typogruby 1.0.5 → 1.0.6

data/README.md

@@ -47,6 +47,14 @@ This will output the contents of your file with all filters applied. Use `typogr
 
 ## Changelog
 
+### 1.0.6
+
+* Bugfix: ignore inline javascript
+
+### 1.0.5
+
+* Manage gem dependencies using Bundler
+
 ### 1.0.4
 
 * Bugfix: no longer regard combination of digits and periods as caps.

data/VERSION

@@ -1 +1 @@
-1.0.5
+1.0.6

data/lib/typogruby.rb

@@ -1,4 +1,5 @@
 require 'rubypants'
+require 'digest/md5'
 
 # A collection of simple helpers for improving web
 # typograhy. Based on TypographyHelper by Luke Hartman and Typogrify.
@@ -54,8 +55,11 @@ module Typogruby
   # @return [String] input text with ampersands wrapped
   def amp(text)
     # $1 is an excluded HTML tag, $2 is the part before the caps and $3 is the amp match
-    text.gsub(/<(code|pre).+?<\/\1>|(\s|&nbsp;)&(?:amp;|#38;)?(\s|&nbsp;)/) {|str|
-    $1 ? str : $2 + '<span class="amp">&amp;</span>' + $3 }.gsub(/(\w+)="(.*?)<span class="amp">&amp;<\/span>(.*?)"/, '\1="\2&amp;\3"')
+    ignore_scripts(text) do |t|
+      t.gsub(/<(code|pre).+?<\/\1>|(\s|&nbsp;)&(?:amp;|#38;)?(\s|&nbsp;)/) { |str|
+        $1 ? str : $2 + '<span class="amp">&amp;</span>' + $3
+      }.gsub(/(\w+)="(.*?)<span class="amp">&amp;<\/span>(.*?)"/, '\1="\2&amp;\3"')
+    end
   end
 
   # replaces space(s) before the last word (or tag before the last word)
@@ -102,14 +106,16 @@ module Typogruby
   # @param [String] text input text
   # @return [String] input text with non-breaking spaces inserted
   def widont(text)
-    text.gsub(%r{
-      ((?:</?(?:a|em|span|strong|i|b)[^>]*>)|[^<>\s]) # must be proceeded by an approved inline opening or closing tag or a nontag/nonspace
-      \s+                                             # the space to replace
-      (([^<>\s]+)                                     # must be flollowed by non-tag non-space characters
-      \s*                                             # optional white space!
-      (</(a|em|span|strong|i|b)>\s*)*                 # optional closing inline tags with optional white space after each
-      ((</(p|h[1-6]|li|dt|dd)>)|$))                   # end with a closing p, h1-6, li or the end of the string
-    }x) { |match| $1 + (match.include?('&nbsp;') ? ' ' : '&nbsp;') + $2 } # Make sure to not add another nbsp before one already there
+    ignore_scripts(text) do |t|
+      t.gsub(%r{
+        ((?:</?(?:a|em|span|strong|i|b)[^>]*>)|[^<>\s]) # must be proceeded by an approved inline opening or closing tag or a nontag/nonspace
+        \s+                                             # the space to replace
+        (([^<>\s]+)                                     # must be flollowed by non-tag non-space characters
+        \s*                                             # optional white space!
+        (</(a|em|span|strong|i|b)>\s*)*                 # optional closing inline tags with optional white space after each
+        ((</(p|h[1-6]|li|dt|dd)>)|$))                   # end with a closing p, h1-6, li or the end of the string
+      }x) { |match| $1 + (match.include?('&nbsp;') ? ' ' : '&nbsp;') + $2 } # Make sure to not add another nbsp before one already there
+    end
   end
 
   # surrounds two or more consecutive captial letters, perhaps with interspersed digits and periods
@@ -132,15 +138,17 @@ module Typogruby
   # @param [String] text input text
   # @return [String] input text with caps wrapped
   def caps(text)
-    # $1 is an excluded HTML tag, $2 is the part before the caps and $3 is the caps match
-    text.gsub(/<(?i)(code|pre)(?-i).+?<(?i)\/\1(?-i)>|(\s|&nbsp;|^|'|"|>)([A-Z\d][A-Z\d\.']{1,})(?!\w)/)  do |str|
-      excluded, before, caps = $1, $2, $3
-      if excluded
-        str
-      elsif $3 =~ /^[\d\.]+$/
-        before + caps
-      else
-        before + '<span class="caps">' + caps + '</span>'
+    ignore_scripts(text) do |t|
+      # $1 is an excluded HTML tag, $2 is the part before the caps and $3 is the caps match
+      t.gsub(/<(?i)(code|pre)(?-i).+?<(?i)\/\1(?-i)>|(\s|&nbsp;|^|'|"|>)([A-Z\d][A-Z\d\.']{1,})(?!\w)/)  do |str|
+        excluded, before, caps = $1, $2, $3
+        if excluded
+          str
+        elsif $3 =~ /^[\d\.]+$/
+          before + caps
+        else
+          before + '<span class="caps">' + caps + '</span>'
+        end
       end
     end
   end
@@ -167,7 +175,9 @@ module Typogruby
   # @return [String] input text with initial quotes wrapped
   def initial_quotes(text)
     # $1 is the initial part of the string, $2 is the quote or entitity, and $3 is the double quote
-    text.gsub(/((?:<(?:h[1-6]|p|li|dt|dd)[^>]*>|^)\s*(?:<(?:a|em|strong|span)[^>]*>)?)('|&#8216;|&lsquo;|("|&#8220;|&ldquo;))/) {$1 + "<span class=\"#{'d' if $3}quo\">#{$2}</span>"}
+    ignore_scripts(text) do |t|
+      t.gsub(/((?:<(?:h[1-6]|p|li|dt|dd)[^>]*>|^)\s*(?:<(?:a|em|strong|span)[^>]*>)?)('|&#8216;|&lsquo;|("|&#8220;|&ldquo;))/) {$1 + "<span class=\"#{'d' if $3}quo\">#{$2}</span>"}
+    end
   end
 
