X-Git-Url: https://codewiz.org/gitweb?a=blobdiff_plain;f=geekigeeki.py;h=eba446715e1e6f08bcdb19d01fbcfe7f3b3c2025;hb=06653e2489969feaf8ec69fc449187eaf30ea525;hp=23b05d5cae67fbef5a7de5695548426e29c8aefe;hpb=c76ddda68748f9bc67641ccba61314d8d8b32b12;p=geekigeeki.git diff --git a/geekigeeki.py b/geekigeeki.py index 23b05d5..eba4467 100755 --- a/geekigeeki.py +++ b/geekigeeki.py @@ -3,7 +3,7 @@ # # Copyright 1999, 2000 Martin Pool # Copyright 2002 Gerardo Poggiali -# Copyright 2007, 2008 Bernie Innocenti +# 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 @@ -30,18 +30,24 @@ import cgi, sys, os, re, errno, stat 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)$", re.IGNORECASE) +img_re = re.compile(r"^.*\.(png|gif|jpg|jpeg|bmp|ico|ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt)$", 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("(?:\[\[|{{)([^\s\|]+)(?:\s*\|\s*([^\]]+)|)(?:\]\]|}})") title_done = False - # 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() @@ -61,8 +67,11 @@ def get_hostname(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 (url_re.match(pathname) or pathname.startswith('/')): + if not is_external_url(pathname): if privileged: url = privileged_path() else: @@ -77,6 +86,13 @@ def permalink(s): 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 '
'
@@ -86,28 +102,10 @@ def send_guru(msg_text, msg_type):
     if msg_type == 'error':
         print '\n      Guru Meditation #DEADBEEF.ABADC0DE'
     print '
' - # FIXME: This little JS snippet is harder to pass than ACID 3.0 - print """ - """ + try: + sendfile(sys.stdout, open('gurumeditation.js', 'rb')) + except IOError, err: + pass def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=False): global title_done @@ -153,11 +151,11 @@ def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=Fal # Navbar print '
' + print link_tag('FrontPage', relative_url(site_icon or 'Home'), 'navlink') if name: print ' ' + link_tag('?fullsearch=' + name, text, 'navlink') + ' ' else: print ' ' + text + ' ' - print ' | ' + link_tag('FrontPage', 'Home', 'navlink') print ' | ' + link_tag('FindPage', 'Find Page', 'navlink') if 'history_url' in globals(): print ' | Recent Changes' @@ -177,7 +175,7 @@ def send_title(name, text="Limbo", msg_text=None, msg_type='error', writable=Fal user = remote_user() if user != 'AnonymousCoward': - print ' | ' + print ' | ' print '
' @@ -186,19 +184,35 @@ 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, ss_class=None, privileged=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 potentially added by spammers or to generated pages - if ss_class == 'external' or ss_class == 'navlink': - classattr += 'rel="nofollow" ' - elif url_re.match(params): - classattr += 'rel="nofollow" ' + elif img_re.match(text): + 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 '%s' % (classattr, relative_url(params, privileged=privileged), text) +def link_inline(name, descr=None, args=''): + if not descr: descr = name + if video_re.match(name): + return '' % name + else: + return '%s' % (name, name + args, descr) + # Search --------------------------------------------------- def handle_fullsearch(needle): @@ -220,7 +234,7 @@ def handle_fullsearch(needle): print "
    " for (count, page_name) in hits: - print '

  • ' + Page(page_name).link_to() + print '

  • ' + link_tag(page_name) print ' . . . . ' + `count` print ['match', 'matches'][count != 1] print '

    • ' @@ -238,7 +252,7 @@ def handle_titlesearch(needle): print "
      " for filename in hits: - print '

    • ' + Page(filename).link_to() + "

      • " + print '

    • ' + link_tag(filename) + "

