X-Git-Url: https://codewiz.org/gitweb?p=geekigeeki.git;a=blobdiff_plain;f=geekigeeki.py;fp=geekigeeki.py;h=4d50a3af5c0fd4aa29c24ec1918e59f801a2d047;hp=faf59027217f4c59f0300d99c22f4a22f487e4b5;hb=41e94725555d64df1b6e56ffd29a733c28e0bfdf;hpb=d810022b52bf2aa2c7a4b4c530dd7a7f03dc9299 diff --git a/geekigeeki.py b/geekigeeki.py index faf5902..4d50a3a 100755 --- a/geekigeeki.py +++ b/geekigeeki.py @@ -22,22 +22,17 @@ __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)$", re.IGNORECASE) video_re = re.compile(r"^.*\.(ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt)$", 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"\.([^\./]+)$") -title_done = False - # CGI stuff --------------------------------------------------------- def script_name(): return os.environ.get('SCRIPT_NAME', '') @@ -83,21 +78,44 @@ def relative_url(pathname, privileged=False): def permalink(s): return re.sub(' ', '-', re.sub('[^a-z0-9_ ]', '', s.lower()).strip()) +def humanlink(s): + return re.search('([^:/\.]+)(?:\.[^/:]+|)$', s).group(1).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: + 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" + print("Content-type: " + mime_type + "; charset=utf-8\n") def send_guru(msg_text, msg_type): if not msg_text: return - print '
'
+    print('
')
     if msg_type == 'error':
-        print '    Software Failure.  Press left mouse button to continue.\n'
-    print msg_text
+        print('    Software Failure.  Press left mouse button to continue.\n')
+    print(msg_text)
     if msg_type == 'error':
         print '\n           Guru Meditation #DEADBEEF.ABADC0DE'
-    print '
'
-    print '' \
-        % relative_url('sys/GuruMeditation.js')
+    print('
' \ + % relative_url('sys/GuruMeditation.js')) def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=False): global title_done @@ -105,80 +123,80 @@ def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=Fal # Head emit_header() - print '' - print '' + print('') + print('') - print "%s: %s" % (site_name, text) - print ' ' + print("%s: %s" % (site_name, text)) + print(' ') if not name: - print ' ' + print(' ') for meta in meta_urls: http_equiv, content = meta - print ' ' % (http_equiv, relative_url(content)) + print(' ' % (http_equiv, relative_url(content))) for link in link_urls: rel, href = link - print ' ' % (rel, relative_url(href)) + print(' ' % (rel, relative_url(href))) if name and writable and privileged_url is not None: - print ' ' \ - % (privileged_path() + '?edit=' + name) + print(' ' \ + % (privileged_path() + '?a=edit&q=' + name)) if history_url is not None: - print ' ' \ - % relative_url(history_url + '?a=rss') + print(' ' \ + % relative_url(history_url + '?a=rss')) - print '' + print('') # Body if name and writable and privileged_url is not None: - print '' + print('') else: - print '' + print('') title_done = True send_guru(msg_text, msg_type) # Navbar - print '
' + print('
') print link_tag('FrontPage', site_icon or 'Home', 'navlink') if name: - print ' ' + link_tag('?fullsearch=' + name, text, 'navlink') + ' ' + print(' ' + link_tag('?fullsearch=' + name, text, 'navlink') + ' ') else: - print ' ' + text + ' ' - print ' | ' + link_tag('FindPage', 'Find Page', 'navlink') + print(' ' + text + ' ') + print(' | ' + link_tag('FindPage', 'Find Page', 'navlink')) if 'history_url' in globals(): - print ' | Recent Changes' + print(' | Recent Changes') if name: - print ' | Page History' + print(' | Page History') if name: - print ' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink') + print(' | ' + link_tag('?raw=' + name, 'Raw Text', 'navlink')) if privileged_url is not None: if writable: - print ' | ' + link_tag('?edit=' + name, 'Edit', 'navlink', privileged=True) + print(' | ' + link_tag('?a=edit&q=' + name, 'Edit', 'navlink', privileged=True)) else: - print ' | ' + link_tag(name, 'Login', 'navlink', privileged=True) + print(' | ' + link_tag(name, 'Login', 'navlink', privileged=True)) else: - print ' | Immutable Page' + print(' | Immutable Page') user = remote_user() if user != 'AnonymousCoward': - print ' | ' + print(' | ') - print '
' + print('
') def send_httperror(status="403 Not Found", query=""): - print "Status: %s" % status + print("Status: %s" % status) send_title(None, msg_text=("%s: on query '%s'" % (status, query))) send_footer() -def link_tag(params, text=None, link_class=None, privileged=False): +def link_tag(params, text=None, link_class=None, privileged=False, **kvargs): if text is None: - text = params # default + text = humanlink(params) elif img_re.match(text): text = '' + text + '' @@ -198,13 +216,13 @@ def link_tag(params, text=None, link_class=None, privileged=False): return '%s' % (classattr, relative_url(params, privileged=privileged), text) -def link_inline(name, descr=None, args=''): - if not descr: descr = name +def link_inline(name, descr=None, kvargs={}): + if not descr: descr = humanlink(name) url = relative_url(name) if video_re.match(name): return '' % url elif img_re.match(name): - return '%s' % (url, url + args, descr) + 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 return Page(name).send_naked() else: @@ -213,8 +231,12 @@ def link_inline(name, descr=None, args=''): # Search --------------------------------------------------- -def handle_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 = [] @@ -230,47 +252,44 @@ def handle_fullsearch(needle): hits.sort() hits.reverse() - print "") print_search_stats(len(hits), len(all_pages)) -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 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 print_search_stats(hits, searched): - print "