   # main function to do all the functions from the method
@@ -177,5 +187,26 @@ module Typogruby
     initial_quotes(caps(smartypants(widont(amp(text)))))
   end
 
+private
+
+  # Hackish text filter that will make sure our text filters leave inline
+  # javascript alone without resorting to a full-blown HTML parser.
+  #
+  # The idea is simple: every text filter is applied as a block to this
+  # method. This will preprocess the text and replace any inline scripts
+  # with a MD5 hash of its entire contents. Then the filter is called,
+  # and then the hashes are replaced back with their original content.
+  def ignore_scripts(text)
+    @ignored_scripts = {}
+    modified_text = text.gsub(/<script[^>]*>.*?<\/script>/mi) do |script|
+      hash = Digest::MD5.hexdigest(script)
+      @ignored_scripts[hash] = script
+      hash
+    end
+    yield(modified_text).gsub(/#{@ignored_scripts.keys.join('|')}/) do |h|
+      @ignored_scripts.delete(h)
+    end
+  end
+
   extend self
 end

data/test/test_typogruby.rb

@@ -12,6 +12,7 @@ class TestTypogruby < Test::Unit::TestCase
   def test_should_ignore_special_amps
     assert_equal 'One <span class="amp">&amp;</span> two', amp('One <span class="amp">&amp;</span> two')
     assert_equal '&ldquo;this&rdquo; <span class="amp">&amp;</span> <a href="/?that&amp;test">that</a>', amp('&ldquo;this&rdquo; & <a href="/?that&amp;test">that</a>')
+    assert_equal %Q{<script>\nvar x = "FOO & BAR";\n</script>"}, initial_quotes(%Q{<script>\nvar x = "FOO & BAR";\n</script>"})
   end
 
   def test_should_ignore_standalone_amps_in_attributes
@@ -28,6 +29,7 @@ class TestTypogruby < Test::Unit::TestCase
     assert_equal '<Pre>CAPS</PRE> with odd tag names <span class="caps">CAPS</span>', caps("<Pre>CAPS</PRE> with odd tag names CAPS")
     assert_equal 'A message from <span class="caps">2KU2</span> with digits', caps("A message from 2KU2 with digits")
     assert_equal 'Dotted caps followed by spaces should never include them in the wrap <span class="caps">D.O.T.</span>   like so.', caps("Dotted caps followed by spaces should never include them in the wrap D.O.T.   like so.")
+    assert_equal %Q{<script>\nvar x = "FOO BAR and BAZ";\n</script>"}, initial_quotes(%Q{<script>\nvar x = "FOO BAR and BAZ";\n</script>"})
   end
 
   def test_should_not_break_caps_with_apostrophes
@@ -43,6 +45,10 @@ class TestTypogruby < Test::Unit::TestCase
     assert_equal '<span class="quo">&lsquo;</span>With manual quotes&rsquo;', initial_quotes('&lsquo;With manual quotes&rsquo;')
   end
 
+  def test_should_not_replace_quotes_in_scripts
+    assert_equal %Q{<script>\nvar x = "foo" +\n"bar";\n</script>"}, initial_quotes(%Q{<script>\nvar x = "foo" +\n"bar";\n</script>"})
+  end
+
   def test_should_apply_smartypants
     assert_equal 'The &#8220;Green&#8221; man', smartypants('The "Green" man')
   end
@@ -76,6 +82,7 @@ class TestTypogruby < Test::Unit::TestCase
   def test_should_ignore_widows_in_special_tags
     assert_equal '<div>Divs get no love!</div>', widont('<div>Divs get no love!</div>')
     assert_equal '<pre>Neither do PREs</pre>', widont('<pre>Neither do PREs</pre>')
+    assert_equal "<script>\nreturn window;\n</script>", widont("<script>\nreturn window;\n</script>")
     assert_equal '<div><p>But divs with paragraphs&nbsp;do!</p></div>', widont('<div><p>But divs with paragraphs do!</p></div>')
   end
 end

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: typogruby
 version: !ruby/object:Gem::Version 
-  hash: 29
+  hash: 27
   prerelease: false
   segments: 
   - 1
   - 0
-  - 5
-  version: 1.0.5
+  - 6
+  version: 1.0.6
 platform: ruby
 authors: 
 - Arjan van der Gaag
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-08-10 00:00:00 +02:00
+date: 2010-08-17 00:00:00 +02:00
 default_executable: typogruby
 dependencies: 
 - !ruby/object:Gem::Dependency