html_press 0.6.5 → 0.6.6

data/Readme.md

@@ -70,34 +70,34 @@ bundle install
 bundle exec jekyll
 ```
 
-## Additional tools
-
-  - https://github.com/aberant/css-spriter
-  - https://github.com/sstephenson/sprockets
-  - https://github.com/documentcloud/jammit
-  - https://github.com/cjohansen/juicer
-  - https://github.com/alexdunae/w3c_validators
-
 ## TODO
 
+  - use parser ([鋸](https://github.com/tenderlove/nokogiri)) instead of regexp's
   - check if all utf-8 symbols, are smaller than html entities
   - ambigious ampersands
-  - bin
-  - Support other minifiers (Closure, YUI compressor)
+  - cli
+  - Support other js/css minifiers (Closure, YUI compressor)
   - htmlTydi
   - add examples of usage with Sinatra and Rails
-  - https://github.com/gfranco/jeanny
-  - https://github.com/aanand/deadweight
 
 ## Alternatives
 
-###Ruby
+  - [html-minifier](https://github.com/kangax/html-minifier) (js), [test suite](https://github.com/kangax/html-minifier/blob/gh-pages/tests/index.html), [ruby wrapper - html_minifier](https://github.com/stereobooster/html_minifier)
+  - [htmlcompressor](http://code.google.com/p/htmlcompressor/) (java), [test suite](http://code.google.com/p/htmlcompressor/source/browse/#svn%2Ftrunk%2Fsrc%2Ftest%2Fresources%2Fhtml%253Fstate%253Dclosed)
+  - PHPTal compress (php), [test suite](https://svn.motion-twin.com/phptal/trunk/tests/CompressTest.php)
+  - [W3 total cache](http://wordpress.org/extend/plugins/w3-total-cache/) - WP plugin from smashingmagazine contains html minifier (php)
 
-  - https://github.com/completelynovel/html_compressor
-  - https://github.com/MadRabbit/frontcompiler
+## Additional tools
 
-###Other
+  - https://github.com/gfranco/jeanny
+  - https://github.com/aanand/deadweight
+  - https://github.com/aberant/css-spriter
+  - https://github.com/sstephenson/sprockets
+  - https://github.com/documentcloud/jammit
+  - https://github.com/cjohansen/juicer
+  - https://github.com/alexdunae/w3c_validators
 
-  - http://code.google.com/p/htmlcompressor/
-  - smarty `strip` tag
-  - W3 total cache (WP plugin from smashingmagazine contains html minifier)
+## Resources
+  - http://perfectionkills.com/experimenting-with-html-minifier
+  - http://perfectionkills.com/optimizing-html
+  - https://developers.google.com/speed/articles/optimizing-html

data/html_press.gemspec

@@ -23,7 +23,7 @@ Gem::Specification.new do |s|
   s.add_development_dependency "rspec"
   s.add_development_dependency "rake"
   
-  s.add_runtime_dependency "css_press", ">=0.3.1"
+  s.add_runtime_dependency "css_press", ">=0.3.2"
   s.add_runtime_dependency "uglifier"
   s.add_runtime_dependency "htmlentities"
 end

data/lib/html_press/css_press.rb

@@ -4,7 +4,7 @@ module HtmlPress
     def self.css_compressor (text)
       text = "a{#{text}}"
       text = CssPress.press text
-      text.gsub(/^a\{/, '').gsub!(/\}$/, '')
+      text.gsub(/^a\{/, '').gsub(/\}$/, '')
     end
     def self.style_compressor (text)
       CssPress.press text

data/lib/html_press/html.rb

@@ -4,11 +4,15 @@ module HtmlPress
     DEFAULTS = {
       :logger => false,
       :unquoted_attributes => false,
-      :dump_empty_values => false 
+      :drop_empty_values => false 
     }
 
     def initialize (options = {})
       @options = DEFAULTS.merge(options)
+      if @options.keys.include? :dump_empty_values
+        @options[:drop_empty_values] = @options.delete(:dump_empty_values)
+        warn "dump_empty_values deprecated use drop_empty_values"
+      end
     end
 
     def log (text)
@@ -37,7 +41,10 @@ module HtmlPress
       # replace SCRIPTs (and minify) with placeholders
       out.gsub! /\s*(<script\b[^>]*?>[\s\S]*?<\/script>)\s*/i do |m|
         m.gsub!(/^\s+|\s+$/, '')
-        js = m.gsub(/\s*<script\b[^>]*?>([\s\S]*?)<\/script>\s*/i , "\\1")
+        m = m.gsub /^<script\s([^>]+)>/i do |m|
+          attrs(m, 'script', true)
