
- print
+def relative_url(pathname, privileged=False):
+ if not is_external_url(pathname):
+ if privileged:
+ url = privileged_path()
+ else:
+ url = script_name()
+ pathname = url + '/' + pathname
+ return pathname
-def send_guru(msg, msg_type):
- if msg is None or len(msg) == 0: return
+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"
+
+def sendfile(dest_file, src_file):
+ """Efficiently copy file data between file descriptors"""
+ while 1:
+ data = src_file.read(65536)
+ if not data: break
+ dest_file.write(data)
+
+def send_guru(msg_text, msg_type):
+ if not msg_text: return
print '<pre id="guru" onclick="this.style.display = \'none\'" class="' + msg_type + '">'
if msg_type == 'error':
print ' Software Failure. Press left mouse button to continue.\n'
- print msg
+ print msg_text
if msg_type == 'error':
- print ' Guru Meditation #DEADBEEF.ABADC0DE'
+ print '\n Guru Meditation #DEADBEEF.ABADC0DE'
print '</pre>'
- # FIXME: This simple JS snippet is harder to pass than ACID 3.0
- print """
- <script language="JavaScript" type="text/javascript">
- var guru = document.getElementById('guru');
- // Firefox 2.0 doesn't take border-color, but returns border-top-color fine
- var color = document.defaultView.getComputedStyle(guru,null).getPropertyValue('border-top-color');
-
- function guruOn() {
- guru.style.setProperty('border-color', color, '');
- setTimeout('guruOff()', 1000);
- }
- function guruOff() {
- guru.style.setProperty('border-color', '#000000', '');
- setTimeout('guruOn()', 1000);
- }
- // Safari 2.0 returns this rgba crap
- // Konqueror 3.5.6 doesn't seem to support computed properties
- if (color && color != 'rgba(0, 0, 0, 0)') {
- //window.alert("enabled! color='" + color + "'");
- guruOn();
- }
- </script>"""
+ try:
+ sendfile(sys.stdout, open('gurumeditation.js', 'rb'))
+ except IOError, err:
+ pass
-def send_title(name, text="Limbo", msg=None, msg_type='error'):
+def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=False):
global title_done
if title_done: return
# Head
emit_header()
- print """<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
- "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
-<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
-"""
+ print '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'
+ print ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">'
+ print '<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">'
+
print "<head><title>%s: %s</title>" % (site_name, text)
- print ' <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />'
+ print ' <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8" />'
if not name:
print ' <meta name="robots" content="noindex,nofollow" />'
- if css_url:
- print ' <link rel="stylesheet" type="text/css" href="%s" />' % css_url
+
+ for meta in meta_urls:
+ http_equiv, content = meta
+ print ' <meta http-equiv="%s" content="%s" />' % (http_equiv, relative_url(content))
+
+ for link in link_urls:
+ rel, href = link
+ print ' <link rel="%s" href="%s" />' % (rel, relative_url(href))
+
+ if name and writable and privileged_url is not None:
+ print ' <link rel="alternate" type="application/x-wiki" title="Edit this page" href="%s" />' \
+ % (privileged_path() + '?edit=' + name)
+
+ if history_url is not None:
+ print ' <link rel="alternate" type="application/rss+xml" title="RSS" href="%s" />' \
+ % relative_url(history_url + '?a=rss')
+
print '</head>'
# Body
- if name and privileged_url is not None:
+ if name and writable and privileged_url is not None:
print '<body ondblclick="location.href=\'' + privileged_path() + '?edit=' + name + '\'">'
else:
print '<body>'
- send_guru(msg, msg_type)
+ title_done = True
+ send_guru(msg_text, msg_type)
# Navbar
- print '<div class="navigator">'
- print ' <b>' + site_name + ': ',
+ print '<div class="nav">'
