X-Git-Url: https://codewiz.org/gitweb?a=blobdiff_plain;f=geekigeeki.py;h=678754e3a9633d29fc779eeed48eeddb8fa9fa87;hb=8d7f457924b6c4bde8486d1fd413516082ceedf9;hp=6170d7394ca6bb7b91d3fdb195d0d2ba6a5b4503;hpb=cd285df131b7e5fe14ba69d7160586694c5f9d5e;p=geekigeeki.git diff --git a/geekigeeki.py b/geekigeeki.py index 6170d73..678754e 100755 --- a/geekigeeki.py +++ b/geekigeeki.py @@ -22,12 +22,10 @@ __version__ = '$Id$'[4:12] from time import clock start_time = clock() +title_done = False import cgi, sys, os, re, errno, stat -# Regular expression defining a WikiWord -# (but this definition is also assumed in other places) -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)$", re.IGNORECASE) @@ -35,8 +33,6 @@ video_re = re.compile(r"^.*\.(ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt)$", re url_re = re.compile(r"^[a-z]{3,8}://[^\s'\"]+\S$") ext_re = re.compile(r"\.([^\./]+)$") -title_done = False - # CGI stuff --------------------------------------------------------- def script_name(): return os.environ.get('SCRIPT_NAME', '') @@ -82,16 +78,19 @@ def relative_url(pathname, privileged=False): 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('|'): - try: - key, val = arg.split('=', 1) - kwargs[key.strip()] = val.strip() - except ValueError: + 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) @@ -162,24 +161,24 @@ def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=Fal # Navbar print('
') - print link_tag('FrontPage', site_icon or 'Home', 'navlink') + print link_tag('FrontPage', site_icon or 'Home', cssclass='navlink') if name: - print(' ' + link_tag('?fullsearch=' + name, text, 'navlink') + ' ') + print(' ' + link_tag('?fullsearch=' + name, text, cssclass='navlink') + ' ') else: print(' ' + text + ' ') - print(' | ' + link_tag('FindPage', 'Find Page', 'navlink')) + print(' | ' + link_tag('FindPage', 'Find Page', cssclass='navlink')) if 'history_url' in globals(): print(' | Recent Changes') if name: print(' | Page History') if name: - print(' | ' + link_tag('?raw=' + name, 'Raw Text', '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', 'navlink', privileged=True)) + print(' | ' + link_tag('?a=edit&q=' + name, 'Edit', cssclass='navlink', privileged=True)) else: - print(' | ' + link_tag(name, 'Login', 'navlink', privileged=True)) + print(' | ' + link_tag(name, 'Login', cssclass='navlink', privileged=True)) else: print(' | Immutable Page') @@ -195,33 +194,34 @@ def send_httperror(status="403 Not Found", query=""): send_title(None, msg_text=("%s: on query '%s'" % (status, query))) send_footer() -def link_tag(params, text=None, link_class=None, privileged=False, **kvargs): +def link_tag(dest, text=None, privileged=False, **kvargs): if text is None: - text = params # default + text = humanlink(dest) elif img_re.match(text): text = '' + text + '' + link_class = kvargs.get('class', kvargs.get('cssclass', None)) if not link_class: - if is_external_url(params): + if is_external_url(dest): link_class = 'external' - elif file_re.match(params) and Page(params).exists(): + elif file_re.match(dest) and Page(dest).exists(): link_class = 'wikilink' else: - params = nonexist_pfx + params + text = nonexist_pfx + text link_class = 'nonexistent' - classattr = 'class="%s" ' % link_class # Prevent crawlers from following links potentially added by spammers or to generated pages + nofollow = '' if link_class == 'external' or link_class == 'navlink': - classattr += 'rel="nofollow"' + nofollow = 'rel="nofollow" ' - return '%s' % (classattr, relative_url(params, privileged=privileged), text) + return '%s' % (link_class, nofollow, relative_url(dest, privileged=privileged), text) def link_inline(name, descr=None, kvargs={}): - if not descr: descr = name + if not descr: descr = humanlink(name) url = relative_url(name) if video_re.match(name): - return '' % url + return '' % url elif img_re.match(name): return '%s' % (url, url + url_args(kvargs), descr) elif file_re.match(name) and not ext_re.search(name): # FIXME: this guesses a wiki page @@ -264,7 +264,6 @@ def handle_fullsearch(query, form): print_search_stats(len(hits), len(all_pages)) def handle_titlesearch(query, form): - # TODO: check needle is legal -- but probably we can just accept any RE needle = form['q'].value send_title(None, 'Title search for "' + needle + '"') @@ -325,7 +324,10 @@ def make_index_key(): links = ['%s' % (ch, ch) for ch in 'abcdefghijklmnopqrstuvwxyz'] return '

' + ' | '.join(links) + '

' -def page_list(dirname = None, re = word_re): +def page_list(dirname = None, re = None): + if re is None: + # FIXME: WikiWord is too restrictive now! + re = re.compile(r"^\b((([A-Z][a-z0-9]+){2,}/)*([A-Z][a-z0-9]+){2,})\b$") return sorted(filter(re.match, os.listdir(dirname or data_dir))) def send_footer(mod_string=None): @@ -414,8 +416,9 @@ class WikiFormatter: name = args.pop(0) if len(args): descr = args.pop(0) - # The "extthumb" nonsense works around a limitation of the HTML block model - return '
' \ + # This double div nonsense works around a limitation of the HTML block model + return '
' \ + + '
' \ + link_inline(name, descr, kvargs) \ + '
' + descr + '
' else: @@ -648,8 +651,8 @@ class Page: 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' + maxwidth_arg = ' | maxwidth=' + str(image_maxwidth) + out += '{{' + self.page_name + '/' + filename + ' | ' + humanlink(filename) + maxwidth_arg + ' | class=thumbleft}}\n' else: out += ' * [[' + self.page_name + '/' + filename + ']]\n' return out