      • " print "
    " print_search_stats(len(hits), len(all_pages)) @@ -290,10 +304,10 @@ def send_footer(mod_string=None): print ''' ' @@ -314,14 +328,14 @@ class WikiFormatter: self.styles = { #wiki html enabled? "//": ["em", False], - "''": ["em", False], "**": ["b", False], - "'''": ["b", False], "##": ["tt", False], - "``": ["tt", False], "__": ["u", False], "^^": ["sup", False], - ",,": ["sub", False] + ",,": ["sub", False], + "''": ["em", False], # LEGACY + "'''": ["b", False], # LEGACY + "``": ["tt", False], # LEGACY } def _b_repl(self, word): @@ -345,19 +359,6 @@ class WikiFormatter: def _rule_repl(self, word): return self._undent() + '\n
    \n' % (len(word) - 2) - def _word_repl(self, word): - return Page(word).link_to() - - def _img_repl(self, word): - pathname = relative_url(word) - return '' % (pathname, pathname) - - def _url_repl(self, word): - if img_re.match(word): - return '' % (word, word) - else: - return '%s' % (word, word) - def _macro_repl(self, word): m = re.compile("\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>").match(word) name = m.group(1) @@ -370,61 +371,70 @@ class WikiFormatter: if not macro: try: execfile("macros/" + name + ".py", globals()) - except IOError, er: - if er.errno == errno.ENOENT: - pass + except IOError, err: + if err.errno == errno.ENOENT: pass macro = globals().get('_macro_' + name) if macro: return macro(argv) else: - return '<<' + '|'.join(argv) + '>>' + msg = '<<' + '|'.join(argv) + '>>' + if not self.in_html: + msg = '' + msg + '' + return msg def _hurl_repl(self, word): m = link_re.match(word) - name = m.group(1) - descr = m.group(2) - if descr is None: - descr = name - elif img_re.match(m.group(2)): - descr = '' - - return link_tag(name, descr, 'wikilink') + return link_tag(m.group(1), m.group(2)) def _inl_repl(self, word): m = link_re.match(word) - name = m.group(1) - descr = m.group(2) or name - name = relative_url(name) - argv = descr.split('|') - descr = argv.pop(0) + name = relative_url(m.group(1)) + descr = m.group(2) - if argv: - args = '?' + '&'.join(argv) - else: + if descr: + argv = descr.split('|') + descr = argv.pop(0) args = '' + if argv: + args = '?' + '&'.join(argv) - if descr: # The "extthumb" nonsense works around a limitation of the HTML block model - return '
    %s
    %s
    ' \ - % (name, name + args, descr, descr) + return '
    ' \ + + link_inline(name, descr, args) \ + + '
    ' + descr + '
    ' else: - return '' % (name, name + args) - - def _email_repl(self, word): - return '%s' % (word, word) + return link_inline(name, name) 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 and s == '>': - self.in_html -= 1 - return '>' + if self.in_html: + return s; # Pass through return {'&': '&', '<': '<', '>': '>'}[s] + def _img_repl(self, word): # LEGACY + return self._inl_repl('{{' + word + '}}') + + 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 '
  • ' @@ -495,58 +505,58 @@ class WikiFormatter: if 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 - # TODO: highlight search words (look at referrer) - scan_re = re.compile( - r"(?:" - # Formatting - + r"(?P\*\*|'''|//|''|##|``|__|\^\^|,,)" - + r"|(?P\={2,6})" - + r"|(?P
    \\\\)" - + r"|(?P^-{3,})" - + r"|(?P\b(FIXME|TODO|DONE)\b)" + scan_re = re.compile(r"""(?: + # Styles and formatting + (?P \*\*|'''|//|''|\#\#|``|__|\^\^|,,) + | (?P \={2,6}) + | (?P
    \\\\) + | (?P ^-{3,}) + | (?P \b( FIXME | TODO | DONE )\b ) # Links - + r"|(?P\<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>)" - + r"|(?P\[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\])" + | (?P \<\<([^\s\|\>]+)(?:\s*\|\s*([^\>]+)|)\>\>) + | (?P \[\[([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\]\]) # Inline HTML - + r"|(?P<(/|)(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])[^>]*>)" - + r"|(?P[<>&])" + | (?P <(br|hr|div|span|form|iframe|input|textarea|a|img|h[1-5])\b ) + | (?P ( /\s*> | ) ) + | (?P [<>&] ) # Auto links (LEGACY) - + r"|(?P\b[a-zA-Z0-9_/-]+\.(png|gif|jpg|jpeg|bmp|ico))" - + r"|(?P\b(?:[A-Z][a-z]+){2,}\b)" - + r"|(?P(http|https|ftp|mailto)\:[^\s'\"]+\S)" - + r"|(?P[-\w._+]+\@[\w.-]+)" + | (?P \b[a-zA-Z0-9_/-]+\.(png|gif|jpg|jpeg|bmp|ico|ogm|ogg|mkv|mpg|mpeg|mp4|avi|asf|flv|wmv|qt)) + | (?P \b(?:[A-Z][a-z]+){2,}\b) + | (?P (http|https|ftp|mailto)\:[^\s'\"]+\S) + | (?P [-\w._+]+\@[\w.-]+) # Lists, divs, spans - + r"|(?P

