xls_html_cleaner 0.0.1

data/ChangeLog

@@ -0,0 +1,4 @@
+== 0.0.1 / 2010-02-15
+
+* initial release
+

data/README

@@ -0,0 +1,31 @@
+
+= xls_html_cleaner
+
+
+== Description
+
+Clean up your Excel generated HTML
+
+== Installation
+
+=== Archive Installation
+
+	rake install
+
+=== Gem Installation
+
+	gem install xls_html_cleaner
+
+
+== Features/Problems
+
+
+== Synopsis
+
+ $ xls_html_cleaner SRC_HTML > DEST_HTML
+
+== Copyright
+
+Author::    wtnabe
+Copyright:: Copyright (c) 2010 wtnabe
+License::   Two-clause BSD

data/Rakefile

@@ -0,0 +1,113 @@
+# -*- mode: ruby -*-
+require 'rubygems'
+require 'rake'
+require 'rake/clean'
+require 'rake/testtask'
+require 'rake/packagetask'
+require 'rake/gempackagetask'
+require 'rake/rdoctask'
+require 'fileutils'
+require 'lib/xls_html_cleaner'
+include FileUtils
+
+NAME              = "xls_html_cleaner"
+AUTHOR            = "wtnabe"
+EMAIL             = "wtnabe@gmail.com"
+DESCRIPTION       = "Clean up your Excel generated HTML"
+RUBYFORGE_PROJECT = "xls_html_cleaner"
+HOMEPATH          = "http://github.com/wtnabe/xls_html_cleaner"
+BIN_FILES         = %w( xls_html_cleaner )
+
+VERS              = XlsHtmlCleaner::VERSION
+REV = File.read(".svn/entries")[/committed-rev="(d+)"/, 1] rescue nil
+CLEAN.include ['**/.*.sw?', '*.gem', '.config']
+RDOC_OPTS = [
+	'--title', "#{NAME} documentation",
+	"--charset", "utf-8",
+	"--opname", "index.html",
+	"--line-numbers",
+	"--main", "README",
+	"--inline-source",
+]
+
+task :default => [:test]
+task :package => [:clean]
+
+Rake::TestTask.new("test") do |t|
+	t.libs   << "test"
+	t.pattern = "test/**/*_test.rb"
+	t.verbose = true
+end
+
+spec = Gem::Specification.new do |s|
+	s.name              = NAME
+	s.version           = VERS
+	s.platform          = Gem::Platform::RUBY
+	s.has_rdoc          = true
+	s.extra_rdoc_files  = ["README", "ChangeLog"]
+	s.rdoc_options     += RDOC_OPTS + ['--exclude', '^(examples|extras)/']
+	s.summary           = DESCRIPTION
+	s.description       = DESCRIPTION
+	s.author            = AUTHOR
+	s.email             = EMAIL
+	s.homepage          = HOMEPATH
+	s.executables       = BIN_FILES
+	s.bindir            = "bin"
+	s.require_path      = "lib"
+	#s.autorequire       = ""
+	s.test_files        = Dir["test/*_test.rb"]
+
+  s.add_dependency('hpricot', '>=1.3.1')
+	#s.required_ruby_version = '>= 1.8.2'
+
+	s.files = %w(README ChangeLog Rakefile) +
+		Dir.glob("{bin,doc,test,lib,templates,generator,extras,website,script}/**/*") + 
+		Dir.glob("ext/**/*.{h,c,rb}") +
+		Dir.glob("examples/**/*.rb") +
+		Dir.glob("tools/*.rb") +
+		Dir.glob("rails/*.rb")
+
+	s.extensions = FileList["ext/**/extconf.rb"].to_a
+end
+
+Rake::GemPackageTask.new(spec) do |p|
+	p.need_tar = true
+	p.gem_spec = spec
+end
+
+desc "install #{NAME}-#{VERS}.gem"
+task :install do
+	name = "#{NAME}-#{VERS}.gem"
+	sh %{rake package}
+	sh %{sudo gem install pkg/#{name}}
+end
+
+desc "uninstall gem"
+task :uninstall => [:clean] do
+	sh %{sudo gem uninstall #{NAME}}
+end
+
+
+Rake::RDocTask.new do |rdoc|
+	rdoc.rdoc_dir = 'html'
+	rdoc.options += RDOC_OPTS
+	rdoc.template = "resh"
+	#rdoc.template = "#{ENV['template']}.rb" if ENV['template']
+	if ENV['DOC_FILES']
+		rdoc.rdoc_files.include(ENV['DOC_FILES'].split(/,\s*/))
+	else
+		rdoc.rdoc_files.include('README', 'ChangeLog')
+		rdoc.rdoc_files.include('lib/**/*.rb')
+		rdoc.rdoc_files.include('ext/**/*.c')
+	end
+end
+
+desc 'Show information about the gem.'
+task :debug_gem do
+	puts spec.to_ruby
+end
+
+desc 'Update gem spec'
+task :gemspec do
+  open("#{NAME}.gemspec", 'w').write spec.to_ruby
+end

