mechanize 2.0.pre.2 → 2.0

data/test/test_mechanize_link.rb

@@ -0,0 +1,83 @@
+require "helper"
+
+class TestMechanizeLink < Test::Unit::TestCase
+
+  def setup
+    @agent = Mechanize.new
+  end
+
+  def test_click
+    page = @agent.get("http://localhost/frame_test.html")
+    link = page.link_with(:text => "Form Test")
+    assert_not_nil(link)
+    assert_equal('Form Test', link.text)
+    page = link.click
+    assert_equal("http://localhost/form_test.html",
+      @agent.history.last.uri.to_s)
+  end
+
+  def test_click_base
+    page = @agent.get("http://google.com/tc_base_link.html")
+    page = page.links.first.click
+    assert @agent.visited?("http://localhost/index.html")
+  end
+
+  def test_click_unsupported_scheme
+    page = @agent.get("http://google.com/tc_links.html")
+    link = page.link_with(:text => 'javascript link')
+    assert_raise(Mechanize::UnsupportedSchemeError) {
+      link.click
+    }
+
+    @agent.scheme_handlers['javascript'] = lambda { |my_link, my_page|
+      URI.parse('http://localhost/tc_links.html')
+    }
+    assert_nothing_raised {
+      link.click
+    }
+  end
+
+  def test_text_alt_text
+    page = @agent.get("http://localhost/alt_text.html")
+    assert_equal(5, page.links.length)
+    assert_equal(1, page.meta_refresh.length)
+
+    assert_equal '',         page.meta_refresh.first.text
+    assert_equal 'alt text', page.link_with(:href => 'alt_text.html').text
+    assert_equal '',         page.link_with(:href => 'no_alt_text.html').text
+    assert_equal 'no image', page.link_with(:href => 'no_image.html').text
+    assert_equal '',         page.link_with(:href => 'no_text.html').text
+    assert_equal '',         page.link_with(:href => 'nil_alt_text.html').text
+  end
+
+  def test_uri_escaped
+    doc = Nokogiri::HTML::Document.new
+
+    node = Nokogiri::XML::Node.new('foo', doc)
+    node['href'] = 'http://foo.bar/%20baz'
+
+    link = Mechanize::Page::Link.new(node, nil, nil)
+
+    assert_equal 'http://foo.bar/%20baz', link.uri.to_s
+  end
+
+  def test_uri_no_path
+    page = @agent.get("http://localhost/relative/tc_relative_links.html")
+    page = page.link_with(:text => 'just the query string').click
+    assert_equal('http://localhost/relative/tc_relative_links.html?a=b',
+                 page.uri.to_s)
+  end
+
+  def test_uri_weird
+    doc = Nokogiri::HTML::Document.new
+
+    node = Nokogiri::XML::Node.new('foo', doc)
+    node['href'] = 'http://foo.bar/ baz'
+
+    link = Mechanize::Page::Link.new(node, nil, nil)
+
+    assert_equal 'http://foo.bar/%20baz', link.uri.to_s
+  end
+
+end
+

data/test/test_mechanize_page_encoding.rb

@@ -0,0 +1,147 @@
+# -*- coding: utf-8 -*-
+require 'helper'
+require 'cgi'
+
+# tests for Page encoding and charset and parsing
+
+class TestMechanizePageEncoding < Test::Unit::TestCase
+
+  MECH_ASCII_ENCODING = Mechanize::Util::NEW_RUBY_ENCODING ? 'US-ASCII' : 'ISO-8859-1'
+
+  def setup
+    @agent = Mechanize.new
+    @uri = URI('http://localhost/')
+    @response_headers = { 'content-type' => 'text/html' }
+    @body = '<title>hi</title>'
+  end
+
+  def util_page body = @body, headers = @response_headers
+    body.force_encoding Encoding::BINARY if body.respond_to? :force_encoding
+    Mechanize::Page.new @uri, headers, body, 200, @agent
+  end
+
+  def test_page_charset
+    charset = Mechanize::Page.charset 'text/html;charset=vAlue'
+    assert_equal 'vAlue', charset
+  end
+
+  def test_page_charset_upcase
+    charset = Mechanize::Page.charset 'TEXT/HTML;CHARSET=UTF-8'
+    assert_equal 'UTF-8', charset
+  end
+
+  def test_page_semicolon