%d hits out of %d pages searched.

" % (hits, searched) - -def handle_raw(pagename): +def handle_raw(pagename, form): if not file_re.match(pagename): send_httperror("403 Forbidden", pagename) return Page(pagename).send_raw() -def handle_edit(pagename): +def handle_edit(pagename, form): if not file_re.match(pagename): send_httperror("403 Forbidden", pagename) return - pg = Page(pagename) + pg = Page(form['q'].value) if 'save' in form: if form['file'].value: pg.save(form['file'].file.read(), form['changelog'].value) @@ -287,12 +306,27 @@ def handle_edit(pagename): 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(): - links = map(lambda ch: '%s' % (ch, ch), 'abcdefghijklmnopqrstuvwxyz') - return '

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

' + 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): @@ -300,16 +334,16 @@ def send_footer(mod_string=None): cgi.print_arguments() cgi.print_form(form) cgi.print_environ() - print ''' + print(''' ' + print('

last modified %s

' % mod_string) + print('') class WikiFormatter: """Object that turns Wiki markup into HTML. @@ -359,48 +393,35 @@ class WikiFormatter: return self._undent() + '\n
\n' % (len(word) - 2) 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: - msg = '<<' + '|'.join(argv) + '>>' + 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 = link_re.match(word) - return link_tag(m.group(1), m.group(2)) + args, kvargs = parse_args(word) + return link_tag(*args, **kvargs) 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 '
' \ - + link_inline(name, descr, args) \ + 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_inline(name, name) + return link_inline(name, None, kvargs) def _html_repl(self, word): if not self.in_html and word.startswith('

' + print('

') scan_re = re.compile(r"""(?: # Styles and formatting @@ -552,9 +573,9 @@ class WikiFormatter: 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()): + #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 self.line.startswith('#'): @@ -562,23 +583,24 @@ class WikiFormatter: self.in_header = False if self.in_pre: - print re.sub(pre_re, self.replace, self.line) + print(re.sub(pre_re, self.replace, self.line)) else: if self.in_table and not tr_re.match(self.line): self.in_table = False - print '

' + print('

') if blank_re.match(self.line): - print '

' + print('

') else: indent = indent_re.match(self.line) - print self._indent_to(len(indent.group(0))) , - print re.sub(scan_re, self.replace, 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 '

' + if self.in_pre: print('') + if self.in_table: print('

') + print(self._undent()) + print('