+        end
+        js = m.gsub(/^<script\b[^>]*?>([\s\S]*?)<\/script>/i , "\\1")
         begin
           js_compressed = HtmlPress.js_compressor js
           m.gsub!(js, js_compressed)
@@ -50,7 +57,10 @@ module HtmlPress
       # replace STYLEs (and minify) with placeholders
       out.gsub! /\s*(<style\b[^>]*?>[\s\S]*?<\/style>)\s*/i do |m|
         m.gsub!(/^\s+|\s+$/, '')
-        css = m.gsub(/\s*<style\b[^>]*?>([\s\S]*?)<\/style>\s*/i, "\\1")
+        m = m.gsub /^<style\s([^>]+)>/i do |m|
+          attrs(m, 'style', true)
+        end
+        css = m.gsub(/^<style\b[^>]*?>([\s\S]*?)<\/style>/i, "\\1")
         begin
           css_compressed = HtmlPress.style_compressor css
           m.gsub!(css, css_compressed)
@@ -150,15 +160,17 @@ module HtmlPress
           attr k, "'", tag
         end
   
+        attributes_compressed = " " + attributes_compressed.strip
+
         if attributes_compressed == " /"
           attributes_compressed = "/"
         elsif attributes_compressed == " "
           attributes_compressed = ""
         end
-        return m.gsub(attributes, attributes_compressed)
+        m.gsub(attributes, attributes_compressed)
+      else
+        m
       end
-
-      return m
     end
 
     def attr(attribute, delimiter, tag)
@@ -170,19 +182,25 @@ module HtmlPress
       name = name_original.downcase
       tag_name = tag.downcase
 
+      if value.size > 0
+        re = "^=" + delimiter + "|" + delimiter + "$"
+        re = Regexp.new re
+        value_original.gsub!(re, "")
+      end
+
       case tag_name
       when "script"
-        if name == "type" || name == "language"
+        if (name == "type" && value_original == "text/javascript") || (name == "language" && value_original == "JavaScript")
           return ""
         elsif name == "async" || name == "defer"
           return name_original
         end
       when "form"
-        if name == "method" && (value == "=\"get\"" || value == "='get'")
+        if name == "method" && value_original == "get"
           return ""
         end
       when /link|style/
-        if name == "type"
+        if name == "type" && value_original == "text/stylesheet"
           return ""
         end
       when /input|textarea|button|select|option|optgroup/
@@ -196,7 +214,7 @@ module HtmlPress
           return name_original
         end
         if tag_name == "input"
-          if name == "type" && (value == "=\"text\"" || value == "='text'")
+          if name == "type" && value_original == "text"
             return ""
           end
           if name == "checked"
@@ -208,14 +226,20 @@ module HtmlPress
         end
       end
 
-      if value.size != 0
-
-        re = "^=" + delimiter + "|" + delimiter + "$"
-        re = Regexp.new re
-        value_original.gsub!(re, "")
+      if value.size > 0
 
         if name == "style"
-          value_original = HtmlPress.css_compressor value_original
+          begin
+            value_original = HtmlPress.css_compressor value_original
+            # TODO what about escaped attribute values?
+            if delimiter == "\""
+              value_original.gsub!("\"", "'")
+            else
+              value_original.gsub!("'", "\"")
+            end
+          rescue Exception => e
+            log e.message
+          end
         end
 
         if name == "class"
@@ -228,16 +252,21 @@ module HtmlPress
           onmouseout onkeypress onkeydown onkeyup]
 
         if events.include? name
-          value_original.gsub! /^javascript:\s+/, ''
-          value_original = HtmlPress.js_compressor value_original
-          if delimiter == "\""
-            value_original.gsub! "\"", "'"
+          value_original.gsub! /^javascript:\s+|;$/, ''
+          begin
+            value_original = HtmlPress.js_compressor value_original
+            # TODO what about escaped attribute values?