+ print link_tag('FrontPage', relative_url(site_icon or 'Home'), 'navlink')
if name:
- print link_tag('?fullsearch=' + name, text) + '</b> '
+ print ' <b>' + link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
else:
- print text + '</b> '
- print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink')
+ print ' <b>' + text + '</b> '
print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
- print ' | <a href="/wikigit/wiki.git" class="navlink">Recent Changes</a>'
+ if 'history_url' in globals():
+ print ' | <a href="' + relative_url(history_url) + '" class="navlink">Recent Changes</a>'
+ if name:
+ print ' | <a href="' + relative_url(history_url + '?a=history;f=' + name) + '" class="navlink">Page History</a>'
if name:
- print ' | <a href="/wikigit/wiki.git?a=history;f=' + name + '" class="navlink">Page History</a>'
print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
if privileged_url is not None:
- print ' | ' + link_tag('?edit=' + name, 'Edit Page', 'navlink', authentication=True)
+ if writable:
+ print ' | ' + link_tag('?edit=' + name, 'Edit', 'navlink', privileged=True)
+ else:
+ print ' | ' + link_tag(name, 'Login', 'navlink', privileged=True)
+
else:
print ' | <i>Immutable Page</i>'
user = remote_user()
if user != 'AnonymousCoward':
- print ' | <i>logged in as <b>' + cgi.escape(user) + '</b></i>'
+ print ' | <span class="login"><i><b>' + link_tag('User/' + user, user) + '</b></i></span>'
- print '</div>'
+ print '<hr /></div>'
- title_done = True
+def send_httperror(status="403 Not Found", query=""):
+ print "Status: %s" % status
+ send_title(None, msg_text=("%s: on query '%s'" % (status, query)))
+ send_footer()
-def link_tag(params, text=None, ss_class=None, authentication=False):
+def link_tag(params, text=None, link_class=None, privileged=False):
if text is None:
text = params # default
- classattr = ''
- if ss_class:
- classattr += 'class="%s" ' % ss_class
- # Prevent crawlers from following links to generated pages
- # and links added by potential spammers
- if ss_class == 'external' or ss_class == 'navlink':
- classattr += 'rel="nofollow" '
- if authentication:
- path = privileged_path()
+ elif img_re.match(text):
+ text = '<img border="0" src="' + text + '" />'
+
+ if not link_class:
+ if is_external_url(params):
+ link_class = 'external'
+ elif file_re.match(params) and Page(params).exists():
+ link_class = 'wikilink'
+ else:
+ params = nonexist_pfx + params
+ link_class = 'nonexistent'
+
+ classattr = 'class="%s" ' % link_class
+ # Prevent crawlers from following links potentially added by spammers or to generated pages
+ if link_class == 'external' or link_class == 'navlink':
+ classattr += 'rel="nofollow"'
+
+ return '<a %shref="%s">%s</a>' % (classattr, relative_url(params, privileged=privileged), text)
+
+def link_inline(name, descr=None, args=''):
+ if not descr: descr = name
+ url = relative_url(name)
+ if video_re.match(name):
+ return '<video src="%s">Your browser does not support the HTML5 video tag</video>' % url
+ elif img_re.match(name):
+ return '<a href="%s"><img border="0" src="%s" alt="%s" /></a>' % (url, url + args, descr)
+ elif file_re.match(name) and not ext_re.search(name): # FIXME: this guesses a wiki page
+ return Page(name).send_naked()
else:
- path = script_name()
- return '<a %shref="%s/%s">%s</a>' % (classattr, path, params, text)
+ return '<iframe width="100%%" scrolling="auto" frameborder="0" src="%s"><a href="%s">%s</a></iframe>' \
+ % (url, url, name)
# 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)
print "<ul>"
for (count, page_name) in hits:
- print '<li><p>' + Page(page_name).link_to()
+ print '<li><p>' + link_tag(page_name)
print ' . . . . ' + `count`
- print ['match', 'matches'][count <> 1]
+ print ['match', 'matches'][count != 1]
print '</p></li>'
print "</ul>"