+    charset = Mechanize::Page.charset 'text/html;charset=UTF-8;'
+    assert_equal 'UTF-8', charset
+  end
+
+  def test_page_charset_no_chaset_token
+    charset = Mechanize::Page.charset 'text/html'
+    assert_nil charset
+  end
+
+  def test_page_charset_returns_nil_when_charset_says_none
+    charset = Mechanize::Page.charset 'text/html;charset=none'
+
+    assert_nil charset
+  end
+
+  def test_page_charset_multiple
+    charset = Mechanize::Page.charset 'text/html;charset=111;charset=222'
+
+    assert_equal '111', charset
+  end
+
+  def test_page_response_header_charset
+    headers = {'content-type' => 'text/html;charset=HEADER'}
+    charsets = Mechanize::Page.response_header_charset(headers)
+
+    assert_equal ['HEADER'], charsets
+  end
+
+  def test_page_response_header_charset_no_token
+    headers = {'content-type' => 'text/html'}
+    charsets = Mechanize::Page.response_header_charset(headers)
+
+    assert_equal [], charsets
+
+    headers = {'X-My-Header' => 'hello'}
+    charsets = Mechanize::Page.response_header_charset(headers)
+
+    assert_equal [], charsets
+  end
+
+  def test_response_header_charset
+    page = util_page nil, {'content-type' => 'text/html;charset=HEADER'}
+
+    assert_equal ['HEADER'], page.response_header_charset
+  end
+
+  def test_page_meta_charset
+    body = '<meta http-equiv="content-type" content="text/html;charset=META">'
+    charsets = Mechanize::Page.meta_charset(body)
+
+    assert_equal ['META'], charsets
+  end
+
+  def test_page_meta_charset_is_empty_when_no_charset_meta
+    body = '<meta http-equiv="refresh" content="5; url=index.html">'
+    charsets = Mechanize::Page.meta_charset(body)
+    assert_equal [], charsets
+  end
+
+  def test_meta_charset
+    body = '<meta http-equiv="content-type" content="text/html;charset=META">'
+    page = util_page body
+
+    assert_equal ['META'], page.meta_charset
+  end
+
+  def test_detected_encoding
+    page = util_page
+
+    assert_equal MECH_ASCII_ENCODING, page.detected_encoding
+  end
+
+  def test_encodings
+    response = {'content-type' => 'text/html;charset=HEADER'}
+    body = '<meta http-equiv="content-type" content="text/html;charset=META">'
+    @agent.default_encoding = 'DEFAULT'
+    page = util_page body, response
+
+    assert_equal true, page.encodings.include?('HEADER')
+    assert_equal true, page.encodings.include?('META')
+    assert_equal true, page.encodings.include?(MECH_ASCII_ENCODING)
+    assert_equal true, page.encodings.include?('DEFAULT')
+  end
+
+  def test_parser_with_default_encoding
+    # pre test
+    assert_equal false, util_page.encodings.include?('Windows-1252')
+
+    @agent.default_encoding = 'Windows-1252'
+    page = util_page
+
+    assert_equal true, page.encodings.include?('Windows-1252')
+  end
+
+  def test_parser_force_default_encoding
+    @agent.default_encoding = 'Windows-1252'
+    @agent.force_default_encoding = true
+    page = util_page
+
+    assert page.encodings.include? 'Windows-1252'
+  end
+
+  def test_parser_encoding_equals_overwrites_force_default_encoding
+    @agent.default_encoding = 'Windows-1252'
+    @agent.force_default_encoding = true
+    page = util_page
+
+    assert_equal 'Windows-1252', page.encoding
+
+    page.encoding = 'ISO-8859-2'
+
+    assert_equal 'ISO-8859-2', page.encoding
+  end
+end

data/test/test_mechanize_page_link.rb

@@ -0,0 +1,379 @@
+# coding: utf-8
+
+require 'helper'
+require 'cgi'
+
+class TestMechanizePage < Test::Unit::TestCase
+
+  WINDOWS_1255 = <<-HTML
+<meta http-equiv="content-type" content="text/html; charset=windows-1255">
+<title>hi</title>
+  HTML
+
+  BAD = <<-HTML