+            if delimiter == "\""
+              value_original.gsub! "\"", "'"
+            end
+          rescue Exception => e
+            log e.message
           end
         end
 
         if value_original.size == 0
           #attribute without value may be dropped by IE7
-          if @options[:dump_empty_values]
+          if @options[:drop_empty_values]
             attribute = name_original
           else
             attribute = name_original + "=" + delimiter + delimiter

data/lib/html_press/version.rb

@@ -1,3 +1,3 @@
 module HtmlPress
-  VERSION = "0.6.5"
+  VERSION = "0.6.6"
 end

data/spec/html_press_spec.rb

@@ -1,218 +1,220 @@
-# encoding: UTF-8
-
-require File.expand_path("../lib/html_press", File.dirname(__FILE__))
-
-class Lg
-  class << self
-    @warns = [] 
-    attr_accessor :warns
-    def log text
-      Lg.warns.push text
-    end
-  end
-end
-
-describe HtmlPress do
-  before :each do
-  end
-
-  it "should leave only one whitespace between inline tags" do
-    HtmlPress.press("<p>lorem  <b>ipsum</b>  <i>dolor</i>  </p>").should eql "<p>lorem <b>ipsum</b> <i>dolor</i></p>"
-  end
-
-  it "should leave no whitespaces between block tags" do
-    HtmlPress.press("<div></div> \t\r\n  <div></div>").should eql "<div></div><div></div>"
-    HtmlPress.press("<div>  <div> \t\r\n  </div>  </div>").should eql "<div><div></div></div>"
-  end
-
-  it "should leave only one whitespace in text" do
-    HtmlPress.press("<p>a  a</p>").should eql "<p>a a</p>"
-  end
-
-  it "should leave newlines in pre tags and remove trailing spaces" do
-    HtmlPress.press("<pre>a \t </pre>").should eql "<pre>a</pre>"
-    HtmlPress.press("<pre>qwe   \nasd   </pre>").should eql "<pre>qwe\nasd</pre>"
-  end
-
-  it "should leave textareas as is" do
-    text = "<textarea> \t </textarea>"
-    HtmlPress.press(text).should eql text
-  end
-
-  it "should compress js in script tags" do
-    script = "  (function(undefined){ \t\n var long_name = '  '; }())  \n \r"
-    pressed_script = "<script>" + HtmlPress.js_compressor(script) + "</script>"
-    script = "  <script>" + script + "</script>  "
-    HtmlPress.press(script).should eql pressed_script
-  end
-
-  it "should compress css in style tags" do
-    style = "  div { margin: 0px 0px; \n}  "
-    pressed_style = "<style>" + HtmlPress.style_compressor(style) + "</style>"
-    style = "  <style>" + style + "</style>  "
-    HtmlPress.press(style).should eql pressed_style
-  end
-
-  it "should remove html comments" do
-    HtmlPress.press("<p></p><!-- comment  --><p></p>").should eql "<p></p><p></p>"
-  end
-
-  it "should leave IE conditional comments" do
-    text = "<!--[if IE]><html class=\"ie\"><![endif]--><div></div>"
-    HtmlPress.press(text).should eql text
-  end
-
-  it "should work with special utf-8 symbols" do
-    HtmlPress.press("✪<p></p>  <p></p>").should eql "✪<p></p><p></p>"
-  end
-
-  it "should work with tags in upper case" do
-    HtmlPress.press("<P>  </p>").should eql "<P></p>"
-  end
-
-  it "should remove whitespaces between IE conditional comments" do
-    text = "<p></p>  <!--[if IE]><html class=\"ie\">      <![endif]--> <!--[if IE]><html class=\"ie1\"><![endif]-->"
-    text2 = "<p></p><!--[if IE]><html class=\"ie\"><![endif]--><!--[if IE]><html class=\"ie1\"><![endif]-->"
-    HtmlPress.press(text).should eql text2
-  end
-
-  it "should treat text inside IE conditional comments as it was without comments" do
-    text = "<div class=\"a\"   id=\"b\">    </div>    <p></p>"
-    text2 = HtmlPress.press(text)
-    text = "<!--[if IE]>" + text + "<![endif]-->"
-    text2 = "<!--[if IE]>" + text2 + "<![endif]-->"
-    HtmlPress.press(text).should eql text2
-    text = "<script> (function(undefined){ var a;}()) </script>"
-    text2 = HtmlPress.press(text)
-    text = "<!--[if IE]>" + text + "<![endif]-->"
-    text2 = "<!--[if IE]>" + text2 + "<![endif]-->"
-    HtmlPress.press(text).should eql text2
-  end
-