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 + '"')
print "<ul>"
for filename in hits:
- print '<li><p>' + Page(filename).link_to() + "</p></li>"
+ print '<li><p>' + link_tag(filename) + "</p></li>"
print "</ul>"
print_search_stats(len(hits), len(all_pages))
-
def print_search_stats(hits, searched):
print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
-def do_edit(pagename):
- Page(pagename).send_editor()
+def handle_raw(pagename):
+ if not file_re.match(pagename):
+ send_httperror("403 Forbidden", pagename)
+ return
-def do_raw(pagename):
Page(pagename).send_raw()
-def do_savepage(pagename):
- if privileged_url is None:
- raise 'editing disallowed for ' + pagename
+def handle_edit(pagename):
+ if not file_re.match(pagename):
+ send_httperror("403 Forbidden", pagename)
+ return
- global form
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(), form['changelog'].value)
+ else:
+ pg.save(form['savetext'].value.replace('\r\n', '\n'), form['changelog'].value)
+ pg.format()
elif 'cancel' in form:
- pg.msg = 'Editing cancelled'
+ 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 = '<p><center>'
- links = map(lambda ch: '<a href="#%s">%s</a>' % (ch, ch),
- string.lowercase)
- s = s + string.join(links, ' | ')
- s = s + '</center></p>'
- return s
+ links = map(lambda ch: '<a href="#%s">%s</a>' % (ch, ch), 'abcdefghijklmnopqrstuvwxyz')
+ return '<p><center>'+ ' | '.join(links) + '</center></p>'
+def page_list(dirname = None, re = word_re):
+ return sorted(filter(re.match, os.listdir(dirname or data_dir)))
-def page_list():
- return filter(word_re.match, os.listdir(text_dir))
-
-
-def send_footer(name, mod_string=None):
- if debug_cgi:
+def send_footer(mod_string=None):
+ if globals().get('debug_cgi', False):
cgi.print_arguments()
- cgi.print_form(cgi.FieldStorage())
+ cgi.print_form(form)
cgi.print_environ()
- print '<div class="footer">'
+ print '''
+<div id="footer"><hr />
+<p class="copyright">
+<a rel="license" href="http://creativecommons.org/licenses/by-sa/3.0/"><img class="license" alt="Creative Commons License" src="%s" /></a>
+<span class="benchmark">generated in %0.3fs</span> by <a href="http://www.codewiz.org/wiki/GeekiGeeki">GeekiGeeki</a> version %s
+</p>
+''' % (relative_url('cc-by-sa.png'), clock() - start_time, __version__)
if mod_string:
- print "last modified %s" % mod_string
+ print '<p class="modified">last modified %s</p>' % mod_string
print '</div></body></html>'
-
-# ----------------------------------------------------------
-# Macros
-def _macro_TitleSearch():
- return _macro_search("titlesearch")
-
-def _macro_FullSearch():
- return _macro_search("fullsearch")
-
-def _macro_search(type):
- if form.has_key('value'):
- default = form["value"].value
- else:
- default = ''
- return """<form method="get"><input name="%s" size="30" value="%s"><input type="submit" value="Go" /></form>""" % (type, default)
-
-def _macro_WordIndex():
- 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 + '; <a name="%s"><h3>%s</h3></a>' % (letter, letter)
- last_letter = letter
-
- s = s + '<b>%s</b><ul>' % word
- links = map[word]
- links.sort()
- last_page = None
- for name in links:
- if name == last_page: continue
- s = s + '<li>' + Page(name).link_to()
- s = s + '</ul>'
- return s
-
-
-def _macro_TitleIndex():
- 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 + '<a name="%s"><h3>%s</h3></a>' % (letter, letter)
- current_letter = letter
- else:
- s = s + '<br />'
- 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
"""
def __init__(self, raw):
self.raw = raw
- self.is_em = self.is_b = 0
self.h_level = 0
- self.list_indents = []
- self.in_pre = 0
- self.in_var = 0
+ self.in_pre = self.in_html = self.in_table = self.in_li = False
self.in_header = True
+ self.list_indents = []
+ self.tr_cnt = 0
+ self.styles = {
+ #wiki html enabled?
