escape_utils 0.1.2 → 0.1.3

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 # Changelog
 
+## 0.1.3 (June 9th, 2010)
+* cleaned some code up, removing duplication
+* moved to a more flexible character encoding scheme using Encoding.defaut_internal for 1.9 users
+
 ## 0.1.2 (June 8th, 2010)
 * forgot to add the ActionView monkey patch for JS escaping ;)
 

data/README.rdoc

@@ -2,7 +2,9 @@
 
 Being as though we're all html escaping everything these days, why not make it faster?
 
-At the moment escape_utils supports escaping and unescaping of HTML, and Javascript but I wanna add URL encoding soon
+At the moment escape_utils supports escaping and unescaping of HTML, and Javascript but I wanna add URL encoding soon.
+
+For character encoding in 1.9, we'll return strings in whatever Encoding.default_internal is set to or utf-8 otherwise.
 
 It has monkey-patches for Rack::Utils, CGI, ERB::Util and Haml and ActionView so you can drop this in and have your app start escaping fast as balls in no time
 
@@ -45,8 +47,9 @@ It has monkey-patches for Rack::Utils, CGI, ERB::Util and Haml and ActionView so
 
 == Benchmarks
 
-In my testing, escaping is around 10-20x faster than the pure ruby implementations in wide use today.
-While unescaping is around 20-40x faster than CGI.unescapeHTML - also pure ruby.
+In my testing, escaping html is around 10-20x faster than the pure ruby implementations in wide use today.
+While unescaping html is around 20-40x faster than CGI.unescapeHTML - also pure ruby.
+Escaping Javascript is around 16-30x faster.
 
 This output is from my laptop using the benchmark scripts in the benchmarks folder.
 

data/VERSION

@@ -1 +1 @@
-0.1.2
+0.1.3

data/escape_utils.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = %q{escape_utils}
-  s.version = "0.1.2"
+  s.version = "0.1.3"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Brian Lopez"]
-  s.date = %q{2010-06-08}
+  s.date = %q{2010-06-09}
   s.email = %q{seniorlopez@gmail.com}
   s.extensions = ["ext/extconf.rb"]
   s.extra_rdoc_files = [

data/ext/escape_utils.c

@@ -1,4 +1,8 @@
 #include <ruby.h>
+#ifdef HAVE_RUBY_ENCODING_H
+#include <ruby/encoding.h>
+static rb_encoding *utf8Encoding;
+#endif
 
 #define APPEND_BUFFER(escape, len, scoot_by)  \
   memcpy(&out[total], &in[offset], i-offset); \
@@ -6,18 +10,17 @@
   offset = i+scoot_by;                        \
   memcpy(&out[total], escape, len);           \
   total += len;                               \
-  break;                                      \
 
 static size_t escape_html(unsigned char *out, const unsigned char *in, size_t in_len) {
   size_t i = 0, offset = 0, total = 0;
 
   for(;i<in_len;i++) {
     switch(in[i]) {
-      case '&':  APPEND_BUFFER("&amp;",  5, 1);
-      case '<':  APPEND_BUFFER("&lt;",   4, 1);
-      case '>':  APPEND_BUFFER("&gt;",   4, 1);
-      case '\'': APPEND_BUFFER("&#39;",  5, 1);
-      case '\"': APPEND_BUFFER("&quot;", 6, 1);
+      case '&':  APPEND_BUFFER("&amp;",  5, 1); break;
+      case '<':  APPEND_BUFFER("&lt;",   4, 1); break;
+      case '>':  APPEND_BUFFER("&gt;",   4, 1); break;
+      case '\'': APPEND_BUFFER("&#39;",  5, 1); break;
+      case '\"': APPEND_BUFFER("&quot;", 6, 1); break;
     }
   }
 