-  it "should remove unnecessary whitespaces inside tag" do
-    HtmlPress.press("<p class=\"a\"   id=\"b\"></p>").should eql "<p class=\"a\" id=\"b\"></p>"
-    HtmlPress.press("<p class=\"a\" ></p>").should eql "<p class=\"a\"></p>"
-    HtmlPress.press("<img src=\"\" />").should eql "<img src=\"\"/>"
-    HtmlPress.press("<br />").should eql "<br/>"
-  end
-
-  it "should work with 'badly' formatted attributes" do
-    HtmlPress.press("<p class='a'   id='b'></p>").should eql "<p class='a' id='b'></p>"
-    # HtmlPress.press("<p class = 'a'></p>").should eql "<p class='a'></p>"
-    # HtmlPress.press("<p class = a></p>").should eql "<p class=a></p>"
-  end
-
-  it "should optimize attributes" do
-    HtmlPress.press("<p class=\"a  b\"></p>").should eql "<p class=\"a b\"></p>"
-    # http(s):// to //
-  end
-
-  it "should compress css in style attributes" do
-    HtmlPress.press("<p style=\"display: none;\"></p>").should eql "<p style=\"display:none\"></p>"
-  end
-
-  it "should work with namespaces" do
-    text = "<html xmlns:og=\"http://ogp.me/ns#\" class=\"a b\"><og:like>like</og:like></html>"
-    HtmlPress.press(text).should eql text
-  end
-
-  it "should not modify input value" do
-    text = "<div>   </div>"
-    text1 = text.dup
-    HtmlPress.press(text).should_not eql text
-    text.should eql text1
-  end
-
-  it "should leave whitespaces inside other attributes" do
-    text = "<a onclick=\"alert('     ')\" unknown_attr='   a  a'>a</a>"
-    HtmlPress.press(text).should eql text
-  end
-
-  it "should report javascript errors" do
-    script_with_error = "<script>function(){</script>"
-    Lg.warns = []
-    HtmlPress.press(script_with_error, {:logger => Lg.method(:log)}).should eql script_with_error
-    Lg.warns.size.should eql 1
-  end
-
-  # it "should report css errors" do
-    # script_with_error = "<style>.clas{margin:</style>"
-    # l = Lg.new
-    # l.warns.size.should eql 0
-    # HtmlPress.press(script_with_error, {:logger => l}).should eql script_with_error
-    # l.warns.size.should eql 1
-  # end
-
-  it "should remove values of boolean attributes" do
-    HtmlPress.press("<option selected=\"selected\">a</option>").should eql "<option selected>a</option>"
-    HtmlPress.press("<input type=\"checkbox\" checked=\"checked\"/>").should eql "<input type=\"checkbox\" checked/>"
-    HtmlPress.press("<input type=\"radio\" checked=\"checked\"/>").should eql "<input type=\"radio\" checked/>"
-    # disabled            (input, textarea, button, select, option, optgroup)
-    HtmlPress.press("<input disabled=\"disabled\"/>").should eql "<input disabled/>"
-    # readonly            (input type=text/password, textarea)
-    HtmlPress.press("<input readonly=\"readonly\"/>").should eql "<input readonly/>"
-    # HtmlPress.press("<script src=\"example.com\" async=\"async\"></script>").should eql "<script src=\"example.com\" async></script>"
-    # HtmlPress.press("<script src=\"example.com\" defer=\"defer\"></script>").should eql "<script src=\"example.com\" async></script>"
-    # HtmlPress.press("<select multiple=\"multiple\"/>").should eql "<select multiple/>"
-    # ismap     isMap     (img, input type=image)
-    # declare             (object; never used)
-    # noresize  noResize  (frame)
-    # nowrap    noWrap    (td, th; deprecated)
-    # noshade   noShade   (hr; deprecated)
-    # compact             (ul, ol, dl, menu, dir; deprecated)
-  end
-
-  it "should remove attributes with default values" do
-    # HtmlPress.press("<script type=\"text/javascript\" language=\"JavaScript\">var a;</script>").should eql "<script>var a;</script>"
-    # HtmlPress.press("<style type=\"text/stylesheet\"></style>").should eql "<style></style>"
-    HtmlPress.press("<link type=\"text/stylesheet\"/>").should eql "<link/>"
-    HtmlPress.press("<form method=\"get\"></form>").should eql "<form></form>"
-    HtmlPress.press("<input type=\"text\"/>").should eql "<input/>"
-    # input value "" ?
