X-Git-Url: https://codewiz.org/gitweb?a=blobdiff_plain;f=geekigeeki.py;h=5a401abde033bc06cd504fdb61c4e00fc41676e2;hb=62db44f1f908f7f3f0d9942f841b89115b8d6c67;hp=e7001abc60ce63edd26aa03c838c6208816b53ec;hpb=76a9e7542e0ed64a3054882db2cf4ae18e4f558e;p=geekigeeki.git
diff --git a/geekigeeki.py b/geekigeeki.py
index e7001ab..5a401ab 100755
--- a/geekigeeki.py
+++ b/geekigeeki.py
@@ -1,10 +1,9 @@
-#! /usr/bin/env python
-"""Quick-quick implementation of WikiWikiWeb in Python
-"""
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
#
-# Copyright (C) 1999, 2000 Martin Pool
-# This version includes additional changes by Gerardo Poggiali (2002)
-# This version includes additional changes by Bernardo Innocenti (2007)
+# Copyright 1999, 2000 Martin Pool
+# Copyright 2002 Gerardo Poggiali
+# Copyright 2007, 2008, 2009 Bernie Innocenti
#
# 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
@@ -19,135 +18,228 @@
# You should have received a copy of the GNU General Public License
# along with this program. If not, see .
-__version__ = '$Revision: 1.63+gerry+bernie $'[11:-2];
-
-import cgi, sys, string, os, re, errno, time, stat
-from os import path, environ
-from socket import gethostbyaddr
-from time import localtime, strftime
-
-# Regular expression defining a WikiWord
-# (but this definition is also assumed in other places)
-file_re = re.compile(r"^\b([A-Za-z0-9_\.\-]+)\b$")
-word_re = re.compile(r"^\b([A-Z][a-z]+){2,}\b$")
-img_re = re.compile(r"^.*\.(png|gif|jpg|jpeg)$", re.IGNORECASE)
-url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$")
+__version__ = '4.0-' + '$Id$'[4:11]
+from time import clock
+start_time = clock()
title_done = False
-def emit_header(type="text/html"):
- print "Content-type: " + type + "; charset=utf-8"
- print
+import cgi, sys, os, re, errno, stat
+
+image_ext = 'png|gif|jpg|jpeg|bmp|ico'
+video_ext = "ogg|ogv|oga" # Not supported by Firefox 3.5: mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt
+image_re = re.compile(r".*\.(" + image_ext + "|" + video_ext + ")", re.IGNORECASE)
+video_re = re.compile(r".*\.(" + video_ext + ")", re.IGNORECASE)
+# FIXME: we accept stuff like foo/../bar and we shouldn't
+file_re = re.compile(r"([A-Za-z0-9_\-][A-Za-z0-9_\.\-/]*)")
+url_re = re.compile(r"[a-z]{3,8}://[^\s'\"]+\S")
+ext_re = re.compile(r"\.([^\./]+)$")
+
+# 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()
-# Formatting stuff --------------------------------------------------
+def remote_user():
+ user = os.environ.get('REMOTE_USER', '')
+ if user is None or user == '' or user == 'anonymous':
+ user = 'AnonymousCoward'
+ return user
-def get_scriptname():
- return environ.get('SCRIPT_NAME', '')
+def remote_host():
+ 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
+
+def is_external_url(pathname):
+ return (url_re.match(pathname) or pathname.startswith('/'))
+
+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 cgi.escape(pathname, quote=True)
+
+def permalink(s):
+ return re.sub(' ', '-', re.sub('[^a-z0-9_ ]', '', s.lower()).strip())
+
+def humanlink(s):
+ return re.sub(r'(?:.*[/:]|)([^:/\.]+)(?:\.[^/:]+|)$', r'\1', s.replace('_', ' '))
+
+# Split arg lists like "blah| blah blah| width=100 | align = center",
+# return a list containing anonymous arguments and a map containing the named arguments
+def parse_args(s):
+ args = []
+ kwargs = {}
+ for arg in s.strip('<[{}]>').split('|'):