+<meta http-equiv="content-type" content="text/html; charset=windows-1255">
+<title>Bia\xB3ystok</title>
+  HTML
+  BAD.force_encoding Encoding::BINARY if defined? Encoding
+
+  SJIS_TITLE = "\x83\x65\x83\x58\x83\x67"
+
+  SJIS_AFTER_TITLE = <<-HTML
+<title>#{SJIS_TITLE}</title>
+<meta http-equiv="Content-Type" content="text/html; charset=Shift_JIS">
+  HTML
+
+  SJIS_AFTER_TITLE.force_encoding Encoding::BINARY if defined? Encoding
+
+  SJIS_BAD_AFTER_TITLE = <<-HTML
+<title>#{SJIS_TITLE}</title>
+<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
+  HTML
+
+  SJIS_BAD_AFTER_TITLE.force_encoding Encoding::BINARY if defined? Encoding
+
+  UTF8_TITLE = 'テスト'
+  UTF8 = <<-HTML
+<title>#{UTF8_TITLE}</title>
+<meta http-equiv="Content-Type" content="text/html; charset=Shift_JIS">
+  HTML
+
+  ENCODING_ERROR_CLASS = Nokogiri::XML::SyntaxError
+
+  def setup
+    @agent = Mechanize.new
+    @uri = URI('http://example')
+    @res = { 'content-type' => 'text/html' }
+    @body = '<title>hi</title>'
+  end
+
+  def util_page body = @body, res = @res
+    body.force_encoding Encoding::BINARY if body.respond_to? :force_encoding
+    Mechanize::Page.new @uri, res, body, 200, @agent
+  end
+
+  def test_initialize_supported_content_type
+    page = Mechanize::Page.new nil, 'content-type' => 'application/xhtml+xml'
+    assert page
+    assert_equal 'application/xhtml+xml', page.content_type
+
+    page = Mechanize::Page.new nil, 'content-type' => 'text/html'
+    assert page
+    assert_equal 'text/html', page.content_type
+
+    page = Mechanize::Page.new nil, 'content-type' => 'application/xhtml+xml;charset=utf-8'
+    assert page
+    assert_equal 'application/xhtml+xml;charset=utf-8', page.content_type
+
+    page = Mechanize::Page.new nil, 'content-type' => 'text/html;charset=utf-8'
+    assert page
+    assert_equal 'text/html;charset=utf-8', page.content_type
+  end
+
+  def test_initialize_unsupported_content_type
+    e = assert_raises Mechanize::ContentTypeError do
+      Mechanize::Page.new nil, 'content-type' => 'text/plain'
+    end
+
+    assert_equal 'text/plain', e.content_type
+
+    e = assert_raises Mechanize::ContentTypeError do
+      Mechanize::Page.new nil, 'content-type' => 'text/plain;charset=utf-8'
+    end
+
+    assert_equal 'text/plain;charset=utf-8', e.content_type
+  end
+
+  def test_override_content_type
+    page = Mechanize::Page.new nil, {'content-type' => 'text/html'}, WINDOWS_1255
+    assert page
+    assert_equal 'text/html; charset=windows-1255', page.content_type
+  end
+
+  def test_canonical_uri
+    page = @agent.get("http://localhost/canonical_uri.html")
+    assert_equal(URI("http://localhost/canonical_uri"), page.canonical_uri)
+
+    page = @agent.get("http://localhost/file_upload.html")
+    assert_equal(nil, page.canonical_uri)
+  end
+
+  def test_canonical_uri_unescaped
+    page = util_page <<-BODY
+<head>
+  <link rel="canonical" href="http://example/white space"/>
+</head>
+    BODY
+
+    assert_equal @uri + '/white%20space', page.canonical_uri
+  end
+
+  def test_charset_from_content_type
+    charset = Mechanize::Page.__send__ :charset_from_content_type, 'text/html;charset=UTF-8'
+
+    assert_equal 'UTF-8', charset
+  end
+
+  def test_charset_from_bad_content_type
+    charset = Mechanize::Page.__send__ :charset_from_content_type, 'text/html'
+
+    assert_nil charset
+  end
+
+  def test_encoding
+    page = util_page WINDOWS_1255
+
+    assert_equal 'windows-1255', page.encoding
+  end
+
+  def test_encoding_charset_after_title
+    page = util_page SJIS_AFTER_TITLE
+
+    assert_equal false, page.encoding_error?
