truncateHTML 0.0.1

data/.gitignore

@@ -0,0 +1,5 @@
+*.gem
+.bundle
+Gemfile.lock
+pkg/*
+nbproject

data/Gemfile

@@ -0,0 +1,4 @@
+source "http://rubygems.org"
+
+# Specify your gem's dependencies in truncateHTML.gemspec
+gemspec

data/Rakefile

@@ -0,0 +1,2 @@
+require 'bundler'
+Bundler::GemHelper.install_tasks

data/lib/truncateHTML/hpricot_truncator.rb

@@ -0,0 +1,47 @@
+module HpricotTruncator
+  module NodeWithChildren
+    def truncate(max_length)
+      return self if inner_text.length <= max_length
+      truncated_node = if self.is_a?(Hpricot::Doc)
+        self.dup
+      else
+        self.class.send(:new, self.name, self.attributes)
+      end
+      truncated_node.children = []
+      each_child do |node|
+        if node.is_a?(Hpricot::Elem) && node.name == "html"
+          node.children.each do |c|
+            # Find the body node and use it. Let us reset earlier truncations
+            # and start afresh with this body tag
+            return c.truncate(max_length) if (c.is_a?(Hpricot::Elem) && c.name == "body")
+          end
+        end
+
+        remaining_length = max_length - truncated_node.inner_text.length
+        break if remaining_length <= 0
+        truncated_node.children << node.truncate(remaining_length)
+      end
+      truncated_node
+    end
+  end
+
+  module TextNode
+    def truncate(max_length)
+      # We're using String#scan because Hpricot doesn't distinguish entities.
+      Hpricot::Text.new(content.scan(/&#?[^\W_]+;|./).first(max_length).join)
+    end
+  end
+
+  module IgnoredTag
+    def truncate(max_length)
+      self
+    end
+  end
+end
+
+Hpricot::Doc.send(:include,       HpricotTruncator::NodeWithChildren)
+Hpricot::Elem.send(:include,      HpricotTruncator::NodeWithChildren)
+Hpricot::Text.send(:include,      HpricotTruncator::TextNode)
+Hpricot::BogusETag.send(:include, HpricotTruncator::IgnoredTag)
+Hpricot::Comment.send(:include,   HpricotTruncator::IgnoredTag)
+Hpricot::DocType.send(:include,   HpricotTruncator::IgnoredTag)

data/lib/truncateHTML/version.rb

@@ -0,0 +1,3 @@
+module Truncatehtml
+  VERSION = "0.0.1"
+end

data/lib/truncateHTML.rb

@@ -0,0 +1,60 @@
+# By Henrik Nyh <http://henrik.nyh.se> 2008-01-30.
+# Free to modify and redistribute with credit.
+
+# modified by Dave Nolan <http://textgoeshere.org.uk> 2008-02-06
+# Ellipsis appended to text of last HTML node
+# Ellipsis inserted after final word break
+
+# modified by Mark Dickson <mark@sitesteaders.com> 2008-12-18
+# Option to truncate to last full word
+# Option to include a 'more' link
+# Check for nil last child
+
+# modified by Ken-ichi Ueda <http://kueda.net> 2009-09-02
+# Rails 2.3 compatability (chars -> mb_chars), via Henrik
+# Hpricot 0.8 compatability (avoid dup on Hpricot::Elem)
+
+# modified by Satyaram B V <http://bvsatyaram.com> 2011-03-24
+# Rails version independent
+# Making this a gem
+# Removed mb_chars
+
+require "hpricot"
+require "truncateHTML/hpricot_truncator"
+
+module TruncateHTML
+  # Like the Rails _truncate_ helper but doesn't break HTML tags, entities, and optionally. words.
+  def self.truncate(text, options={})
+    return if text.nil?
+
+    max_length = options[:max_length] || 40
+    ellipsis = options[:ellipsis] || "..."
+    words = options[:words] || false
+    status = options[:status] || false
+    # use :link => link_to('more', post_path), or something to that effect
+
+    doc = Hpricot(text.to_s)
+    ellipsis_length = Hpricot(ellipsis).inner_text.length
+    content_length = doc.inner_text.length
+    actual_length = max_length - ellipsis_length
+
+    if content_length > max_length
+      truncated_doc = doc.truncate(actual_length)
+
+      if words
+        word_length = actual_length - (truncated_doc.inner_html.length - truncated_doc.inner_html.rindex(' '))
+        truncated_doc = doc.truncate(word_length)
+      end
+
+      #XXX The check here has to be blank as the inner_html for text node is blank
+      return_string = truncated_doc.inner_html + ellipsis
+      return_string += options[:link] if options[:link]
+      return_status = true
+    else
+      return_string = text.to_s
+      return_status = false
+    end
+
+    return status ? [return_string, return_status] : return_string
+  end
+end