-  end
-
-  it "should convert html entities to utf-8 symbols" do
-    HtmlPress.press("&lt; &#60; &gt; &#62; &amp; &#38;").should eql "&lt; &#60; &gt; &#62; &amp; &#38;"
-    HtmlPress.press("&eacute;lan").should eql "élan"
-  end
-
-  it "should remove unnecessary quotes for attribute values" do
-    HtmlPress.press("<img src=\"\">", {:unquoted_attributes => true}).should eql "<img src=\"\">"
-    HtmlPress.press("<p id=\"a\"></p>", {:unquoted_attributes => true}).should eql "<p id=a></p>"
-    text = "<p id=\"a b\"></p>"
-    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
-    text = "<p id=\"a=\"></p>"
-    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
-    text = "<p id=\"a'\"></p>"
-    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
-    text = "<p id=\"a`\"></p>"
-    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
-    text = "<p id='a\"'></p>"
-    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
-    text = "<p id=\"a\t\"></p>"
-    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
-  end
-
-  it "should remove empty attribute values" do
-    HtmlPress.press("<img src=\"\">", {:dump_empty_values => true}).should eql "<img src>"
-  end
-
-  it "should compress javascript in event attributes" do
-    HtmlPress.press("<a onclick=\"javacript: alert('  ');\"></a>").should eql "<a onclick=\"alert('  ')\"></a>"
-    # onfocus
-    # onblur
-    # onselect
-    # onchange
-    # onclick
-    # ondblclick
-    # onmousedown
-    # onmouseup
-    # onmouseover
-    # onmousemove
-    # onmouseout
-    # onkeypress
-    # onkeydown
-    # onkeyup
-  end
-
+# encoding: UTF-8
+
+require File.expand_path("../lib/html_press", File.dirname(__FILE__))
+
+class Lg
+  class << self
+    @warns = [] 
+    attr_accessor :warns
+    def log text
+      Lg.warns.push text
+    end
+  end
+end
+
+describe HtmlPress do
+  before :each do
+  end
+
+  it "should leave only one whitespace between inline tags" do
+    HtmlPress.press("<p>lorem  <b>ipsum</b>  <i>dolor</i>  </p>").should eql "<p>lorem <b>ipsum</b> <i>dolor</i></p>"
+  end
+
+  it "should leave no whitespaces between block tags" do
+    HtmlPress.press("<div></div> \t\r\n  <div></div>").should eql "<div></div><div></div>"
+    HtmlPress.press("<div>  <div> \t\r\n  </div>  </div>").should eql "<div><div></div></div>"
+  end
+
+  it "should leave only one whitespace in text" do
+    HtmlPress.press("<p>a  a</p>").should eql "<p>a a</p>"
+  end
+
+  it "should leave newlines in pre tags and remove trailing spaces" do
+    HtmlPress.press("<pre>a \t </pre>").should eql "<pre>a</pre>"
+    HtmlPress.press("<pre>qwe   \nasd   </pre>").should eql "<pre>qwe\nasd</pre>"
+  end
+
+  it "should leave textareas as is" do
+    text = "<textarea> \t </textarea>"
+    HtmlPress.press(text).should eql text
+  end
+
+  it "should compress js in script tags" do
+    script = "  (function(undefined){ \t\n var long_name = '  '; }())  \n \r"
+    pressed_script = "<script>" + HtmlPress.js_compressor(script) + "</script>"
+    script = "  <script>" + script + "</script>  "
+    HtmlPress.press(script).should eql pressed_script
+  end
+
+  it "should compress css in style tags" do
+    style = "  div { margin: 0px 0px; \n}  "
+    pressed_style = "<style>" + HtmlPress.style_compressor(style) + "</style>"
+    style = "  <style>" + style + "</style>  "
+    HtmlPress.press(style).should eql pressed_style
+  end
+
+  it "should remove html comments" do
+    HtmlPress.press("<p></p><!-- comment  --><p></p>").should eql "<p></p><p></p>"
+  end
+
+  it "should leave IE conditional comments" do
+    text = "<!--[if IE]><html class=\"ie\"><![endif]--><div></div>"
+    HtmlPress.press(text).should eql text
+  end
+
+  it "should work with special utf-8 symbols" do
+    HtmlPress.press("✪<p></p>  <p></p>").should eql "✪<p></p><p></p>"