+
+    assert_equal 'Shift_JIS', page.encoding
+  end
+
+  def test_encoding_charset_after_title_bad
+    page = util_page UTF8
+
+    assert_equal false, page.encoding_error?
+
+    assert_equal 'UTF-8', page.encoding
+  end
+
+  def test_encoding_charset_after_title_double_bad
+    page = util_page SJIS_BAD_AFTER_TITLE
+
+    assert_equal false, page.encoding_error?
+
+    assert_equal 'SHIFT_JIS', page.encoding
+  end
+
+  def test_encoding_charset_bad
+    page = util_page "<title>#{UTF8_TITLE}</title>"
+    page.encodings.replace %w[
+      UTF-8
+      Shift_JIS
+    ]
+
+    assert_equal false, page.encoding_error?
+
+    assert_equal 'UTF-8', page.encoding
+  end
+
+  def test_encoding_meta_charset
+    page = util_page "<meta charset='UTF-8'>"
+
+    assert_equal 'UTF-8', page.encoding
+  end
+
+  def test_encoding_equals
+    page = util_page
+    page.meta_refresh
+    assert page.instance_variable_get(:@meta_refresh)
+
+    page.encoding = 'UTF-8'
+
+    assert_nil page.instance_variable_get(:@meta_refresh)
+
+    assert_equal 'UTF-8', page.encoding
+    assert_equal 'UTF-8', page.parser.encoding
+  end
+
+  def test_page_encoding_error?
+    page = util_page
+    page.parser.errors.clear
+    assert_equal false, page.encoding_error?
+  end
+
+  def test_detect_libxml2error_indicate_encoding
+    page = util_page
+    page.parser.errors.clear
+
+    # error in libxml2-2.7.8/parser.c, HTMLparser.c or parserInternals.c
+    page.parser.errors = [ENCODING_ERROR_CLASS.new("Input is not proper UTF-8, indicate encoding !\n")]
+    assert_equal true, page.encoding_error?
+  end
+
+  def test_detect_libxml2error_invalid_char
+    page = util_page
+    page.parser.errors.clear
+
+    # error in libxml2-2.7.8/HTMLparser.c
+    page.parser.errors = [ENCODING_ERROR_CLASS.new("Invalid char in CDATA 0x%X\n")]
+    assert_equal true, page.encoding_error?
+  end
+
+  def test_detect_libxml2error_input_conversion_failed
+    page = util_page
+    page.parser.errors.clear
+
+    # error in libxml2-2.7.8/encoding.c
+    page.parser.errors = [ENCODING_ERROR_CLASS.new("input conversion failed due to input error\n")]
+    assert_equal true, page.encoding_error?
+  end
+
+  def test_detect_libxml2error_which_unsupported_by_mechanize
+    page = util_page
+    page.parser.errors.clear
+
+    # error in libxml2-2.7.8/HTMLparser.c
+    page.parser.errors = [ENCODING_ERROR_CLASS.new("encoder error\n")]
+    assert_equal false, page.encoding_error?
+  end
+
+  def test_encoding_equals_before_parser
+    # document has a bad encoding information - windows-1255
+    page = util_page BAD
+
+    # encoding is wrong, so user wants to force ISO-8859-2
+    page.encoding = 'ISO-8859-2'
+
+    assert_equal false, page.encoding_error?