') class Page: def __init__(self, page_name): @@ -594,13 +616,13 @@ class Page: return os.path.join(data_dir, self.page_name) def _tmp_filename(self): - return os.path.join(data_dir, ('#' + self.page_name.replace('/','_') + '.' + `os.getpid()` + '#')) + return os.path.join(data_dir, ('#' + self.page_name.replace('/','_') + '.' + str(os.getpid()) + '#')) def exists(self): try: os.stat(self._filename()) return True - except OSError, err: + except OSError as err: if err.errno == errno.ENOENT: return False raise err @@ -608,10 +630,10 @@ class Page: def get_raw_body(self, default=None): try: return open(self._filename(), 'rb').read() - except IOError, err: + except IOError as err: if err.errno == errno.ENOENT: if default is None: - default = '//[[?edit=%s|Describe %s]]//' % (self.page_name, self.page_name) + default = '//[[%s|Describe %s|action=edit]]//' % (self.page_name, self.page_name) return default if err.errno == errno.EISDIR: return self.format_dir() @@ -628,8 +650,8 @@ 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 @@ -638,17 +660,17 @@ class Page: if not '_pragmas' in self.__dict__: self._pragmas = {} try: - f = open(self._filename(), 'rt') + file = open(self._filename(), 'rt') attr_re = re.compile(r"^#(\S*)(.*)$") - for line in f: + 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: + #print "bernie: pragmas[" + m.group(1) + "] = " + m.group(2) + "
\n" + except IOError as err: if err.errno != errno.ENOENT and err.errno != errno.EISDIR: - raise err + raise er return self._pragmas def pragma(self, name, default): @@ -697,7 +719,7 @@ class Page: try: from time import localtime, strftime modtime = localtime(os.stat(self._filename())[stat.ST_MTIME]) - except OSError, err: + except OSError as err: if err.errno != errno.ENOENT: raise err return None @@ -713,17 +735,17 @@ class Page: if 'file' in form: filename = form['file'].value - print ('

Editing ' + self.page_name + print(('

Editing ' + self.page_name + ' for ' + cgi.escape(remote_user()) + ' from ' + cgi.escape(get_hostname(remote_host())) - + '

') - print '
' % relative_url(self.page_name) - print '' % (self.page_name) - print '
' % (self.page_name) - print '' \ - % cgi.escape(preview or self.get_raw_body(default='')) - print ' ' % filename - print """ + + '

')) + print('
' % relative_url(self.page_name)) + print('' % (self.page_name)) + print('
' % (self.page_name)) + print('' \ + % cgi.escape(preview or self.get_raw_body(default=''))) + print(' ' % filename) + print("""
@@ -736,12 +758,12 @@ class Page: document.editform.savetext.focus() //--> - """ - print "

" + link_tag('EditingTips') + "

" + """) + print("

" + link_tag('EditingTips') + "

") if preview: - print "
" + print("
") WikiFormatter(preview).print_html() - print "
" + print("
") send_footer() def send_raw(self, mimetype='text/plain', args=[]): @@ -758,7 +780,7 @@ class Page: else: body = self.get_raw_body() emit_header(mimetype) - print body + print(body) def _write_file(self, data): tmp_filename = self._tmp_filename() @@ -768,7 +790,7 @@ class Page: # Bad Bill! POSIX rename ought to replace. :-( try: os.remove(name) - except OSError, err: + except OSError as err: if err.errno != errno.ENOENT: raise err os.rename(tmp_filename, name) @@ -794,29 +816,16 @@ class Page: self.msg_text = 'Thank you for your contribution. Your attention to detail is appreciated.' self.msg_type = 'success' -def main(): - for cmd in form: - handler = globals().get('handle_' + cmd) - if handler: - handler(form[cmd].value) - break - else: - query = query_string() - if file_re.match(query): - # 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: - Page(query).format() - else: - send_httperror("403 Forbidden", query) - try: - execfile("geekigeeki.conf.py") + exec(open("geekigeeki.conf.py").read()) form = cgi.FieldStorage() - main() + action = form.getvalue('a', 'get') + handler = globals().get('handle_' + action) + if handler: + handler(query_string(), form) + else: + send_httperror("403 Forbidden", query_string()) + except Exception: import traceback msg_text = traceback.format_exc()