# 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$")
-word_re = re.compile(r"^\b((([A-Z][a-z]+){2,}/)*([A-Z][a-z]+){2,})\b$")
+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)$", re.IGNORECASE)
url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$")
pathname = url + '/' + pathname
return pathname
-# Formatting stuff --------------------------------------------------
+def permalink(s):
+ return re.sub(' ', '-', re.sub('[^a-z0-9_ ]', '', s.lower()).strip())
+# Formatting stuff --------------------------------------------------
def emit_header(mime_type="text/html"):
print "Content-type: " + mime_type + "; charset=utf-8\n"
# Navbar
print '<div class="nav">'
- print ' <b>' + site_name + ': ',
if name:
- print link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
+ print ' <b>' + link_tag('?fullsearch=' + name, text, 'navlink') + '</b> '
else:
- print text + '</b> '
+ print ' <b>' + text + '</b> '
print ' | ' + link_tag('FrontPage', 'Front Page', 'navlink')
print ' | ' + link_tag('FindPage', 'Find Page', 'navlink')
- if globals().has_key('history_url'):
+ if 'history_url' in globals():
print ' | <a href="' + history_url + '" class="navlink">Recent Changes</a>'
if name:
print ' | <a href="' + history_url + '?a=history;f=' + name + '" class="navlink">Page History</a>'
# 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_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 + '"')
def print_search_stats(hits, searched):
print "<p>%d hits out of %d pages searched.</p>" % (hits, searched)
-def do_raw(pagename):
+def handle_raw(pagename):
Page(pagename).send_raw()
-def do_edit(pagename):
+def handle_edit(pagename):
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())
+ else:
+ pg.save(form['savetext'].value.replace('\r\n', '\n'))
+ pg.format()
elif 'cancel' in form:
pg.msg_text = 'Editing canceled'
pg.msg_type = 'notice'
- pg.send_page()
- else:
- pg.send_editor()
+ pg.format()
+ else: # preview or edit
+ text = None
+ if 'preview' in form:
+ text = form['savetext'].value
+ pg.send_editor(text)
def make_index_key():
links = map(lambda ch: '<a href="#%s">%s</a>' % (ch, ch), 'abcdefghijklmnopqrstuvwxyz')
def send_footer(name, 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 id="footer"><hr />'
print ('<p class="copyright"><span class="benchmark">generated in %0.3fs</span>' +
print '<p class="modified">last modified %s</p>' % mod_string
print '</div></body></html>'
-# ----------------------------------------------------------
-# Macros
-def _macro_TitleSearch(*vargs):
- return _macro_search("titlesearch")
-
-def _macro_FullSearch(*vargs):
- 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="Search" /></form>""" % (type, default)
-
-def _macro_WordIndex(*vargs):
- 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 = word[0].lower()
- 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(*vargs):
- s = make_index_key()
- pages = list(page_list())
- pages.sort()
- current_letter = None
- for name in pages:
- letter = name[0].lower()
- if letter != current_letter:
- s += '<a name="%s"><h3>%s</h3></a>' % (letter, letter)
- current_letter = letter
- else:
- s += '<br />'
- s += Page(name).link_to()
- return s
-
-
-# ----------------------------------------------------------
class PageFormatter:
"""Object that turns Wiki markup into HTML.
def __init__(self, raw):
self.raw = raw
self.h_level = 0
- self.in_pre = self.in_table = False
+ self.in_pre = self.in_html = self.in_table = self.in_li = False
self.in_header = True
self.list_indents = []
- self.tr_cnt = self.h_cnt = 0
+ self.tr_cnt = 0
self.styles = {
#wiki html enabled?