+  end
+
+  it "should work with tags in upper case" do
+    HtmlPress.press("<P>  </p>").should eql "<P></p>"
+  end
+
+  it "should remove whitespaces between IE conditional comments" do
+    text = "<p></p>  <!--[if IE]><html class=\"ie\">      <![endif]--> <!--[if IE]><html class=\"ie1\"><![endif]-->"
+    text2 = "<p></p><!--[if IE]><html class=\"ie\"><![endif]--><!--[if IE]><html class=\"ie1\"><![endif]-->"
+    HtmlPress.press(text).should eql text2
+  end
+
+  it "should treat text inside IE conditional comments as it was without comments" do
+    text = "<div class=\"a\"   id=\"b\">    </div>    <p></p>"
+    text2 = HtmlPress.press(text)
+    text = "<!--[if IE]>" + text + "<![endif]-->"
+    text2 = "<!--[if IE]>" + text2 + "<![endif]-->"
+    HtmlPress.press(text).should eql text2
+    text = "<script> (function(undefined){ var a;}()) </script>"
+    text2 = HtmlPress.press(text)
+    text = "<!--[if IE]>" + text + "<![endif]-->"
+    text2 = "<!--[if IE]>" + text2 + "<![endif]-->"
+    HtmlPress.press(text).should eql text2
+  end
+
+  it "should remove unnecessary whitespaces inside tag" do
+    HtmlPress.press("<p class=\"a\"   id=\"b\"></p>").should eql "<p class=\"a\" id=\"b\"></p>"
+    HtmlPress.press("<p class=\"a\" ></p>").should eql "<p class=\"a\"></p>"
+    HtmlPress.press("<img src=\"\" />").should eql "<img src=\"\"/>"
+    HtmlPress.press("<br />").should eql "<br/>"
+  end
+
+  it "should work with 'badly' formatted attributes" do
+    HtmlPress.press("<p class='a'   id='b'></p>").should eql "<p class='a' id='b'></p>"
+    # HtmlPress.press("<p class = 'a'></p>").should eql "<p class='a'></p>"
+    # HtmlPress.press("<p class = a></p>").should eql "<p class=a></p>"
+  end
+
+  it "should optimize attributes" do
+    HtmlPress.press("<p class=\"a  b\"></p>").should eql "<p class=\"a b\"></p>"
+    # http(s):// to //
+  end
+
+  it "should compress css in style attributes" do
+    HtmlPress.press("<p style=\"display: none;\"></p>").should eql "<p style=\"display:none\"></p>"
+    HtmlPress.press("<p style=\"\"></p>").should eql "<p style=\"\"></p>"
+    #FIX those tests can be broken if algorithm of css_press will be changed
+    HtmlPress.press("<p style=\"font-family:Arial ,'Helvetica Neue'\"></p>").should eql "<p style=\"font-family:Arial ,'Helvetica Neue'\"></p>"
+    HtmlPress.press("<p style='font-family:Arial ,\"Helvetica Neue\"'></p>").should eql "<p style='font-family:Arial ,\"Helvetica Neue\"'></p>"
+  end
+
+  it "should work with namespaces" do
+    text = "<html xmlns:og=\"http://ogp.me/ns#\" class=\"a b\"><og:like>like</og:like></html>"
+    HtmlPress.press(text).should eql text
+  end
+
+  it "should not modify input value" do
+    text = "<div>   </div>"
+    text1 = text.dup
+    HtmlPress.press(text).should_not eql text
+    text.should eql text1
+  end
+
+  it "should leave whitespaces inside other attributes" do
+    text = "<a onclick=\"alert('     ')\" unknown_attr='   a  a'>a</a>"
+    HtmlPress.press(text).should eql text
+  end
+
+  it "should report javascript errors" do
+    ["<script>function(){</script>", "<a onclick=\"return false\"></a>"].each do |script_with_error|
+      Lg.warns = []
+      HtmlPress.press(script_with_error, {:logger => Lg.method(:log)}).should eql script_with_error
+      Lg.warns.size.should eql 1
+    end
+  end
+
+  it "should report css errors" do
+    ["<style>.clas{margin:</style>", "<a style=\"#asd\">link</a>"].each do |style_with_error|
+      Lg.warns = []
+      HtmlPress.press(style_with_error, {:logger => Lg.method(:log)}).should eql style_with_error
+      Lg.warns.size.should eql 1
+    end
+  end
+
+  it "should remove values of boolean attributes" do
+    HtmlPress.press("<option selected=\"selected\">a</option>").should eql "<option selected>a</option>"
