utf8 0.1.5 → 0.1.6

data/ext/utf8/string_utf8.c

@@ -350,12 +350,13 @@ static VALUE rb_cString_UTF8_clean(VALUE self) {
   replace = REPLACEMENT_CHAR;
   out = xmalloc(len);
 
-  for(i=0; i<len; i++) {
+  for(i=0; i<len; i+=curCharLen) {
     curCharLen = utf8CharLen(str+i, len);
     if (curCharLen < 0) {
       *(out+i) = replace;
+      curCharLen = 1;
     } else {
-      *(out+i) = *(str+i);
+      memcpy(out+i, str+i, curCharLen);
     }
   }
 

data/lib/utf8/version.rb

@@ -1,5 +1,5 @@
 class String
   class UTF8 < ::String
-    VERSION = "0.1.5"
+    VERSION = "0.1.6"
   end
 end

data/spec/string_spec.rb

@@ -56,6 +56,10 @@ describe String::UTF8 do
     orig.as_utf8.clean.should eql(clean)
   end
 
+  it "clean should not replace valid utf8 chars with '?'" do
+    '→'.as_utf8.clean.should eql('→')
+  end
+
   context "#length and #size" do
     it "should be utf8-aware" do
       @utf8.length.should eql(@utf8_len)

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: utf8
 version: !ruby/object:Gem::Version 
-  hash: 17
+  hash: 23
   prerelease: 
   segments: 
   - 0
   - 1
-  - 5
-  version: 0.1.5
+  - 6
+  version: 0.1.6
 platform: ruby
 authors: 
 - Brian Lopez
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-07-21 00:00:00 -07:00
+date: 2011-07-25 00:00:00 -04:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -129,7 +129,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: 
-rubygems_version: 1.3.10
+rubygems_version: 1.6.2
 signing_key: 
 specification_version: 3
 summary: A lightweight UTF8-aware String class meant for use with Ruby 1.8