data/bin/xls_html_cleaner

@@ -0,0 +1,5 @@
+#! /usr/bin/env ruby
+
+require File.dirname( __FILE__ ) + '/../lib/xls_html_cleaner'
+
+puts XlsHtmlCleaner.new.clean!( readlines.to_s )

data/lib/xls_html_cleaner.rb

@@ -0,0 +1,30 @@
+require 'rubygems' unless defined? RubyGems
+require 'hpricot'
+
+class XlsHtmlCleaner
+  VERSION = '0.0.1'
+
+  ALLOW_TAGS = %w( html head title body 
+                   table thead tbody tfoot tr th td col colgroup )
+
+  def initialize
+    @parser = nil
+  end
+
+  def clean!( str )
+    @parser = Hpricot( str )
+    @parser.traverse_all_element { |e|
+      if ( e.elem? and !ALLOW_TAGS.include?( e.name.downcase ) )
+        e.swap( e.inner_html.size > 0 ? e.inner_html : "\n" )
+      end
+      if ( e.comment? )
+        e.swap( "\n" )
+      end
+      if ( e.elem? )
+        e.attributes.to_hash.each_key { |a|
+          e.remove_attribute( a )
+        }
+      end
+    }.to_s
+  end
+end

data/test/html/simple_ooo.html

@@ -0,0 +1,44 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
+
+<HTML>
+<HEAD>
+	
+	<META HTTP-EQUIV="CONTENT-TYPE" CONTENT="text/html; charset=utf-8">
+	<TITLE></TITLE>
+	<META NAME="GENERATOR" CONTENT="OpenOffice.org 3.1  (Unix)">
+	<META NAME="CREATED" CONTENT="20100213;23342400">
+	<META NAME="CHANGED" CONTENT="0;0">
+	
+	<STYLE>
+		<!-- 
+		BODY,DIV,TABLE,THEAD,TBODY,TFOOT,TR,TH,TD,P { font-family:"ヒラギノ角ゴ ProN W3"; font-size:x-small }
+		 -->
+	</STYLE>
+	
+</HEAD>
+
+<BODY TEXT="#000000">
+<TABLE FRAME=VOID CELLSPACING=0 COLS=3 RULES=NONE BORDER=0>
+	<COLGROUP><COL WIDTH=100><COL WIDTH=100><COL WIDTH=100></COLGROUP>
+	<TBODY>
+		<TR>
+			<TD WIDTH=100 HEIGHT=18 ALIGN=LEFT><FONT FACE="Arial">a</FONT></TD>
+			<TD WIDTH=100 ALIGN=LEFT><FONT FACE="Arial">d</FONT></TD>
+			<TD WIDTH=100 ALIGN=LEFT><FONT FACE="Arial">g</FONT></TD>
+		</TR>
+		<TR>
+			<TD HEIGHT=18 ALIGN=LEFT><FONT FACE="Arial">b</FONT></TD>
+			<TD ALIGN=LEFT><FONT FACE="Arial">e</FONT></TD>
+			<TD ALIGN=LEFT><FONT FACE="Arial">h</FONT></TD>
+		</TR>
+		<TR>
+			<TD HEIGHT=18 ALIGN=LEFT><FONT FACE="Arial">c</FONT></TD>
+			<TD ALIGN=LEFT><FONT FACE="Arial">f</FONT></TD>
+			<TD ALIGN=LEFT><FONT FACE="Arial">I</FONT></TD>
+		</TR>
+	</TBODY>
+</TABLE>
+<!-- ************************************************************************** -->
+</BODY>
+
+</HTML>