+    HtmlPress.press("<input type=\"checkbox\" checked=\"checked\"/>").should eql "<input type=\"checkbox\" checked/>"
+    HtmlPress.press("<input type=\"radio\" checked=\"checked\"/>").should eql "<input type=\"radio\" checked/>"
+    # disabled            (input, textarea, button, select, option, optgroup)
+    HtmlPress.press("<input disabled=\"disabled\"/>").should eql "<input disabled/>"
+    # readonly            (input type=text/password, textarea)
+    HtmlPress.press("<input readonly=\"readonly\"/>").should eql "<input readonly/>"
+    # HtmlPress.press("<script src=\"example.com\" async=\"async\"></script>").should eql "<script src=\"example.com\" async></script>"
+    # HtmlPress.press("<script src=\"example.com\" defer=\"defer\"></script>").should eql "<script src=\"example.com\" async></script>"
+    # HtmlPress.press("<select multiple=\"multiple\"/>").should eql "<select multiple/>"
+    # ismap     isMap     (img, input type=image)
+    # declare             (object; never used)
+    # noresize  noResize  (frame)
+    # nowrap    noWrap    (td, th; deprecated)
+    # noshade   noShade   (hr; deprecated)
+    # compact             (ul, ol, dl, menu, dir; deprecated)
+  end
+
+  it "should remove attributes with default values" do
+    HtmlPress.press("<script type=\"text/javascript\" language=\"JavaScript\">var a</script>").should eql "<script>var a</script>"
+    HtmlPress.press("<style type=\"text/stylesheet\"></style>").should eql "<style></style>"
+    HtmlPress.press("<link type=\"text/stylesheet\"/>").should eql "<link/>"
+    HtmlPress.press("<link rel=\"alternate\" type=\"application/rss+xml\"/>").should eql "<link rel=\"alternate\" type=\"application/rss+xml\"/>"
+    HtmlPress.press("<form method=\"get\"></form>").should eql "<form></form>"
+    HtmlPress.press("<input type=\"text\"/>").should eql "<input/>"
+    # input value "" ?
+  end
+
+  it "should convert html entities to utf-8 symbols" do
+    HtmlPress.press("&lt; &#60; &gt; &#62; &amp; &#38;").should eql "&lt; &#60; &gt; &#62; &amp; &#38;"
+    HtmlPress.press("&eacute;lan").should eql "élan"
+  end
+
+  it "should remove unnecessary quotes for attribute values" do
+    HtmlPress.press("<img src=\"\">", {:unquoted_attributes => true}).should eql "<img src=\"\">"
+    HtmlPress.press("<p id=\"a\"></p>", {:unquoted_attributes => true}).should eql "<p id=a></p>"
+    text = "<p id=\"a b\"></p>"
+    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
+    text = "<p id=\"a=\"></p>"
+    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
+    text = "<p id=\"a'\"></p>"
+    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
+    text = "<p id=\"a`\"></p>"
+    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
+    text = "<p id='a\"'></p>"
+    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
+    text = "<p id=\"a\t\"></p>"
+    HtmlPress.press(text, {:unquoted_attributes => true}).should eql text
+  end
+
+  it "should remove empty attribute values" do
+    HtmlPress.press("<img src=\"\">", {:drop_empty_values => true}).should eql "<img src>"
+  end
+
+  it "should compress javascript in event attributes" do
+    %w[onfocus onblur onselect onchange onclick
+      ondblclick onmousedown onmouseup onmouseover onmousemove
+      onmouseout onkeypress onkeydown onkeyup
+    ].each do |evt|
+      HtmlPress.press("<a #{evt}=\"javacript: alert('  ');\"></a>").should eql "<a #{evt}=\"alert('  ')\"></a>"
+      HtmlPress.press("<a #{evt}=\"\"></a>").should eql "<a #{evt}=\"\"></a>"
+    end
+  end
+
+  # it "should concatenate adjecent script and style tags" do
+  #   all stylle tags can be collected, concatneated and placed in header
+  # end
+
 end

metadata

@@ -1,78 +1,105 @@
---- !ruby/object:Gem::Specification
+--- !ruby/object:Gem::Specification 
 name: html_press
-version: !ruby/object:Gem::Version
-  version: 0.6.5
