More improvements to mediawiki fragment escaping

It turns out that the PHP urlencode encodes more characters than
the ruby URI.encode does by default, so specify our own characters
to encode based on http://php.net/manual/en/function.urlencode.php
with the extra exclusion of colon, which the mediawiki code reverses
after the encoding is done.

Fixes #967

diff --git a/app/helpers/browse_helper.rb b/app/helpers/browse_helper.rb
index a2d0e440ddbd9700667544f60b9a46fb32e5c4cf..5d0a18fd96373b86d9705a9e7514a2883b8b9fe2 100644 (file)
--- a/app/helpers/browse_helper.rb
+++ b/app/helpers/browse_helper.rb
@@ -140,7 +140,7 @@ module BrowseHelper
       # Must break it up to correctly build the url
       value = $1
       section = "#" + $2
-      encoded_section = "#" + URI.encode($2.gsub(" ", "_")).gsub("%3A", ":").gsub("%", ".")
+      encoded_section = "#" + URI.encode($2.gsub(/ +/, "_"), /[^A-Za-z0-9:_]/).gsub("%", ".")
     else
       section = ""
       encoded_section = ""

diff --git a/test/helpers/browse_helper_test.rb b/test/helpers/browse_helper_test.rb
index c4f0e8c31d97c46248605bce0b7a298c6874861f..7878d10e082eddef78f32f4a3990a77cfde61284 100644 (file)
--- a/test/helpers/browse_helper_test.rb
+++ b/test/helpers/browse_helper_test.rb
@@ -196,6 +196,10 @@ class BrowseHelperTest < ActionView::TestCase
     assert_equal "http://de.wikipedia.org/wiki/de:Alte Brücke (Heidelberg)?uselang=en#Br.C3.BCckenaffe", link[:url]
     assert_equal "de:Alte Brücke (Heidelberg)#Brückenaffe", link[:title]
 
+    link = wikipedia_link("wikipedia", "de:Liste der Baudenkmäler in Eichstätt#Brückenstraße 1, Ehemaliges Bauernhaus")
+    assert_equal "http://de.wikipedia.org/wiki/de:Liste der Baudenkmäler in Eichstätt?uselang=en#Br.C3.BCckenstra.C3.9Fe_1.2C_Ehemaliges_Bauernhaus", link[:url]
+    assert_equal "de:Liste der Baudenkmäler in Eichstätt#Brückenstraße 1, Ehemaliges Bauernhaus", link[:title]
+
     I18n.locale = "pt-BR"
 
     link = wikipedia_link("wikipedia", "zh-classical:Test#Section")