data/test/output/simple_ooo.html

@@ -0,0 +1,48 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
+
+<html>
+<head>
+	
+	
+
+	<title></title>
+	
+
+	
+
+	
+
+	
+	
+		
+
+	
+	
+</head>
+
+<body>
+<table>
+	<colgroup><col /><col /><col /></colgroup>
+	<tbody>
+		<tr>
+			<td>a</td>
+			<td>d</td>
+			<td>g</td>
+		</tr>
+		<tr>
+			<td>b</td>
+			<td>e</td>
+			<td>h</td>
+		</tr>
+		<tr>
+			<td>c</td>
+			<td>f</td>
+			<td>I</td>
+		</tr>
+	</tbody>
+</table>
+
+
+</body>
+
+</html>

data/test/test_helper.rb

@@ -0,0 +1,3 @@
+require 'test/unit'
+require File.dirname(__FILE__) + '/../lib/xls_html_cleaner'
+

data/test/xls_html_cleaner_test.rb

@@ -0,0 +1,23 @@
+require File.dirname(__FILE__) + '/test_helper.rb'
+
+require "test/unit"
+class XlsHtmlCleanerTest < Test::Unit::TestCase
+  def setup
+    @obj = XlsHtmlCleaner.new
+  end
+
+  def test_clean!
+    Dir.glob( File.dirname( __FILE__ ) + '/html/*' ) { |html|
+      begin
+        open( File.join( File.dirname( __FILE__ ),
+                         'output',
+                         File.basename( html ) ), 'wb' ) { |f|
+          f.write( @obj.clean!( open( html ).read ) )
+        }
+        assert( true )
+      rescue e
+        raise e
+      end
+    }
+  end
+end

metadata

@@ -0,0 +1,84 @@
+--- !ruby/object:Gem::Specification 
+name: xls_html_cleaner
+version: !ruby/object:Gem::Version 
+  version: 0.0.1
+platform: ruby
+authors: 
+- wtnabe
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2010-02-15 00:00:00 +09:00
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: hpricot
+  type: :runtime
+  version_requirement: 
+  version_requirements: !ruby/object:Gem::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        version: 1.3.1
+    version: 
+description: Clean up your Excel generated HTML
+email: wtnabe@gmail.com
+executables: 
+- xls_html_cleaner
+extensions: []
+
+extra_rdoc_files: 
+- README
+- ChangeLog
+files: 
+- README
+- ChangeLog
+- Rakefile
+- bin/xls_html_cleaner
+- test/html/simple_ooo.html
+- test/output/simple_ooo.html
+- test/test_helper.rb
+- test/xls_html_cleaner_test.rb
+- lib/xls_html_cleaner.rb
+has_rdoc: true
+homepage: http://github.com/wtnabe/xls_html_cleaner
+licenses: []
+
+post_install_message: 
+rdoc_options: 
+- --title
+- xls_html_cleaner documentation
+- --charset
+- utf-8
+- --opname
+- index.html
+- --line-numbers
+- --main
+- README
+- --inline-source
+- --exclude
+- ^(examples|extras)/
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.5
+signing_key: 
+specification_version: 3
+summary: Clean up your Excel generated HTML
+test_files: 
+- test/xls_html_cleaner_test.rb