"//": ["em", False],
def _tit_repl(self, word):
if self.h_level:
- result = '</h%d>' % self.h_level
+ result = '</h%d><p>\n' % self.h_level
self.h_level = 0
else:
self.h_level = len(word) - 1
- self.h_cnt += 1
- #abridged = re.sub('[^a-z_]', '', word.lower().replace(' ', '_'))
- result = '<h%d id="%d"><a class="heading" href="#%d">¶</a> ' % (self.h_level, self.h_cnt, self.h_cnt)
+ link = permalink(self.line)
+ result = '\n</p><h%d id="%s"><a class="heading" href="#%s">¶</a> ' % (self.h_level, link, link)
return result
def _br_repl(self, word):
else:
return '<a href="%s" rel="nofollow" class="external">%s</a>' % (word, word)
- def _hurl_repl(self, word):
- m = re.compile("\[\[([^ \t\n\r\f\v\|]+)(?:\s*\|\s*([^\]]+)|)\]\]").match(word)
+ def _macro_repl(self, word):
+ m = re.compile("\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>").match(word)
name = m.group(1)
- descr = m.group(2) or name
+ 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:
+ execfile("macros/" + name + ".py", globals())
+ macro = globals().get('_macro_' + name)
if macro:
- return apply(macro, (name, descr))
- elif img_re.match(name):
+ return macro(argv)
+ else:
+ return '<strong class="error"><<' + '|'.join(argv) + '>></strong>'
+
+ def _hurl_repl(self, word):
+ m = re.compile("\[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\]").match(word)
+ name = m.group(1)
+ descr = m.group(2) or name
+
+ if img_re.match(name):
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)
return '<a href="mailto:%s">%s</a>' % (word, word)
def _html_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 '>'
return {'&': '&',
'<': '<',
'>': '>'}[s]
def _li_repl(self, match):
- return '<li>'
+ if self.in_li:
+ return '</li><li>'
+ else:
+ self.in_li = True
+ return '<li>'
def _pre_repl(self, word):
if word == '{{{' and not self.in_pre:
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)
def replace(self, match):
for type, hit in match.groupdict().items():
if hit:
- return apply(getattr(self, '_' + type + '_repl'), (hit,))
+ return getattr(self, '_' + type + '_repl')(hit)
else:
raise "Can't handle match " + `match`
+ r"|(?P<tit>\={2,6})"
+ r"|(?P<br>\\\\)"
+ r"|(?P<rule>^-{3,})"
- + r"|(?P<html><(/|)(div|span|iframe)[^<>]*>)"
- + r"|(?P<ent>[<>&])"
+ r"|(?P<hi>\b(FIXME|TODO|DONE)\b)"
# 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>\[\[([^ \t\n\r\f\v\|]+)(?:\s*\|\s*([^\]]+)|)\]\])"
- + r"|(?P<url>(http|https|ftp|mailto)\:[^\s'\"]+\S)"
- + r"|(?P<email>[-\w._+]+\@[\w.-]+)"
+ + r"|(?P<macro>\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)"
+ + r"|(?P<hurl>\[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])"
+
+ # Inline HTML
+ + r"|(?P<html><(/|)(br|hr|div|form|iframe|input|span))"
+ + r"|(?P<ent>[<>&])"
+
+ # Auto links
+ + r"|(?P<img>\b[a-zA-Z0-9_-]+\.(png|gif|jpg|jpeg|bmp))" # LEGACY
+ + r"|(?P<word>\b(?:[A-Z][a-z]+){2,}\b)" # LEGACY
+ + r"|(?P<url>(http|https|ftp|mailto)\:[^\s'\"]+\S)" # LEGACY
+ + r"|(?P<email>[-\w._+]+\@[\w.-]+)" # LEGACY
# Lists, divs, spans
+ r"|(?P<li>^\s+[\*#] +)"
+ r"|(?P<pre>\{\{\{|\s*\}\}\})"
+ + r"|(?P<inl>\{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})" #TODO
# Tables
+ r"|(?P<tr>^\s*\|\|(=|)\s*)"
indent_re = re.compile(r"^\s*")
tr_re = re.compile(r"^\s*\|\|")
eol_re = re.compile(r"\r?\n")
- for line in eol_re.split(self.raw.expandtabs()):
+ for self.line in eol_re.split(self.raw.expandtabs()):
# Skip ACLs
if self.in_header:
- if line.startswith('#'):
+ if self.line.startswith('#'):
continue
self.in_header = False
if self.in_pre:
- print re.sub(pre_re, self.replace, line)
+ print re.sub(pre_re, self.replace, self.line)
else:
- if self.in_table and not tr_re.match(line):
+ if self.in_table and not tr_re.match(self.line):
self.in_table = False
print '</tbody></table><p>'
- if blank_re.match(line):
+ if blank_re.match(self.line):
print '</p><p>'
else:
- indent = indent_re.match(line)
+ indent = indent_re.match(self.line)
print self._indent_to(len(indent.group(0)))
