merged roels unicsss.py code

etherdump

@@ -1,4 +1,7 @@
 #!/usr/bin/env python
+# License: AGPL
+#
+
 from __future__ import print_function
 # stdlib
 import json, sys, os, re
@@ -47,9 +50,10 @@ p.add_argument("--showurls", default=False, action="store_true", help="flag to d
 p.add_argument("--hidepaths", default=False, action="store_true", help="flag to not display paths")
 p.add_argument("--pretend", default=False, action="store_true", help="flag to not actually save")
 p.add_argument("--add-images", default=False, action="store_true", help="flag to add image tags")
+p.add_argument("--authors-css", default="authors.css", help="filename to save collected authorship css (nb: etherdump will overwrite this file!)")
 
 # TODO css from pad --- ie specify a padid for a stylesheet!!!!!!
-p.add_argument("--css", default="styles.css", help="padid of stylesheet")
+# p.add_argument("--css", default="styles.css", help="padid of stylesheet")
 
 
 args = p.parse_args()
@@ -77,6 +81,7 @@ todo = args.padid
 done = set()
 count = 0
 data = {}
+authors_css_rules = {}
 data['apikey'] = info['apikey']
 
 if args.allpads:
@@ -245,6 +250,14 @@ while len(todo) > 0:
         # extract the style tag (with authorship colors)
         style = t.find(".//style")
         if style != None:
+            if args.authors_css:
+                for i in style.text.splitlines(): 
+                    if len(i):
+                        selector, rule = i.split(' ',1)
+                        authors_css_rules[selector] = rule
+                style = '' # strip the individual style tag from each page (only exports to authors-css file)
+                # nb: it's up to the template to refer to the authors-css file
+            else:
                 style = ET.tostring(style, method="html")
         else:
             style = ""
@@ -274,3 +287,9 @@ while len(todo) > 0:
             break
     except TypeError:
         print ("ERROR, skipping!", file=sys.stderr)
+
+# Write the unified CSS with authors
+if args.authors_css:
+    with open(args.authors_css, 'w') as css:
+        for selector, rule in sorted(authors_css_rules.items()):
+            css.write(selector+' '+rule+'\n')

linkify.py

@@ -13,12 +13,20 @@ def filename_to_padid (t):
     t = re.sub(r"\.html$", "", t)
     return t
 
+def normalize_pad_name (n):
+    if '?' in n:
+        n = n.split('?', 1)[0]
+    if '/' in n:
+        n = n.split('/', 1)[0]
+    return n
+
 def linkify (src, urlify=urlify):
 
     collect = []
 
     def s (m):
         contents = strip_tags(m.group(1))
+        contents = normalize_pad_name(contents)
         collect.append(contents)
         link = urlify(contents)
         # link = link.split("?", 1)[0]
@@ -27,7 +35,7 @@ def linkify (src, urlify=urlify):
     # src = re.sub(r"\[\[([\w_\- ,]+?)\]\]", s, src)
     ## question marks are ignored by etherpad, so split/strip it
     ## strip slashes as well!! (/timeslider)
-	src = re.sub(r"\[\[(.+?)(\/.*)?(\?.*)?\]\]", s, src)
+    src = re.sub(r"\[\[(.+?)\]\]", s, src)
     return (src, collect)