  • ^\s+[\*#] +)" - + r"|(?P
    \{\{\{|\s*\}\}\})"
-            + r"|(?P\{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})"
+            | (?P
  •     ^\s+[\*\#]\s+)
+            | (?P
       \{\{\{|\s*\}\}\})
+            | (?P   \{\{([^\s\|]+)(?:\s*\|\s*([^\]]+)|)\}\})
 
             # Tables
-            + r"|(?P^\s*\|\|(=|)\s*)"
-            + r"|(?P\s*\|\|(=|)\s*$)"
-            + r"|(?P\s*\|\|(=|)\s*)"
-            + r")")
-        pre_re = re.compile(
-            r"(?:"
-            + r"(?P
    \s*\}\}\})"
-            + r"|(?P[<>&])"
-            + r")")
+            | (?P    ^\s*\|\|(=|)\s*)
+            | (?P   \s*\|\|(=|)\s*$)
+            | (?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
         for self.line in eol_re.split(self.raw.expandtabs()):
-            # Skip ACLs
+            # Skip pragmas
             if self.in_header:
                 if self.line.startswith('#'):
                     continue
@@ -591,27 +601,20 @@ class Page:
         try:
             os.stat(self._filename())
             return True
-        except OSError, er:
-            if er.errno == errno.ENOENT:
+        except OSError, err:
+            if err.errno == errno.ENOENT:
                 return False
-            raise er
-
-    def link_to(self):
-        word = self.page_name
-        if self.exists():
-            return link_tag(word, word, 'wikilink')
-        else:
-            return link_tag(word, nonexist_pfx + word, 'nonexistent')
+            raise err
 
     def get_raw_body(self):
         try:
             return open(self._filename(), 'rb').read()
-        except IOError, er:
-            if er.errno == errno.ENOENT:
+        except IOError, err:
+            if err.errno == errno.ENOENT:
                 return '' # just doesn't exist, use default
-            if er.errno == errno.EISDIR:
+            if err.errno == errno.EISDIR:
                 return self.format_dir()
-            raise er
+            raise err
 
     def format_dir(self):
         out = '== '
@@ -629,32 +632,32 @@ class Page:
             else:
                 out += ' * [[' + self.page_name + '/' + filename + ']]\n'
         return out
-    def get_attrs(self):
-        if 'attrs' in self.__dict__:
-            return self.attrs
-        self.attrs = {}
-        try:
-            f = open(self._filename(), 'rt')
-            attr_re = re.compile(r"^#(\S*)(.*)$")
-            for line in f:
-                m = attr_re.match(line)
-                if not m:
-                    break
-                self.attrs[m.group(1)] = m.group(2).strip()
-                #print "bernie: attrs[" + m.group(1) + "] = " + m.group(2) + "
    \n"
-        except IOError, er:
-            if er.errno != errno.ENOENT and er.errno != errno.EISDIR:
-                raise er
-        return self.attrs
-
-    def get_attr(self, name, default):
-        return self.get_attrs().get(name, default)
+
+    def pragmas(self):
+        if not '_pragmas' in self.__dict__:
+            self._pragmas = {}
+            try:
+                f = open(self._filename(), 'rt')
+                attr_re = re.compile(r"^#(\S*)(.*)$")
+                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) + "
    \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:
             #acl SomeUser:read,write All:read
-            acl = self.get_attr("acl", None)
+            acl = self.pragma("acl", None)
             for rule in acl.split():
                 (user, perms) = rule.split(':')
                 if user == remote_user() or user == "All":
@@ -679,7 +682,7 @@ class Page:
 