@@ -31,38 +34,26 @@ static size_t unescape_html(unsigned char *out, const unsigned char *in, size_t
   size_t i = 0, offset = 0, total = 0;
   
   for(;i<in_len;i++) {
-    switch(in[i]) {
-      case '&':
-        if (i+5 <= in_len) {
-          if (memcmp(&in[i], "&amp;", 5) == 0) {
-            APPEND_BUFFER("&", 1, 5);
-          } else if (memcmp(&in[i], "&lt;", 4) == 0) {
-            APPEND_BUFFER("<", 1, 4);
-          } else if (memcmp(&in[i], "&gt;", 4) == 0) {
-            APPEND_BUFFER(">", 1, 4);
-          } else if (memcmp(&in[i], "&#39;", 5) == 0) {
-            APPEND_BUFFER("\'", 1, 5);
-          } else if (memcmp(&in[i], "&quot;", 6) == 0) {
-            APPEND_BUFFER("\"", 1, 6);
-          }
-        } else if (i+4 <= in_len) {
-          if (memcmp(&in[i], "&amp;", 5) == 0) {
-            APPEND_BUFFER("&", 1, 5);
-          } else if (memcmp(&in[i], "&lt;", 4) == 0) {
-            APPEND_BUFFER("<", 1, 4);
-          } else if (memcmp(&in[i], "&gt;", 4) == 0) {
-            APPEND_BUFFER(">", 1, 4);
-          } else if (memcmp(&in[i], "&#39;", 5) == 0) {
-            APPEND_BUFFER("\'", 1, 5);
-          }
-        } else if (i+3 <= in_len) {
-          if (memcmp(&in[i], "&lt;", 4) == 0) {
-            APPEND_BUFFER("<", 1, 4);
-          } else if (memcmp(&in[i], "&gt;", 4) == 0) {
-            APPEND_BUFFER(">", 1, 4);
-          }
+    if(in[i] == '&') {
+      if (i+3 <= in_len) {
+        if (memcmp(&in[i], "&lt;", 4) == 0) {
+          APPEND_BUFFER("<", 1, 4);
+        } else if (memcmp(&in[i], "&gt;", 4) == 0) {
+          APPEND_BUFFER(">", 1, 4);
+        }
+      }
+      if (i+4 <= in_len) {
+        if (memcmp(&in[i], "&amp;", 5) == 0) {
+          APPEND_BUFFER("&", 1, 5);
+        } else if (memcmp(&in[i], "&#39;", 5) == 0) {
+          APPEND_BUFFER("\'", 1, 5);
         }
-      break;
+      }
+      if (i+5 <= in_len) {
+        if (memcmp(&in[i], "&quot;", 6) == 0) {
+          APPEND_BUFFER("\"", 1, 6);
+        }
+      }
     }
   }
 
@@ -77,7 +68,7 @@ static size_t escape_javascript(unsigned char *out, const unsigned char *in, siz
 
   for(;i<in_len;i++) {
     switch(in[i]) {
-      case '\\':  APPEND_BUFFER("\\\\", 2, 1);
+      case '\\':  APPEND_BUFFER("\\\\", 2, 1); break;
       case '<':
         if (i+1 <= in_len && in[i+1] == '/') {
           APPEND_BUFFER("<\\/", 3, 2);
@@ -90,9 +81,9 @@ static size_t escape_javascript(unsigned char *out, const unsigned char *in, siz
           APPEND_BUFFER("\\n", 2, 1);
         }
         break;
-      case '\n': APPEND_BUFFER("\\n", 2, 1);
-      case '\"': APPEND_BUFFER("\\\"", 2, 1);
-      case '\'': APPEND_BUFFER("\\'", 2, 1);
+      case '\n': APPEND_BUFFER("\\n", 2, 1); break;
+      case '\"': APPEND_BUFFER("\\\"", 2, 1); break;
+      case '\'': APPEND_BUFFER("\\'", 2, 1); break;
     }
   }
 
@@ -106,6 +97,10 @@ static VALUE rb_escape_html(VALUE self, VALUE str) {
   Check_Type(str, T_STRING);
 
