# HG changeset patch
# User Paul Boddie <paul@boddie.org.uk>
# Date 1370864729 -7200
# Node ID 4a10cbd14a491bc903e2e3607f004067acc1a931
# Parent  589001c9539edb37d32f29c16c708a160735f3eb
Introduced page title truncation in links in order to find translated pages.

diff -r 589001c9539e -r 4a10cbd14a49 TO_DO.txt
--- a/TO_DO.txt	Mon Jun 10 13:20:27 2013 +0200
+++ b/TO_DO.txt	Mon Jun 10 13:45:29 2013 +0200
@@ -4,15 +4,6 @@
   that would cause conflicts in the generated Moin markup should be translated
   with spaces separating them
 
-DOC/How do I turn off HTML messages/attachments? (4030675)
-
-  Link targets need shortening for long page names:
-  4.22 Messages that are translated from HTML to plain text are empty or have a body like ""-root-8e8Ta4- Permission denied"
-
-DOC/Where do I go for help? (4816921)
-
-  XHTML: need to combine ac:anchor with ri:content-title to produce a complete link
-
 Handle page renaming (add redirects, perhaps support comments on old revisions):
 
   DOC/Making Sure Your Lists Are Private (10715350) renamed from
diff -r 589001c9539e -r 4a10cbd14a49 wikiparser.py
--- a/wikiparser.py	Mon Jun 10 13:20:27 2013 +0200
+++ b/wikiparser.py	Mon Jun 10 13:45:29 2013 +0200
@@ -430,6 +430,7 @@
                 prefix = ""
                 space, rest = target.split(":", 1)
                 if space not in URL_SCHEMES:
+                    rest = get_page_title(rest)
                     target = "%s/%s" % (space, rest)
 
             # Detect anchors.
@@ -452,6 +453,8 @@
                 if not label:
                     label = target
 
+                target = get_page_title(target)
+
             if not label and not title:
                 return "[[%s%s]]" % (prefix, target)
             elif not title:
diff -r 589001c9539e -r 4a10cbd14a49 xmlparser.py
--- a/xmlparser.py	Mon Jun 10 13:20:27 2013 +0200
+++ b/xmlparser.py	Mon Jun 10 13:45:29 2013 +0200
@@ -282,12 +282,23 @@
             for attrname in link_target_tags[name]:
                 attrvalue = self.attributes[-1].get(attrname)
                 if attrvalue:
+
+                    # Obtain a link label.
+
+                    if attrname in link_label_attributes and not self.label:
+                        self.label = attrvalue
+
+                    # Validate any page title.
+
+                    if attrname == "ri:content-title":
+                        attrvalue = get_page_title(attrvalue)
                     target_details.append(attrvalue)
+
+                    # Insert any prefix required for the link.
+
                     prefix = link_target_prefixes.get(attrname)
                     if prefix:
                         target_details.insert(0, prefix)
-                    if attrname in link_label_attributes and not self.label:
-                        self.label = attrvalue
 
             # Make a link based on the details.