+ "//": ["em", False],
+ "**": ["b", False],
+ "##": ["tt", False],
+ "__": ["u", False],
+ "^^": ["sup", False],
+ ",,": ["sub", False],
+ "''": ["em", False], # LEGACY
+ "'''": ["b", False], # LEGACY
+ "``": ["tt", False], # LEGACY
+ }
- def _emph_repl(self, word):
- if len(word) == 3:
- self.is_b = not self.is_b
- return ['</b>', '<b>'][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:
- 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
- result = "<h%d>" % self.h_level
- return result;
-
- def _rule_repl(self, word):
- s = self._undent()
- if len(word) <= 3:
- s = s + "\n<hr size='1' noshade=\"noshade\" />\n"
- else:
- s = s + "\n<hr size='%d' noshade=\"noshade\" />\n" % (len(word) - 2 )
- return s
+ 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 _word_repl(self, word):
- return Page(word).link_to()
+ def _br_repl(self, word):
+ return '<br />'
- def _img_repl(self, word):
- return '<img border="0" src="%s/%s" />' % (script_name(), word)
+ def _rule_repl(self, word):
+ return self._undent() + '\n<hr size="%d" noshade="noshade" />\n' % (len(word) - 2)
- def _url_repl(self, word):
- if img_re.match(word):
- return '<img border="0" src="%s" />' % word
+ def _macro_repl(self, word):
+ m = re.compile("\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>").match(word)
+ name = m.group(1)
+ 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:
+ try:
+ execfile("macros/" + name + ".py", globals())
+ except IOError, err:
+ if err.errno == errno.ENOENT: pass
+ macro = globals().get('_macro_' + name)
+ if macro:
+ return macro(argv)
else:
- return '<a href="%s" rel="nofollow" class="external">%s</a>' % (word, word)
+ msg = '<<' + '|'.join(argv) + '>>'
+ if not self.in_html:
+ msg = '<strong class="error">' + msg + '</strong>'
+ return msg
def _hurl_repl(self, word):
- m = re.compile("\[\[(\S+)\ (.+)\]\]").match(word)
- anchor = m.group(1)
- descr = m.group(2)
- if img_re.match(anchor):
- return '<img border="0" src="%s" alt="%s" />' % (anchor, descr)
- elif url_re.match(anchor):
- return '<a href="%s" rel="nofollow" class="external">%s</a>' % (anchor, descr)
- elif anchor.startswith('/'):
- return '<a href="%s">%s</a>' % (anchor, descr)
+ m = link_re.match(word)
+ return link_tag(m.group(1), m.group(2))
+
+ def _inl_repl(self, word):
+ (name, descr) = link_re.match(word).groups()
+
+ if descr:
+ argv = descr.split('|')
+ descr = argv.pop(0)
+ args = ''
+ if argv:
+ args = '?' + '&'.join(argv)
+
+ # The "extthumb" nonsense works around a limitation of the HTML block model
+ return '<div class="extthumb"><div class="thumb">' \
+ + link_inline(name, descr, args) \
+ + '<div class="caption">' + descr + '</div></div></div>'
else:
- return link_tag(anchor, descr)
+ return link_inline(name, name)
- def _email_repl(self, word):
- return '<a href="mailto:%s">%s</a>' % (word, word)
+ def _html_repl(self, word):
+ 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:
+ return s; # Pass through
return {'&': '&',
'<': '<',
'>': '>'}[s]
+ def _img_repl(self, word): # LEGACY
+ return self._inl_repl('{{' + word + '}}')
- def _li_repl(self, match):
- return '<li>'
+ def _word_repl(self, word): # LEGACY
+ if self.in_html: return word # pass through
+ return link_tag(word)
+
+ def _url_repl(self, word): # LEGACY
+ if self.in_html: return word # pass through
+ return link_tag(word)
+ def _email_repl(self, word): # LEGACY