   VALUE rb_output_buf;
+#ifdef HAVE_RUBY_ENCODING_H
+  rb_encoding *default_internal_enc = rb_default_internal_encoding();
+  rb_encoding *original_encoding = rb_enc_get(str);
+#endif
   unsigned char *inBuf = (unsigned char*)RSTRING_PTR(str);
   size_t len = RSTRING_LEN(str), new_len = 0;
 
@@ -122,6 +117,14 @@ static VALUE rb_escape_html(VALUE self, VALUE str) {
   // free the temporary C string
   free(outBuf);
 
+#ifdef HAVE_RUBY_ENCODING_H
+  rb_enc_associate(rb_output_buf, original_encoding);
+  if (default_internal_enc) {
+    rb_output_buf = rb_str_export_to_enc(rb_output_buf, default_internal_enc);
+  } else {
+    rb_output_buf = rb_str_export_to_enc(rb_output_buf, utf8Encoding);
+  }
+#endif
   return rb_output_buf;
 }
 
@@ -129,6 +132,10 @@ static VALUE rb_unescape_html(VALUE self, VALUE str) {
   Check_Type(str, T_STRING);
 
   VALUE rb_output_buf;
+#ifdef HAVE_RUBY_ENCODING_H
+  rb_encoding *default_internal_enc = rb_default_internal_encoding();
+  rb_encoding *original_encoding = rb_enc_get(str);
+#endif
   unsigned char *inBuf = (unsigned char*)RSTRING_PTR(str);
   size_t len = RSTRING_LEN(str), new_len = 0;
 
@@ -145,6 +152,14 @@ static VALUE rb_unescape_html(VALUE self, VALUE str) {
   // free the temporary C string
   free(outBuf);
 
+#ifdef HAVE_RUBY_ENCODING_H
+  rb_enc_associate(rb_output_buf, original_encoding);
+  if (default_internal_enc) {
+    rb_output_buf = rb_str_export_to_enc(rb_output_buf, default_internal_enc);
+  } else {
+    rb_output_buf = rb_str_export_to_enc(rb_output_buf, utf8Encoding);
+  }
+#endif
   return rb_output_buf;
 }
 
@@ -156,6 +171,10 @@ static VALUE rb_escape_javascript(VALUE self, VALUE str) {
   Check_Type(str, T_STRING);
 
   VALUE rb_output_buf;
+#ifdef HAVE_RUBY_ENCODING_H
+  rb_encoding *default_internal_enc = rb_default_internal_encoding();
+  rb_encoding *original_encoding = rb_enc_get(str);
+#endif
   unsigned char *inBuf = (unsigned char*)RSTRING_PTR(str);
   size_t len = RSTRING_LEN(str), new_len = 0;
 
@@ -172,6 +191,14 @@ static VALUE rb_escape_javascript(VALUE self, VALUE str) {
   // free the temporary C string
   free(outBuf);
 
+#ifdef HAVE_RUBY_ENCODING_H
+  rb_enc_associate(rb_output_buf, original_encoding);
+  if (default_internal_enc) {
+    rb_output_buf = rb_str_export_to_enc(rb_output_buf, default_internal_enc);
+  } else {
+    rb_output_buf = rb_str_export_to_enc(rb_output_buf, utf8Encoding);
+  }
+#endif
   return rb_output_buf;
 }
 
@@ -184,4 +211,8 @@ void Init_escape_utils_ext() {
   rb_define_module_function(mEscape,  "unescape_html",  rb_unescape_html, 1);
   rb_define_method(mEscape,           "escape_javascript",  rb_escape_javascript, 1);
   rb_define_module_function(mEscape,  "escape_javascript",  rb_escape_javascript, 1);
+
+#ifdef HAVE_RUBY_ENCODING_H
+  utf8Encoding = rb_utf8_encoding();
+#endif
 }

data/lib/escape_utils.rb

@@ -4,5 +4,5 @@ require 'escape_utils_ext'
 