     def format(self):
         #css foo.css
-        value = self.get_attr("css", None)
+        value = self.pragma("css", None)
         if value:
             global link_urls
             link_urls += [ [ "stylesheet", value ] ]
@@ -693,9 +696,9 @@ class Page:
         try:
             from time import localtime, strftime
             modtime = localtime(os.stat(self._filename())[stat.ST_MTIME])
-        except OSError, er:
-            if er.errno != errno.ENOENT:
-                raise er
+        except OSError, err:
+            if err.errno != errno.ENOENT:
+                raise err
             return None
         return strftime(datetime_fmt, modtime)
 
@@ -732,22 +735,18 @@ class Page:
             //-->
             
             """
-        print "
    " + Page('EditingTips').link_to() + "
    "
+        print "
    " + link_tag('EditingTips') + "
    "
         if preview:
             print "
    "
             WikiFormatter(preview).print_html()
             print "
    "
         send_footer()
 
-    def send_raw(self, mimetype='text/plain'):
-        if self.can_read():
-            body = self.get_raw_body()
-            emit_header(mimetype)
-            print body
-        else:
+    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
 
-    def send_image(self, mimetype, args=[]):
         if 'maxwidth' in args:
             import subprocess
             emit_header(mimetype)
@@ -755,7 +754,9 @@ class Page:
             subprocess.check_call(['gm', 'convert', self._filename(),
                 '-scale', args['maxwidth'].value + ' >', '-'])
         else:
-            self.send_raw(mimetype)
+            body = self.get_raw_body()
+            emit_header(mimetype)
+            print body
 
     def _write_file(self, data):
         tmp_filename = self._tmp_filename()
@@ -765,8 +766,8 @@ class Page:
             # Bad Bill!  POSIX rename ought to replace. :-(
             try:
                 os.remove(name)
-            except OSError, er:
-                if er.errno != errno.ENOENT: raise er
+            except OSError, err:
+                if err.errno != errno.ENOENT: raise err
         os.rename(tmp_filename, name)
 
     def save(self, newdata, changelog):
@@ -778,19 +779,14 @@ class Page:
         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
-                + " '" + data_dir + '/' + self.page_name
-                + "' '" + remote_user()
-                + "' '" + remote_host()
-		+ "' '" + changelog + "'"
-            )
-            out = os.popen(cmd)
-            output = 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_text += "Post-editing hook returned %d.\n" % rc
-            self.msg_text += 'Command was: ' + cmd + '\n'
+            self.msg_text += 'Command was: ' + ' '.join(cmd) + '\n'
             if output:
                 self.msg_text += 'Output follows:\n' + output
         else:
@@ -804,25 +800,15 @@ def main():
             handler(form[cmd].value)
             break
     else:
-        path_info = os.environ.get('PATH_INFO', '')
-        if len(path_info) and path_info[0] == '/':
-            query = path_info[1:] or 'FrontPage'
-        else:
-            query = os.environ.get('QUERY_STRING', '') or 'FrontPage'
-
+        query = query_string()
         if file_re.match(query):
-            if word_re.match(query):
-                Page(query).format()
+            # 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:
-                from mimetypes import MimeTypes
-                mimetype, encoding = MimeTypes().guess_type(query)
-                if mimetype:
-                    if mimetype.startswith('image/'):
-                        Page(query).send_image(mimetype=mimetype, args=form)
-                    else:
-                        Page(query).send_raw(mimetype=mimetype)
-                else:
-                    Page(query).format()
+                Page(query).format()
         else:
             send_httperror("403 Forbidden", query)