From 6260572237fd99265ceaa4ad1588f052e8056b8b Mon Sep 17 00:00:00 2001
From: Michael Murtaugh <mm@automatist.org>
Date: Thu, 27 Aug 2015 17:02:05 +0200
Subject: [PATCH] merged roels unicsss.py code

---
 etherdump  | 25 ++++++++++++++++++++++---
 linkify.py | 52 ++++++++++++++++++++++++++++++----------------------
 2 files changed, 52 insertions(+), 25 deletions(-)

diff --git a/etherdump b/etherdump
index f589d2c..96636d4 100755
--- a/etherdump
+++ b/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:
@@ -245,7 +250,15 @@ while len(todo) > 0:
         # extract the style tag (with authorship colors)
         style = t.find(".//style")
         if style != None:
-            style = ET.tostring(style, method="html")
+            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 = ""
         # and extract the contents of the body
@@ -273,4 +286,10 @@ while len(todo) > 0:
         if args.limit and count >= args.limit:
             break
     except TypeError:
-        print ("ERROR, skipping!", file=sys.stderr)
\ No newline at end of file
+        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')
\ No newline at end of file
diff --git a/linkify.py b/linkify.py
index 3609b71..981af11 100644
--- a/linkify.py
+++ b/linkify.py
@@ -3,39 +3,47 @@ import re, sys
 
 
 def strip_tags (text):
-	return re.sub(r"<.*?>", "", text)
+    return re.sub(r"<.*?>", "", text)
 
 def urlify (t, ext=".html"):
-	return t.replace(" ", "_") + ext
+    return t.replace(" ", "_") + ext
 
 def filename_to_padid (t):
-	t = t.replace("_", " ")
-	t = re.sub(r"\.html$", "", t)
-	return t
+    t = t.replace("_", " ")
+    t = re.sub(r"\.html$", "", 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 = []
+    collect = []
 
-	def s (m):
-		contents = strip_tags(m.group(1))
-		collect.append(contents)
-		link = urlify(contents)
-		# link = link.split("?", 1)[0]
-		return "[[<a class=\"wikilink\" href=\"{0}\">{1}</a>]]".format(link, contents)
+    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]
+        return "[[<a class=\"wikilink\" href=\"{0}\">{1}</a>]]".format(link, contents)
 
-	# 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)
-	return (src, collect)
+    # 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)
+    return (src, collect)
 
 
 if __name__ == "__main__":
-	src = sys.stdin.read()
-	src, links = linkify(src)
+    src = sys.stdin.read()
+    src, links = linkify(src)
 
-	for l in links:
-		print (l)
+    for l in links:
+        print (l)
 
-	print (src)
+    print (src)