data/test/truncateHTML_test.rb

@@ -0,0 +1,74 @@
+$:.unshift File.join(File.dirname(__FILE__),'..','lib')
+
+require 'test/unit'
+require 'truncateHTML'
+
+class TruncateHTMLTest < Test::Unit::TestCase
+  def test_truncate_html_ignores_doc_type
+    msg = <<-DOCMSG
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
+<html>
+<body bgcolor="#ffffff" text="#000000">
+Take this bottle and go fetch me some water please. Another of those mobiles is going bad.
+</body>
+</html>
+DOCMSG
+    assert_equal("Take this bottle and go fetch me some...", TruncateHTML.truncate(msg, :max_length => 40))
+  end
+
+  def test_truncate_html_with_body_node
+    msg = <<-MSG
+<html>
+<head>
+  <meta content="text/html;charset=UTF-8" http-equiv="Content-Type">
+  <title></title>
+</head>
+<body bgcolor="#ffffff" text="#000000">
+Dear Ramana,<br>
+Satyaram is currently setting up his account. The fourth estate magazine
+is a big time hit of all time<br>
+</body>
+</html>
+MSG
+    assert_equal("Dear Ramana,<br />Satyaram is currently...", TruncateHTML.truncate(msg, :max_length => 40, :words => true))
+  end
+
+  def test_truncate_html_plain_text
+    text = "this is some exmaple test"
+    assert_equal("this is...", TruncateHTML.truncate(text, :max_length => 10))
+  end
+
+  def test_truncate_html_ordinary_nodes
+    text = "<p><b><i>this is some</i> exmaple </b></p> <p>test</p>"
+    assert_equal("<p><b><i>this is</i></b></p>...", TruncateHTML.truncate(text, :max_length => 10))
+  end
+
+  def test_truncate_html_with_html_with_multiple_bodies
+    msg = <<-MSG
+<html>
+<head>
+  <meta content="text/html;charset=UTF-8" http-equiv="Content-Type">
+  <title></title>
+</head>
+<body bgcolor="#ffffff" text="#000000">
+Dear Ramana,<br>
+Satyaram is currently setting up his account. The fourth estate magazine
+is a big time hit of all time<br>
+</body>
+</html>
+
+<html>
+<head>
+  <meta content="text/html;charset=UTF-8" http-equiv="Content-Type">
+  <title></title>
+</head>
+<body bgcolor="#ffffff" text="#000000">
+Dear Ramana,<br>
+Satyaram is currently setting up his account. The fourth estate magazine
+is a big time hit of all time<br>
+</body>
+</html>
+MSG
+    assert_equal("Dear Ramana,<br />Satyaram is currently set...", TruncateHTML.truncate(msg))
+  end
+end

data/truncateHTML.gemspec

@@ -0,0 +1,23 @@
+# -*- encoding: utf-8 -*-
+$:.push File.expand_path("../lib", __FILE__)
+require "truncateHTML/version"
+
+Gem::Specification.new do |s|
+  s.name        = "truncateHTML"
+  s.version     = Truncatehtml::VERSION
+  s.platform    = Gem::Platform::RUBY
+  s.authors     = ["Satyaram B V"]
+  s.email       = ["bvsatyaram AT gmail DOT com"]
+  s.homepage    = "http://bvsatyaram.com"
+  s.summary     = %q{Truncate HTML without breaking HTML tags, entities, and optionally words}
+  s.description = %q{Truncate HTML without breaking HTML tags, entities, and optionally words}
+
+  s.add_dependency "hpricot"
+
+  s.rubyforge_project = "truncateHTML"
+
+  s.files         = `git ls-files`.split("\n")
+  s.test_files    = `git ls-files -- {test,spec,features}/*`.split("\n")
+  s.executables   = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
+  s.require_paths = ["lib"]
+end

metadata

@@ -0,0 +1,88 @@
+--- !ruby/object:Gem::Specification 
+name: truncateHTML
+version: !ruby/object:Gem::Version 
+  hash: 29
+  prerelease: false
+  segments: 
+  - 0
+  - 0
+  - 1
+  version: 0.0.1
+platform: ruby
+authors: 
+- Satyaram B V
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2011-03-26 00:00:00 +05:30
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: hpricot
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 3
+        segments: 
+        - 0
+        version: "0"
+  type: :runtime
+  version_requirements: *id001
+description: Truncate HTML without breaking HTML tags, entities, and optionally words
+email: 
+- bvsatyaram AT gmail DOT com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: []
+
+files: 
+- .gitignore
+- Gemfile
+- Rakefile
+- lib/truncateHTML.rb
+- lib/truncateHTML/hpricot_truncator.rb
+- lib/truncateHTML/version.rb
+- test/truncateHTML_test.rb
+- truncateHTML.gemspec
+has_rdoc: true
+homepage: http://bvsatyaram.com
+licenses: []
+
+post_install_message: 
+rdoc_options: []
+
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+requirements: []
+
+rubyforge_project: truncateHTML
+rubygems_version: 1.3.7
+signing_key: 
+specification_version: 3
+summary: Truncate HTML without breaking HTML tags, entities, and optionally words
+test_files: []
+