+ if self.in_html: return word # pass through
+ return '<a href="mailto:%s">%s</a>' % (word, word)
+
+ def _li_repl(self, match):
+ 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:
- self.in_pre = 1
+ self.in_pre = True
return '<pre>'
elif self.in_pre:
- self.in_pre = 0
+ self.in_pre = False
return '</pre>'
- else:
- return ''
-
- def _var_repl(self, word):
- if word == '{{' and not self.in_var:
- self.in_var = 1
- return '<code>'
- elif self.in_var:
- self.in_var = 0
- return '</code>'
- else:
- return ''
- def _macro_repl(self, word):
- macro_name = word[2:-2]
- # TODO: Somehow get the default value into the search field
- return apply(globals()['_macro_' + macro_name], ())
+ return ''
+
+ def _hi_repl(self, word):
+ return '<strong class="highlight ' + word + '">' + word + '</strong>'
+
+ def _tr_repl(self, word):
+ out = ''
+ if not self.in_table:
+ self.in_table = True
+ self.tr_cnt = 0
+ out = '</p><table><tbody>\n'
+ self.tr_cnt += 1
+ out = out + '<tr class="' + ['even', 'odd'][self.tr_cnt % 2] + '">'
+ return out + ['<td>', '<th>'][word.strip() == '||=']
+
+ def _td_repl(self, word):
+ if self.in_table:
+ return ['</td><td>', '</th><th>'][word.strip() == '||=']
+ return ''
+
+ def _tre_repl(self, word):
+ if self.in_table:
+ return ['</td></tr>', '</th></tr>'][word.strip() == '||=']
+ return ''
def _indent_level(self):
return len(self.list_indents) and self.list_indents[-1]
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)
return res
def replace(self, match):
- for type, hit in match.groupdict().items():
+ for rule, hit in match.groupdict().items():
if hit:
- return apply(getattr(self, '_' + type + '_repl'), (hit,))
+ return getattr(self, '_' + rule + '_repl')(hit)
else:
- raise "Can't handle match " + `match`
+ raise "Can't handle match " + repr(match)
def print_html(self):
- print "<div class='wiki'><p>"
-
- # For each line, we scan through looking for magic
- # strings, outputting verbatim any intervening text
- scan_re = re.compile(
- r"(?:"
- + r"(?P<emph>'{2,3})"
- + r"|(?P<tit>\={2,6})"
- + r"|(?P<ent>[<>&])"
- + 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<rule>^-{3,})"
- + r"|(?P<hurl>\[\[\S+\s+.+\]\])"
- + r"|(?P<url>(http|ftp|nntp|news|mailto)\:[^\s'\"]+\S)"
- + r"|(?P<email>[-\w._+]+\@[\w.-]+)"
- + r"|(?P<li>^\s+\*)"
- + r"|(?P<pre>(\{\{\{|\s*\}\}\}))"
- + r"|(?P<var>(\{\{|\}\}))"
- + r"|(?P<macro>\[\[(TitleSearch|FullSearch|WordIndex|TitleIndex)\]\])"
- + r")")
- pre_re = re.compile(
- r"(?:"
- + r"(?P<pre>\s*\}\}\})"
- + r")")
- blank_re = re.compile("^\s*$")
- indent_re = re.compile("^\s*")
- eol_re = re.compile(r'\r?\n')
- raw = string.expandtabs(self.raw)
- for line in eol_re.split(raw):
- # Skip ACLs
+ print '<div class="wiki"><p>'
+
+ 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
+ | (?P<macro> \<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)
+ | (?P<hurl> \[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])
+
+ # Inline HTML
+ | (?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)
+ | (?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
+ | (?P<li> ^\s+[\*\#]\s+)
+ | (?P<pre> \{\{\{|\s*\}\}\})
+ | (?P<inl> \{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})
+
+ # Tables
+ | (?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:
- 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:
- # XXX: Should we check these conditions in this order?