+ m = re.match('\s*(\w+)\s*=\s*(.+)\s*', arg)
+ if m is not None:
+ kwargs[m.group(1)] = m.group(2)
+ else:
+ args.append(arg.strip())
+ return (args, kwargs)
+
+def url_args(kvargs):
+ argv = []
+ for k, v in kvargs.items():
+ argv.append(k + '=' + v)
+ if argv:
+ return '?' + '&'.join(argv)
+ return ''
+
+# Formatting stuff --------------------------------------------------
+def emit_header(mime_type="text/html"):
+ print("Content-type: " + mime_type + "; charset=utf-8\n")
-def send_guru(msg, msg_type):
- if msg is None or len(msg) == 0: return
- print ''
+def send_guru(msg_text, msg_type):
+ if not msg_text: return
+ print('')
if msg_type == 'error':
- print ' Software Failure. Press left mouse button to continue.\n'
- print msg
+ print(' Software Failure. Press left mouse button to continue.\n')
+ print(msg_text)
if msg_type == 'error':
- print ' Guru Meditation #DEADBEEF.ABADC0DE'
- print ' '
- # FIXME: This simple JS code is harder to pass than ACID 3.0
- print """
- """
+ print '\n Guru Meditation #DEADBEEF.ABADC0DE'
+ print(' ' \
+ % relative_url('sys/GuruMeditation.js'))
-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 """
-
-"""
- print "%s: %s " % (site_name, text)
- print ' '
+ print('')
+ print('')
+
+ print("%s: %s " % (site_name, text))
+ print(' ')
if not name:
- print ' '
- if css_url:
- print ' ' % css_url
- print ''
+ print(' ')
+
+ for meta in meta_urls:
+ http_equiv, content = meta
+ print(' ' % (http_equiv, relative_url(content)))
+
+ for link in link_urls:
+ rel, href = link
+ print(' ' % (rel, relative_url(href)))
+
+ if name and writable and privileged_url is not None:
+ print(' ' \
+ % (privileged_path() + '?a=edit&q=' + name))
+
+ if history_url is not None:
+ print(' ' \
+ % relative_url(history_url + '?a=rss'))
+
+ print('')
# Body
- if name and allow_edit:
- print ''
+ if name and writable and privileged_url is not None:
+ print('')
else:
- print ''
+ print('')
- send_guru(msg, msg_type)
+ title_done = True
+ send_guru(msg_text, msg_type)
# Navbar
- print ''
- print '
' + site_name + ': '
+ print('')
+ print link_tag('FrontPage', site_icon or 'Home', cssclass='navlink')
if name:
- print link_tag('?fullsearch=' + name, text) + ' '
+ print('
' + link_tag('?fullsearch=' + name, text, cssclass='navlink') + ' ')
else:
- print text + ' '
- print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink')
- print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
- print ' |
Recent Changes '
+ print('
' + text + ' ')
+ print(' | ' + link_tag('FindPage', 'Find Page', cssclass='navlink'))
+ if 'history_url' in globals():
+ print(' |
Recent Changes ')
+ if name:
+ print(' |
Page History ')
+
if name:
- print ' |
Page History '
- print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')
- if allow_edit:
- print ' | ' + link_tag('?edit=' + name, 'Edit Page', 'navlink')
+ print(' | ' + link_tag(name + '?a=raw', 'Raw Text', cssclass='navlink'))
+ if privileged_url is not None:
+ if writable:
+ print(' | ' + link_tag('?a=edit&q=' + name, 'Edit', cssclass='navlink', privileged=True))
+ else:
+ print(' | ' + link_tag(name, 'Login', cssclass='navlink', privileged=True))
+
else:
- print ' |
Immutable Page '
- print '
'
+ print(' | Immutable Page ')
- title_done = True
+ user = remote_user()
+ if user != 'AnonymousCoward':
+ print(' | ' + link_tag('User/' + user, user) + ' ')
+ print(' ')