+version: !ruby/object:Gem::Version 
+  hash: 11
   prerelease: 
+  segments: 
+  - 0
+  - 6
+  - 6
+  version: 0.6.6
 platform: ruby
-authors:
+authors: 
 - stereobooster
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-04-28 00:00:00.000000000Z
-dependencies:
-- !ruby/object:Gem::Dependency
-  name: rspec
-  requirement: &28618932 !ruby/object:Gem::Requirement
+
+date: 2012-08-07 00:00:00 +03:00
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  type: :development
+  version_requirements: &id001 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
-        version: '0'
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 3
+        segments: 
+        - 0
+        version: "0"
+  requirement: *id001
+  prerelease: false
+  name: rspec
+- !ruby/object:Gem::Dependency 
   type: :development
+  version_requirements: &id002 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 3
+        segments: 
+        - 0
+        version: "0"
+  requirement: *id002
   prerelease: false
-  version_requirements: *28618932
-- !ruby/object:Gem::Dependency
   name: rake
-  requirement: &28618680 !ruby/object:Gem::Requirement
+- !ruby/object:Gem::Dependency 
+  type: :runtime
+  version_requirements: &id003 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 23
+        segments: 
+        - 0
+        - 3
+        - 2
+        version: 0.3.2
+  requirement: *id003
   prerelease: false
-  version_requirements: *28618680
-- !ruby/object:Gem::Dependency
   name: css_press
-  requirement: &28618380 !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
-        version: 0.3.1
+- !ruby/object:Gem::Dependency 
   type: :runtime
+  version_requirements: &id004 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 3
+        segments: 
+        - 0
+        version: "0"
+  requirement: *id004
   prerelease: false
-  version_requirements: *28618380
-- !ruby/object:Gem::Dependency
   name: uglifier
-  requirement: &28618128 !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
-        version: '0'
+- !ruby/object:Gem::Dependency 
   type: :runtime
-  prerelease: false
-  version_requirements: *28618128
-- !ruby/object:Gem::Dependency
-  name: htmlentities
-  requirement: &28617852 !ruby/object:Gem::Requirement
+  version_requirements: &id005 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :runtime
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 3
+        segments: 
+        - 0
+        version: "0"
+  requirement: *id005
   prerelease: false
-  version_requirements: *28617852
+  name: htmlentities
 description: Ruby gem for compressing html
-email:
+email: 
 - stereobooster@gmail.com
 executables: []
+
 extensions: []
+
 extra_rdoc_files: []
-files:
+
+files: 
 - .gitignore
 - .travis.yml
 - Gemfile
@@ -86,34 +113,39 @@ files:
 - lib/html_press/uglifier.rb
 - lib/html_press/version.rb
 - spec/html_press_spec.rb
+has_rdoc: true
 homepage: https://github.com/stereobooster/html_press
 licenses: []
+
 post_install_message: 
 rdoc_options: []
-require_paths:
+
+require_paths: 
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement
+required_ruby_version: !ruby/object:Gem::Requirement 
   none: false
-  requirements:
-  - - ! '>='
-    - !ruby/object:Gem::Version
-      version: '0'
-      segments:
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
       - 0
-      hash: -910783917
-required_rubygems_version: !ruby/object:Gem::Requirement
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
   none: false
-  requirements:
-  - - ! '>='
-    - !ruby/object:Gem::Version
-      version: '0'
-      segments:
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
       - 0
-      hash: -910783917
+      version: "0"
 requirements: []
+
 rubyforge_project: html_press
-rubygems_version: 1.8.15
+rubygems_version: 1.5.2
 signing_key: 
 specification_version: 3
 summary: Compress html
 test_files: []
+