- if blank_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(self.line):
print '</p><p>'
- continue
- indent = indent_re.match(line)
- print self._indent_to(len(indent.group(0)))
- print re.sub(scan_re, self.replace, line)
+ else:
+ indent = indent_re.match(self.line)
+ print self._indent_to(len(indent.group(0))) ,
+ 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>"
+ print '</p></div>'
-# ----------------------------------------------------------
class Page:
def __init__(self, page_name):
self.page_name = page_name
- self.msg = ''
+ self.msg_text = ''
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)
+ def _filename(self):
+ return os.path.join(data_dir, self.page_name)
def _tmp_filename(self):
- return path.join(text_dir, ('#' + self.page_name + '.' + `os.getpid()` + '#'))
+ return os.path.join(data_dir, ('#' + self.page_name.replace('/','_') + '.' + `os.getpid()` + '#'))
def exists(self):
try:
- os.stat(self._text_filename())
- return 1
- except OSError, er:
- if er.errno == errno.ENOENT:
- return 0
- else:
- raise er
-
- def link_to(self):
- word = self.page_name
- if self.exists():
- return link_tag(word, word, 'wikilink')
- else:
- if nonexist_qm:
- return link_tag(word, '?', 'nonexistent') + word
- else:
- return link_tag(word, word, 'nonexistent')
-
-
- def get_raw_body(self):
+ os.stat(self._filename())
+ return True
+ except OSError, err:
+ if err.errno == errno.ENOENT:
+ return False
+ raise err
+
+ def get_raw_body(self, default=None):
try:
- return open(self._text_filename(), 'rt').read()
- except IOError, er:
- if er.errno == errno.ENOENT:
- # just doesn't exist, use default
- return 'Describe %s here.' % self.page_name
+ return open(self._filename(), 'rb').read()
+ except IOError, err:
+ if err.errno == errno.ENOENT:
+ if default is None:
+ default = '//[[?edit=%s|Describe %s]]//' % (self.page_name, self.page_name)
+ return default
+ if err.errno == errno.EISDIR:
+ return self.format_dir()
+ raise err
+
+ def format_dir(self):
+ out = '== '
+ pathname = ''
+ for dirname in self.page_name.split('/'):
+ pathname = (pathname + '/' + dirname) if pathname else dirname
+ out += '[[' + pathname + '|' + dirname + ']]/'
+ out += ' ==\n'
+
+ for filename in page_list(self._filename(), file_re):
+ if img_re.match(filename):
+ if image_maxwidth:
+ maxwidth_arg = '|maxwidth=' + str(image_maxwidth)
+ out += '{{' + self.page_name + '/' + filename + '|' + filename + maxwidth_arg + '}}\n'
else:
- raise er
+ out += ' * [[' + self.page_name + '/' + filename + ']]\n'
+ return out
- def get_attrs(self):
- if self.attrs:
- return self.attrs
- try:
- file = open(self._text_filename(), 'rt')
- attr_re = re.compile(r"^#(\S*)(.*)$")
- for line in file:
- m = attr_re.match(line)
- if not m:
- break
- self.attrs[m.group(1)] = m.group(2).strip()
- #print "bernie: attrs[" + m.group(1) + "] = " + m.group(2) + "<br>\n"
- except IOError, er:
- if er.errno != errno.ENOENT:
- raise er
- return self.attrs
-
- def can_edit(self):
- attrs = self.get_attrs()
+ def pragmas(self):
+ if not '_pragmas' in self.__dict__:
+ self._pragmas = {}
+ try:
+ f = open(self._filename(), 'rt')
+ attr_re = re.compile(r"^#(\S*)(.*)$")
+ for line in f:
+ m = attr_re.match(line)
+ if not m:
+ break
+ self._pragmas[m.group(1)] = m.group(2).strip()
+ #print "bernie: _pragmas[" + m.group(1) + "] = " + m.group(2) + "<br>\n"
+ except IOError, err:
+ if err.errno != errno.ENOENT and err.errno != errno.EISDIR:
+ raise err
+ return self._pragmas
+
+ def pragma(self, name, default):
+ return self.pragmas().get(name, default)