-def link_tag(params, text=None, ss_class=None):
+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(dest, text=None, privileged=False, **kvargs):
if text is None:
- text = params # default
- classattr = ''
- # 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 ss_class:
- classattr += 'class="%s" ' % ss_class
- return '%s ' % (classattr, get_scriptname(),
- params, text)
+ text = humanlink(dest)
+ elif image_re.match(text):
+ text = ' '
+
+ link_class = kvargs.get('class', kvargs.get('cssclass', None))
+ if not link_class:
+ if is_external_url(dest):
+ link_class = 'external'
+ elif file_re.match(dest) and Page(dest).exists():
+ link_class = 'wikilink'
+ else:
+ text = nonexist_pfx + text
+ link_class = 'nonexistent'
+
+ # Prevent crawlers from following links potentially added by spammers or to generated pages
+ nofollow = ''
+ if link_class == 'external' or link_class == 'navlink':
+ nofollow = 'rel="nofollow" '
+
+ return '%s ' % (link_class, nofollow, relative_url(dest, privileged=privileged), text)
+
+def link_inline(name, descr=None, kvargs={}):
+ if not descr: descr = humanlink(name)
+ url = relative_url(name)
+ if video_re.match(name):
+ return 'Your browser does not support the HTML5 video tag ' % url
+ elif image_re.match(name):
+ return ' ' % (url, url + url_args(kvargs), descr)
+ elif file_re.match(name) and not ext_re.search(name): # FIXME: this guesses a wiki page
+ return Page(name).send_naked()
+ else:
+ return '' \
+ % (url, url, name)
# Search ---------------------------------------------------
-def do_fullsearch(needle):
- send_title(None, 'Full text search for "%s"' % (needle))
+def print_search_stats(hits, searched):
+ print("%d hits out of %d pages searched.
" % (hits, searched))
+
+def handle_fullsearch(query, form):
+ needle = form['q'].value
+ send_title(None, 'Full text search for "' + needle + '"')
needle_re = re.compile(needle, re.IGNORECASE)
hits = []
@@ -163,146 +255,101 @@ def do_fullsearch(needle):
hits.sort()
hits.reverse()
- print "")
print_search_stats(len(hits), len(all_pages))
-
-def do_titlesearch(needle):
- # TODO: check needle is legal -- but probably we can just accept any RE
- send_title(None, "Title search for \"" + needle + '"')
+def handle_titlesearch(query, form):
+ needle = form['q'].value
+ send_title(None, 'Title search for "' + needle + '"')
needle_re = re.compile(needle, re.IGNORECASE)
all_pages = page_list()
- hits = filter(needle_re.search, all_pages)
+ hits = list(filter(needle_re.search, all_pages))
- print "")
print_search_stats(len(hits), len(all_pages))
+def handle_raw(pagename, form):
+ if not file_re.match(pagename):
+ send_httperror("403 Forbidden", pagename)
+ return
-def print_search_stats(hits, searched):
- print "%d hits out of %d pages searched.
" % (hits, searched)
-
-def do_edit(pagename):
- Page(pagename).send_editor()
-
-def do_raw(pagename):
Page(pagename).send_raw()
-def do_savepage(pagename):
- if not allow_edit:
- raise 'editing disallowed for ' + pagename
-
- 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()
+def handle_edit(pagename, form):
+ if not file_re.match(pagename):
+ send_httperror("403 Forbidden", pagename)
+ return
+
+ pg = Page(form['q'].value)
+ 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 handle_get(pagename, form):
+ if file_re.match(pagename):
+ # FIMXE: this is all bullshit, MimeTypes bases its guess on the extension!