 EscapeUtils.send(:extend, EscapeUtils)
 module EscapeUtils
-  VERSION = "0.1.2"
+  VERSION = "0.1.3"
 end

data/spec/html/escape_spec.rb

@@ -21,4 +21,18 @@ describe EscapeUtils, "escape_html" do
   it "should escape the & character" do
     EscapeUtils.escape_html("<b>Bourbon & Branch</b>").should eql("&lt;b&gt;Bourbon &amp; Branch&lt;/b&gt;")
   end
+
+  if RUBY_VERSION =~ /^1.9/
+    it "should default to utf-8 if Encoding.default_internal is nil" do
+      Encoding.default_internal = nil
+      EscapeUtils.escape_html("<b>Bourbon & Branch</b>").encoding.should eql(Encoding.find('utf-8'))
+    end
+
+    it "should use Encoding.default_internal" do
+      Encoding.default_internal = Encoding.find('utf-8')
+      EscapeUtils.escape_html("<b>Bourbon & Branch</b>").encoding.should eql(Encoding.default_internal)
+      Encoding.default_internal = Encoding.find('us-ascii')
+      EscapeUtils.escape_html("<b>Bourbon & Branch</b>").encoding.should eql(Encoding.default_internal)
+    end
+  end
 end

data/spec/html/unescape_spec.rb

@@ -21,4 +21,18 @@ describe EscapeUtils, "unescape_html" do
   it "should unescape the & character" do
     EscapeUtils.unescape_html("&lt;b&gt;Bourbon &amp; Branch&lt;/b&gt;").should eql("<b>Bourbon & Branch</b>")
   end
+
+  if RUBY_VERSION =~ /^1.9/
+    it "should default to utf-8 if Encoding.default_internal is nil" do
+      Encoding.default_internal = nil
+      EscapeUtils.unescape_html("&lt;b&gt;Bourbon &amp; Branch&lt;/b&gt;").encoding.should eql(Encoding.find('utf-8'))
+    end
+
+    it "should use Encoding.default_internal" do
+      Encoding.default_internal = Encoding.find('utf-8')
+      EscapeUtils.unescape_html("&lt;b&gt;Bourbon &amp; Branch&lt;/b&gt;").encoding.should eql(Encoding.default_internal)
+      Encoding.default_internal = Encoding.find('us-ascii')
+      EscapeUtils.unescape_html("&lt;b&gt;Bourbon &amp; Branch&lt;/b&gt;").encoding.should eql(Encoding.default_internal)
+    end
+  end
 end

data/spec/javascript/escape_spec.rb

@@ -10,7 +10,7 @@ describe EscapeUtils, "escape_javascript" do
   it "should return an empty string if passed nil" do
     EscapeUtils.escape_javascript(nil).should eql("")
   end
-  
+
   it "should escape quotes and newlines" do
     EscapeUtils.escape_javascript(%(This "thing" is really\n netos')).should eql(%(This \\"thing\\" is really\\n netos\\'))
   end
@@ -22,4 +22,18 @@ describe EscapeUtils, "escape_javascript" do
   it "should escape closed html tags" do
     EscapeUtils.escape_javascript(%(dont </close> tags)).should eql(%(dont <\\/close> tags))
   end
+
+  if RUBY_VERSION =~ /^1.9/
+    it "should default to utf-8 if Encoding.default_internal is nil" do
+      Encoding.default_internal = nil
+      EscapeUtils.escape_javascript(%(dont </close> tags)).encoding.should eql(Encoding.find('utf-8'))
+    end
+
+    it "should use Encoding.default_internal" do
+      Encoding.default_internal = Encoding.find('utf-8')
+      EscapeUtils.escape_javascript(%(dont </close> tags)).encoding.should eql(Encoding.default_internal)
+      Encoding.default_internal = Encoding.find('us-ascii')
+      EscapeUtils.escape_javascript(%(dont </close> tags)).encoding.should eql(Encoding.default_internal)
+    end
+  end
 end

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 0
   - 1
-  - 2
-  version: 0.1.2
+  - 3
+  version: 0.1.3
 platform: ruby
 authors: 
 - Brian Lopez
@@ -14,7 +14,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-06-08 00:00:00 -07:00
+date: 2010-06-09 00:00:00 -07:00
 default_executable: 
 dependencies: []