+
+ def can(self, action, default=True):
+ acl = None
try:
- # SomeUser:read,write All:read
- acl = attrs["acl"]
+ #acl SomeUser:read,write All:read
+ acl = self.pragma("acl", None)
for rule in acl.split():
- (user,perms) = acl.split(':')
+ (user, perms) = rule.split(':')
if user == remote_user() or user == "All":
- if 'write' in perms.split(','):
- return True
+ return action in perms.split(',')
return False
- except:
- pass
- return True
-
- def send_page(self):
- page_name = None
- if self.can_edit():
- page_name = self.page_name
- send_title(page_name, self.split_title(), msg=self.msg, msg_type=self.msg_type)
- PageFormatter(self.get_raw_body()).print_html()
- send_footer(page_name, self._last_modified())
+ except Exception:
+ if acl:
+ self.msg_text = 'Illegal acl line: ' + acl
+ return default
+
+ def can_write(self):
+ return self.can("write", True)
+
+ def can_read(self):
+ return self.can("read", True)
+
+ def send_naked(self):
+ if self.can_read():
+ WikiFormatter(self.get_raw_body()).print_html()
+ else:
+ send_guru("Read access denied by ACLs", "notice")
+
+ def format(self):
+ #css foo.css
+ value = self.pragma("css", None)
+ if value:
+ global link_urls
+ link_urls += [ [ "stylesheet", value ] ]
+
+ send_title(self.page_name, self.split_title(),
+ msg_text=self.msg_text, msg_type=self.msg_type, writable=self.can_write())
+ self.send_naked()
+ send_footer(self._last_modified())
def _last_modified(self):
- if not self.exists():
+ try:
+ from time import localtime, strftime
+ modtime = localtime(os.stat(self._filename())[stat.ST_MTIME])
+ except OSError, err:
+ if err.errno != errno.ENOENT:
+ raise err
return None
- from time import localtime, strftime
- modtime = localtime(os.stat(self._text_filename())[stat.ST_MTIME])
return strftime(datetime_fmt, modtime)
def send_editor(self, preview=None):
- send_title(None, 'Edit ' + self.split_title(), msg=self.msg, msg_type=self.msg_type)
+ send_title(None, 'Edit ' + self.split_title(), msg_text=self.msg_text, msg_type=self.msg_type)
+ if not self.can_write():
+ send_guru("Write access denied by ACLs", "error")
+ return
+
+ filename = ''
+ if 'file' in form:
+ filename = form['file'].value
print ('<p><b>Editing ' + self.page_name
+ ' 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 name="editform" method="post" enctype="multipart/form-data" action="%s">' % relative_url(self.page_name)
+ print '<input type="hidden" name="edit" value="%s">' % (self.page_name)
+ print '<input type="input" id="editor" name="changelog" value="Edit page %s" accesskey="c" /><br />' % (self.page_name)
+ 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(default=''))
+ print '<label for="file" accesskey="u">Or Upload a file:</label> <input type="file" name="file" value="%s" />' % filename
print """
<br />
- <input type="submit" name="save" value="Save" />
- <input type="submit" name="preview" value="Preview" />
+ <input type="submit" name="save" value="Save" accesskey="s">
+ <input type="submit" name="preview" value="Preview" accesskey="p" />
<input type="reset" value="Reset" />
<input type="submit" name="cancel" value="Cancel" />
<br />
- </form></div>"""
- print "<p>" + Page('EditingTips').link_to() + "</p>"
+ </form></div>
+ <script language="javascript">
+ <!--
+ document.editform.savetext.focus()
+ //-->
+ </script>
+ """
+ print "<p>" + link_tag('EditingTips') + "</p>"
if preview:
print "<div class='preview'>"
- PageFormatter(preview).print_html()
+ WikiFormatter(preview).print_html()
print "</div>"
- send_footer(self.page_name)
-