+ from mimetypes import MimeTypes
+ mimetype, encoding = MimeTypes().guess_type(pagename)
+ if mimetype:
+ Page(pagename).send_raw(mimetype=mimetype, args=form)
+ else:
+ Page(pagename).format()
+ else:
+ send_httperror("403 Forbidden", pagename)
+# Used by macros/WordIndex and macros/TitleIndex
def make_index_key():
- s = '
'
- links = map(lambda ch: '%s ' % (ch, ch),
- string.lowercase)
- s = s + string.join(links, ' | ')
- s = s + '
'
- return s
-
-
-def page_list():
- return filter(word_re.match, os.listdir(text_dir))
-
-
-def send_footer(name, mod_string=None):
- print '')
-def _macro_search(type):
- if form.has_key('value'):
- default = form["value"].value
- else:
- default = ''
- return """""" % (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 + '; %s ' % (letter, letter)
- last_letter = letter
-
- s = s + '%s ' % word
- links = map[word]
- links.sort()
- last_page = None
- for name in links:
- if name == last_page: continue
- s = s + '' + Page(name).link_to()
- s = s + ' '
- 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 + '%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
@@ -310,100 +357,150 @@ class PageFormatter:
"""
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],
+ "--": ["del", 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 ['', ''][self.is_b]
- else:
- self.is_em = not self.is_em
- return ['', ''][self.is_em]
+ def _b_repl(self, word):
+ style = self.styles[word]
+ style[1] = not style[1]
+ return ['', '<'][style[1]] + style[0] + '>'
+
+ def _glyph_repl(self, word):
+ 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
- result = "" % self.h_level
- return result;
+ link = permalink(self.line)
+ result = '\n
¶ ' % (self.h_level, link, link)
+ return result
- def _rule_repl(self, word):
- s = self._undent()
- if len(word) <= 3:
- s = s + "\n \n"
- else:
- s = s + "\n \n" % (len(word) - 2 )
- return s
+ def _br_repl(self, word):
+ return ' '
- def _word_repl(self, word):
- return Page(word).link_to()
-
- def _img_repl(self, word):
- return ' ' % (get_scriptname(), word)
+ def _rule_repl(self, word):
+ return self._undent() + '\n \n' % (len(word) - 2)
- def _url_repl(self, word):
- if img_re.match(word):
- return ' ' % word
- else:
- return '%s ' % (word, word)
+ def _macro_repl(self, word):
+ try:
+ args, kwargs = parse_args(word)
+ macro = globals().get('_macro_' + args[0])
+ if not macro:
+ exec(open("macros/" + name + ".py").read(), globals())
+ macro = globals().get('_macro_' + name)
+ return macro(*args, **kwargs)
+ except Exception:
+ msg = cgi.escape(word)
+ if not self.in_html:
+ msg = '' + msg + ' '
+ 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 ' ' % (anchor, descr)
- elif url_re.match(anchor):
- return '%s ' % (anchor, descr)
- elif anchor.startswith('/'):
- return '%s ' % (anchor, descr)
+ args, kvargs = parse_args(word)
+ return link_tag(*args, **kvargs)
+
+ def _inl_repl(self, word):
+ args, kvargs = parse_args(word)
+ name = args.pop(0)
+ if len(args):
+ descr = args.pop(0)
+ # This double div nonsense works around a limitation of the HTML block model
+ return '' \
+ + '
' \
+ + link_inline(name, descr, kvargs) \
+ + '
' + descr + '
'
else:
- return link_tag(anchor, descr)
+ return link_inline(name, None, kvargs)
- def _email_repl(self, word):
- return '%s ' % (word, word)
+ def _html_repl(self, word):
+ if not self.in_html and word.startswith(''
+ 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 '
'
+ 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 '%s ' % (word, word)
+
+ def _li_repl(self, match):
+ if self.in_li:
+ return ' '
+ else:
+ self.in_li = True
+ return ' '
def _pre_repl(self, word):
if word == '{{{' and not self.in_pre:
- self.in_pre = 1
+ self.in_pre = True
return ''
elif self.in_pre:
- self.in_pre = 0
+ self.in_pre = False
return ' '
- else:
- return ''
-
- def _var_repl(self, word):