- print re.sub(scan_re, self.replace, line)
+ print re.sub(scan_re, self.replace, self.line)
if self.in_pre: print '</pre>'
if self.in_table: print '</tbody></table><p>'
print self._undent()
print '</p></div>'
-# ----------------------------------------------------------
class Page:
def __init__(self, page_name):
self.page_name = page_name
raise er
def get_attrs(self):
- if self.__dict__.has_key('attrs'):
+ if 'attrs' in self.__dict__:
return self.attrs
self.attrs = {}
try:
return self.get_attrs().get(name, default)
def can(self, action, default=True):
+ acl = None
try:
#acl SomeUser:read,write All:read
acl = self.get_attr("acl", None)
return action in perms.split(',')
return False
except Exception, er:
- pass
+ if acl:
+ self.msg_text = 'Illegal acl line: ' + acl
return default
def can_write(self):
def can_read(self):
return self.can("read", True)
- def send_page(self):
+ def send_naked(self):
+ if self.can_read():
+ PageFormatter(self.get_raw_body()).print_html()
+ else:
+ send_guru("Read access denied by ACLs", "notice")
+
+ def format(self):
page_name = None
if self.can_write():
page_name = self.page_name
css_url = self.get_attr("css", "").split() + css_url
send_title(page_name, self.split_title(), msg_text=self.msg_text, msg_type=self.msg_type)
- if self.can_read():
- PageFormatter(self.get_raw_body()).print_html()
- else:
- send_guru("Read access denied by ACLs", "notice")
+ self.send_naked()
send_footer(page_name, self._last_modified())
def _last_modified(self):
send_guru("Write access denied by ACLs", "error")
return
+ file = ''
+ if 'file' in form:
+ file = 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">' % relative_url(self.page_name)
+ print '<div class="editor"><form 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 """<textarea wrap="off" spellcheck="true" id="editor" name="savetext" rows="17" cols="100">%s</textarea>""" % (preview or self.get_raw_body())
+ print '<textarea wrap="off" spellcheck="true" id="editor" name="savetext" rows="17" cols="100">%s</textarea>' % (preview or self.get_raw_body())
+ print 'Or upload a file: <input type="file" name="file" value="%s" />' % file
print """
<br />
<input type="submit" name="save" value="Save" />
else:
send_title(None, msg_text='Read access denied by ACLs', msg_type='notice')
- 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._text_filename()
if os.name == 'nt':
# Bad Bill! POSIX rename ought to replace. :-(
try:
- os.remove(text)
+ os.remove(name)
except OSError, er:
if er.errno != errno.ENOENT: raise er
- os.rename(tmp_filename, text)
+ os.rename(tmp_filename, name)
- def save_text(self, newtext):
+ def save(self, newdata):
if not self.can_write():
self.msg_text = 'Write access denied by ACLs'
self.msg_type = 'error'
return
- self._write_file(newtext)
+ self._write_file(newdata)
rc = 0
if post_edit_hook:
# FIXME: what's the std way to perform shell quoting in python?
try:
execfile("geekigeeki.conf.py")
form = cgi.FieldStorage()
- handlers = {
- 'fullsearch': do_fullsearch,
- 'titlesearch': do_titlesearch,
- 'edit': do_edit,
- 'raw': do_raw,
- }
-
- for cmd in handlers.keys():
- if form.has_key(cmd):
- apply(handlers[cmd], (form[cmd].value,))
+
+ for cmd in form:
+ handler = globals().get('handle_' + cmd)
+ if handler:
+ handler(form[cmd].value)
break
else:
path_info = environ.get('PATH_INFO', '')
if file_re.match(query):
if word_re.match(query):
- Page(query).send_page()
- elif img_re.match(query):
- #FIXME: use correct mime type
- Page(query).send_raw(mimetype='image/jpeg')
+ Page(query).format()
else:
- Page(query).send_raw(mimetype='application/octet-stream')
+ from mimetypes import MimeTypes
+ type, encoding = MimeTypes().guess_type(query)
+ type = type or 'text/plain'
+ Page(query).send_raw(mimetype=type)
else:
print "Status: 404 Not Found"
send_title(None, msg_text='Can\'t work out query: ' + query)
+ send_footer(None)
except Exception:
import traceback
msg_text = traceback.format_exc()