- def send_raw(self):
- emit_header("text/plain")
- print self.get_raw_body()
+ send_footer()
+
+ def send_raw(self, mimetype='text/plain', args=[]):
+ if not self.can_read():
+ send_title(None, msg_text='Read access denied by ACLs', msg_type='notice')
+ return
+
+ if 'maxwidth' in args:
+ import subprocess
+ emit_header(mimetype)
+ sys.stdout.flush()
+ subprocess.check_call(['gm', 'convert', self._filename(),
+ '-scale', args['maxwidth'].value + ' >', '-'])
+ else:
+ body = self.get_raw_body()
+ emit_header(mimetype)
+ print body
- 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)
- except OSError, er:
- if er.errno <> errno.ENOENT: raise er
- os.rename(tmp_filename, text)
-
- def save_text(self, newtext):
- self._write_file(newtext)
+ os.remove(name)
+ except OSError, err:
+ if err.errno != errno.ENOENT: raise err
+ os.rename(tmp_filename, name)
+
+ def save(self, newdata, changelog):
+ if not self.can_write():
+ self.msg_text = 'Write access denied by ACLs'
+ self.msg_type = 'error'
+ return
+
+ self._write_file(newdata)
rc = 0
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
- + "' '" + remote_user()
- + "' '" + remote_host() + "'"
- )
- out = os.popen(cmd)
- msg = out.read()
- rc = out.close()
+ import subprocess
+ cmd = [ post_edit_hook, data_dir + '/' + self.page_name, remote_user(), remote_host(), changelog]
+ child = subprocess.Popen(cmd, stdout=subprocess.PIPE, close_fds=True)
+ output = child.stdout.read()
+ rc = child.wait()
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. Command was:\n'%s'\n" % (rc, "' '".join(cmd))
+ if output:
+ self.msg_text += 'Output follows:\n' + output
else:
- self.msg = 'Thankyou 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'
-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:
- # Configuration values
- site_name = 'Codewiz'
-
- # set to None for read-only sites
- # leave empty ('') to allow anonymous edits
- # otherwise, set to a URL that requires authentication
- privileged_url = 'https://www.codewiz.org/~bernie/wiki'
-
- data_dir = '/home/bernie/public_html/wiki'
- text_dir = path.join(data_dir, 'text')
- css_url = '../wikidata/geekigeeki.css' # optional stylesheet link
- post_edit_hook = './post_edit_hook.sh'
- datetime_fmt = '%a %d %b %Y %I:%M %p'
- allow_edit = True # Is it possible to edit pages?
- show_hosts = True # show hostnames?
- nonexist_qm = False # show '?' for nonexistent?
- debug_cgi = False # Set to True for CGI var dump
-
- 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,))
+def main():
+ for cmd in form:
+ handler = globals().get('handle_' + cmd)
+ if handler:
+ handler(form[cmd].value)
break
else:
- path_info = environ.get('PATH_INFO', '')
- if len(path_info) and path_info[0] == '/':
- query = path_info[1:] or 'FrontPage'
- else:
- query = environ.get('QUERY_STRING', '') or 'FrontPage'
-
+ query = query_string()
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')
+ # FIMXE: this is all bullshit, MimeTypes bases its guess on the extension!
+ from mimetypes import MimeTypes
+ mimetype, encoding = MimeTypes().guess_type(query)
+ if mimetype:
+ Page(query).send_raw(mimetype=mimetype, args=form)
else:
- send_verbatim(query)
+ Page(query).format()
else:
- # TODO: return 404?
- send_title(None, msg='Can\'t work out query: ' + query)
-except:
+ send_httperror("403 Forbidden", query)
+
+try:
+ execfile("geekigeeki.conf.py")
+ form = cgi.FieldStorage()
+ main()
+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_footer(None)
+ send_title(None, msg_text=msg_text)
+ send_footer()
sys.stdout.flush()