- if word == '{{' and not self.in_var:
- self.in_var = 1
- return ''
- elif self.in_var:
- self.in_var = 0
- return '
'
- 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 '' + word + ' '
+
+ def _tr_repl(self, word):
+ out = ''
+ if not self.in_table:
+ self.in_table = True
+ self.tr_cnt = 0
+ out = '\n'
+ self.tr_cnt += 1
+ out = out + ''
+ return out + ['', ' '][word.strip() == '||=']
+
+ def _td_repl(self, word):
+ if self.in_table:
+ return [' ', ' '][word.strip() == '||=']
+ return ''
+
+ def _tre_repl(self, word):
+ if self.in_table:
+ return [' ', ''][word.strip() == '||=']
+ return ''
def _indent_level(self):
return len(self.list_indents) and self.list_indents[-1]
@@ -414,6 +511,9 @@ 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)
@@ -429,280 +529,322 @@ class PageFormatter:
return res
def replace(self, match):
- for type, hit in match.groupdict().items():
+ for rule, hit in list(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 ""
-
- # For each line, we scan through looking for magic
- # strings, outputting verbatim any intervening text
- scan_re = re.compile(
- r"(?:"
- + r"(?P'{2,3})"
- + r"|(?P\={2,6})"
- + r"|(?P[<>&])"
- + r"|(?P \b[a-zA-Z0-9_-]+\.(png|gif|jpg|jpeg|bmp))"
- + r"|(?P\b(?:[A-Z][a-z]+){2,}\b)"
- + r"|(?P^-{3,})"
- + r"|(?P\[\[\S+\s+.+\]\])"
- + r"|(?P(http|ftp|nntp|news|mailto)\:[^\s'\"]+\S)"
- + r"|(?P[-\w._+]+\@[\w.-]+)"
- + r"|(?P^\s+\*)"
- + r"|(?P(\{\{\{|\s*\}\}\}))"
- + r"|(?P(\{\{|\}\}))"
- + r"|(?P\[\[(TitleSearch|FullSearch|WordIndex|TitleIndex)\]\])"
- + r")")
- pre_re = re.compile(
- r"(?:"
- + r"(?P\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('')
+
+ scan_re = re.compile(r"""(?:
+ # Styles and formatting ("--" must cling to a word to disambiguate it from the dash)
+ (?P \*\* | // | \#\# | __ | --\b | \b-- | \^\^ | ,, | ''' | '' | `` )
+ | (?P \={2,6})
+ | (?P \\\\)
+ | (?P ^-{3,})
+ | (?P \b( FIXME | TODO | DONE )\b )
+ | (?P --)
+
+ # Links
+ | (?P \<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)
+ | (?P \[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])
+
+ # Inline HTML
+ | (?P <(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])\b )
+ | (?P ( /\s*> | (br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])> ) )
+ | (?P [<>&] )
+
+ # Auto links (LEGACY)
+ | (?P \b[a-zA-Z0-9_/-]+\.(""" + image_ext + "|" + video_ext + r"""))
+ | (?P \b(?:[A-Z][a-z]+){2,}\b)
+ | (?P (http|https|ftp|mailto)\:[^\s'\"]+\S)
+ | (?P [-\w._+]+\@[\w.-]+)
+
+ # Lists, divs, spans and inline objects
+ | (?P ^\s+[\*\#]\s+)
+ | (?P \{\{\{|\s*\}\}\})
+ | (?P \{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})
+
+ # Tables
+ | (?P ^\s*\|\|(=|)\s*)
+ | (?P \s*\|\|(=|)\s*$)
+ | (?P \s*\|\|(=|)\s*)
+
+ # TODO: highlight search words (look at referrer)
+ )""", re.VERBOSE)
+ pre_re = re.compile("""(?:
+ (?P\s*\}\}\})
+ | (?P[<>&])"
+ )""", re.VERBOSE)
+ blank_re = re.compile(r"^\s*$")
+ 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
+ #3.0: for self.line in eol_re.split(str(self.raw.expandtabs(), 'utf-8')):
+ for self.line in eol_re.split(str(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):
- print ''
- continue
- indent = indent_re.match(line)
- print self._indent_to(len(indent.group(0)))
- print re.sub(scan_re, self.replace, line)
- if self.in_pre: print '
'
- print self._undent()
- print ""
-
-# ----------------------------------------------------------
+ if self.in_table and not tr_re.match(self.line):
+ self.in_table = False
+ print('
')
+
+ if blank_re.match(self.line):
+ print('
')
+ else:
+ indent = indent_re.match(self.line)
+ #3.0: print(self._indent_to(len(indent.group(0))), end=' ')
+ print(self._indent_to(len(indent.group(0))))
+ print(re.sub(scan_re, self.replace, self.line))
+
+ if self.in_pre: print('')
+ if self.in_table: print('
')
+ print(self._undent())
+ print('
')
+
class Page:
def __init__(self, page_name):
self.page_name = page_name
- self.remote_user = 'AnonymousCoward'
- self.remote_host = environ.get('REMOTE_ADDR', '')
- 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('/','_') + '.' + str(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 = '//[[?a=edit&q=%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.strip('/').split('/'):
+ pathname = (pathname + '/' + dirname) if pathname else dirname
+ out += '[[' + pathname + '|' + dirname + ']]/'
+ out += ' ==\n'
+
+ for filename in page_list(self._filename(), file_re):
+ if image_re.match(filename):
+ if image_maxwidth:
+ maxwidth_arg = ' | maxwidth=' + str(image_maxwidth)
+ out += '{{' + self.page_name + '/' + filename + ' | ' + humanlink(filename) + maxwidth_arg + ' | class=thumbleft}}\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*)(.*)$")
- for line in file:
- m = attr_re.match(line)
- if not m:
- break
- self.attrs[m.group(1)] = m.group(2).strip()
- #print "bernie: attrs[" + m.group(1) + "] = " + m.group(2) + " \n"
- except IOError, er:
- if er.errno != errno.ENOENT:
- raise er
- return self.attrs
-
- def can_edit(self):
- attrs = self.get_attrs()
+ def pragmas(self):
+ if not '_pragmas' in self.__dict__:
+ self._pragmas = {}
+ try:
+ file = open(self._filename(), 'rt')
+ attr_re = re.compile(r"^#(\S*)(.*)$")
+ for line in file:
+ 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) + " \n"
+ except IOError, err:
+ if err.errno != errno.ENOENT and err.errno != errno.EISDIR:
+ raise er
+ 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(':')
- if user == self.remote_user or user == "All":
- if 'write' in perms.split(','):
- return True
+ (user, perms) = rule.split(':')
+ if user == remote_user() or user == "All":
+ 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
- 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)
-
- print ('Editing ' + self.page_name
- + ' for ' + self.remote_user
- + ' from ' + get_hostname(self.remote_host)
- + '
')
- print '
+
+ """)
+ print("" + link_tag('EditingTips') + "
")
if preview:
- print ""
- PageFormatter(preview).print_html()
- print "
"
-
- def send_raw(self):
- emit_header("text/plain")
- print self.get_raw_body()
+ print("")
+ WikiFormatter(preview).print_html()
+ print("
")
+ 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
+ path = os.path.split(name)[0]
+ if not os.path.exists(path):
+ os.makedirs(path)
+ 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:
- cmd = ( post_edit_hook
- + ' ' + text_dir + '/' + self.page_name
- + ' ' + self.remote_user
- + ' ' + self.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)
-
-# ---------------------------------------------------------------
try:
- # Configuration values
- data_dir = '/home/bernie/public_html/wiki'
- text_dir = path.join(data_dir, 'text')
- allow_edit = True # Is it possible to edit pages?
- site_name = 'codewiz.org'
- changed_time_fmt = '[%I:%M %p] '
- date_fmt = '%a %d %b %Y'
- datetime_fmt = '%a %d %b %Y %I:%M %p'
- show_hosts = True # show hostnames?
- css_url = '../wikidata/pikiplus.css' # stylesheet link, or ''
- nonexist_qm = False # show '?' for nonexistent?
- post_edit_hook = './post_edit_hook.sh'
-
+ exec(open("geekigeeki.conf.py").read())
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,))
- break
+ action = form.getvalue('a', 'get')
+ handler = globals().get('handle_' + action)
+ if handler:
+ handler(query_string(), form)
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'
+ send_httperror("403 Forbidden", query_string())
- if file_re.match(query):
- if word_re.match(query):
- Page(query).send_page()
- elif img_re.match(query):
- send_verbatim(query, 'image/jpeg')
- else:
- send_verbatim(query)
- else:
- # TODO: return 404?
- send_title(None, msg='Can\'t work out query: ' + query)
-except:
+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()