From d3e79ea84af243d066bcd52bf15eb711baa9f210 Mon Sep 17 00:00:00 2001
From: Jacques Distler <distler@golem.ph.utexas.edu>
Date: Mon, 14 Dec 2009 17:41:28 -0600
Subject: [PATCH] Make truncate() Unicode-aware

---
 app/helpers/application_helper.rb       | 13 ++++++++-----
 app/views/wiki/list.rhtml               |  6 +++---
 lib/stringsupport.rb                    | 25 +++++++++++++++++++++++++
 test/functional/wiki_controller_test.rb |  6 +++++-
 4 files changed, 41 insertions(+), 9 deletions(-)

diff --git a/app/helpers/application_helper.rb b/app/helpers/application_helper.rb
index 893faa70..62ce7a22 100644
--- a/app/helpers/application_helper.rb
+++ b/app/helpers/application_helper.rb
@@ -1,5 +1,6 @@
 # The methods added to this helper will be available to all templates in the application.
 module ApplicationHelper
+require 'stringsupport'
 
   # Accepts a container (hash, array, enumerable, your type) and returns a string of option tags. Given a container 
   # where the elements respond to first and last (such as a two-element array), the "lasts" serve as option values and
@@ -111,15 +112,17 @@ module ApplicationHelper
     PageRenderer.new(page.revisions.last).display_content
   end
 
-  def truncate(text, length = 30, truncate_string = '...')
-    return text if text.length <= length
-    len = length - truncate_string.length
+  def truncate(text, *args)
+    options = args.extract_options!
+    options.reverse_merge!(:length => 30, :omission => "...")
+    return text if text.num_chars <= options[:length]
+    len = options[:length] - options[:omission].as_utf8.num_chars
     t = ''
     text.split.collect do |word|
-      if t.length + word.length <= len
+      if t.num_chars + word.num_chars <= len
         t << word + ' '
       else 
-        return t.chop + truncate_string
+        return t.chop + options[:omission]
       end
     end
   end
diff --git a/app/views/wiki/list.rhtml b/app/views/wiki/list.rhtml
index fb42a0bd..12dc2571 100644
--- a/app/views/wiki/list.rhtml
+++ b/app/views/wiki/list.rhtml
@@ -13,7 +13,7 @@
 <ul>
   <%- @pages_in_category.each do |page| -%>
     <li>
-      <%= link_to_existing_page page, truncate(page.plain_name, 35) %>
+      <%= link_to_existing_page page, truncate(page.plain_name, :length => 35) %>
     </li>
   <%- end -%>
 </ul>
@@ -37,7 +37,7 @@
   <ul style="margin-bottom: 10px">
   <%- @page_names_that_are_wanted.each do |wanted_page_name| -%>
     <li>
-      <%= link_to_page(wanted_page_name, @web, truncate(WikiWords.separate(wanted_page_name), 35)) %>
+      <%= link_to_page(wanted_page_name, @web, truncate(WikiWords.separate(wanted_page_name), :length => 35)) %>
       wanted by
       <%= @web.select.pages_that_reference(wanted_page_name).collect { |referring_page|            
             link_to_existing_page referring_page
@@ -57,7 +57,7 @@
   <ul style="margin-bottom: 35px">
     <%- @pages_that_are_orphaned.each do |orphan_page| -%>
       <li>
-        <%= link_to_existing_page orphan_page, truncate(orphan_page.plain_name, 35) %>
+        <%= link_to_existing_page orphan_page, truncate(orphan_page.plain_name, :length => 35) %>
       </li>
     <%- end -%>
   </ul>
diff --git a/lib/stringsupport.rb b/lib/stringsupport.rb
index 658283dd..29d763ef 100644
--- a/lib/stringsupport.rb
+++ b/lib/stringsupport.rb
@@ -2,8 +2,30 @@
 
 class String
 
+# Return the number of unicode characters in a string
+#
+# :call-seq:
+#    string.num_chars    -> integer
+#
+# Because Rails 2.3.5's String#mb_chars.length is broken,
+# we provide this method.
+#--
+if "".respond_to?(:force_encoding)
+  def num_chars
+    length
+  end
+else
+  def num_chars
+    unpack('U*').length
+  end
+end
+
+#++
 # A method to allow byte-oriented operations in both Ruby 1.8 and Ruby 1.9
 #
+# :call-seq:
+#    string.to_utf_8 -> string (with the encoding set to "ASCII-8BIT")
+#
 # Under 1.8, this is a NOOP. Under 1.9, it sets the encoding to "ASCII-8BIT"
 #--
 if "".respond_to?(:force_encoding)
@@ -19,6 +41,9 @@ end
 #++
 # A method to allow string-oriented operations in both Ruby 1.8 and Ruby 1.9
 #
+# :call-seq:
+#    string.to_utf_8 -> string (with the encoding set to "UTF-8")
+#
 # Under 1.8, this is a NOOP. Under 1.9, it sets the encoding to "UTF-8"
 #--
 if "".respond_to?(:force_encoding)
diff --git a/test/functional/wiki_controller_test.rb b/test/functional/wiki_controller_test.rb
index 7b54a359..bdb47ef8 100644
--- a/test/functional/wiki_controller_test.rb
+++ b/test/functional/wiki_controller_test.rb
@@ -40,8 +40,12 @@ class WikiControllerTest < ActionController::TestCase
   def test_truncate_page_name
     wanted_page_name = 'This is a very, very, very, very, VERY long page name'
     evil_page_name = 'This page has plenty of fun &amp; games'
-    assert_equal 'This is a very, very, very,...',  @controller.truncate(WikiWords.separate(wanted_page_name), 35)
+    unicode_page_name = "This p\xF0\x9D\x94\xB8\xF0\x9D\x94\xBE\xF0\x9D\x94\xBC has plenty of fun &amp; games"
+    assert_equal 'This is a very, very, very,...',  @controller.truncate(WikiWords.separate(wanted_page_name), :length => 35)
     assert_equal 'This page has plenty of fun...',  @controller.truncate(WikiWords.separate(evil_page_name))
+    truncated = "".respond_to?(:force_encoding) ? "This p\u{1D538}\u{1D53E}\u{1D53C} has plenty of fun\u2633\u2633" :
+                 "This p\xF0\x9D\x94\xB8\xF0\x9D\x94\xBE\xF0\x9D\x94\xBC has plenty of fun\xE2\x98\xB3\xE2\x98\xB3"
+    assert_equal truncated, @controller.truncate(WikiWords.separate(unicode_page_name), :omission =>"\xE2\x98\xB3\xE2\x98\xB3")
   end
 
   def test_authenticate