+    assert_equal 'ISO-8859-2', page.encoding
+    assert_equal 'ISO-8859-2', page.parser.encoding
+  end
+
+  def test_encoding_equals_after_parser
+    # document has a bad encoding information - windows-1255
+    page = util_page BAD
+    page.parser
+
+    # autodetection sets encoding to windows-1255
+    assert_equal 'windows-1255', page.encoding
+    # believe in yourself, not machine
+    assert_equal false, page.encoding_error?
+
+    # encoding is wrong, so user wants to force ISO-8859-2
+    page.encoding = 'ISO-8859-2'
+
+    assert_equal false, page.encoding_error?
+    assert_equal 'ISO-8859-2', page.encoding
+    assert_equal 'ISO-8859-2', page.parser.encoding
+  end
+
+  def test_frames_with
+    page = @agent.get("http://localhost/frame_test.html")
+    assert_equal(3, page.frames.size)
+
+    find_orig = page.frames.find_all { |f| f.name == 'frame1' }
+
+    find1 = page.frames_with(:name => 'frame1')
+
+    find_orig.zip(find1).each { |a,b|
+      assert_equal(a, b)
+    }
+  end
+
+  def test_links_with_dom_id
+    page = @agent.get("http://localhost/tc_links.html")
+    link = page.links_with(:dom_id => 'bold_aaron_link')
+    link_by_id = page.links_with(:id => 'bold_aaron_link')
+    assert_equal(1, link.length)
+    assert_equal('Aaron Patterson', link.first.text)
+    assert_equal(link, link_by_id)
+  end
+
+  def test_link_with_encoded_space
+    page = @agent.get("http://localhost/tc_links.html")
+    link = page.link_with(:text => 'encoded space')
+    page = @agent.click link
+  end
+
+  def test_link_with_space
+    page = @agent.get("http://localhost/tc_links.html")
+    link = page.link_with(:text => 'not encoded space')
+    page = @agent.click link
+  end
+
+  def test_link_with_unusual_characters
+    page = @agent.get("http://localhost/tc_links.html")
+    link = page.link_with(:text => 'unusual characters')
+    assert_nothing_raised { @agent.click link }
+  end
+
+  def test_links
+    page = @agent.get("http://localhost/find_link.html")
+    assert_equal(18, page.links.length)
+  end
+
+  def test_links_with_bold
+    page = @agent.get("http://localhost/tc_links.html")
+    link = page.links_with(:text => /Bold Dude/)
+    assert_equal(1, link.length)
+    assert_equal('Bold Dude', link.first.text)
+    assert_equal [], link.first.rel
+    assert !link.first.rel?('me')
+    assert !link.first.rel?('nofollow')
+
+    link = page.links_with(:text => 'Aaron James Patterson')
+    assert_equal(1, link.length)
+    assert_equal('Aaron James Patterson', link.first.text)
+    assert_equal ['me'], link.first.rel
+    assert link.first.rel?('me')
+    assert !link.first.rel?('nofollow')
+
+    link = page.links_with(:text => 'Aaron Patterson')
+    assert_equal(1, link.length)
+    assert_equal('Aaron Patterson', link.first.text)
+    assert_equal ['me', 'nofollow'], link.first.rel
+    assert link.first.rel?('me')
+    assert link.first.rel?('nofollow')
+
+    link = page.links_with(:text => 'Ruby Rocks!')
+    assert_equal(1, link.length)
+    assert_equal('Ruby Rocks!', link.first.text)
+  end
+
+  def test_meta_refresh
+    page = @agent.get("http://localhost/find_link.html")
+    assert_equal(3, page.meta_refresh.length)
+    assert_equal(%w{
+      http://www.drphil.com/
+      http://www.upcase.com/
+      http://tenderlovemaking.com/ }.sort,
+      page.meta_refresh.map { |x| x.href.downcase }.sort)
+  end
+
+  def test_title
+    page = util_page
+
+    assert_equal('hi', page.title)
+  end
+
+  def test_title_none
+    page = util_page '' # invalid HTML
+
+    assert_equal(nil, page.title)
+  end
+
+  def test_page_decoded_with_charset
+    page = util_page @body, 'content-type' => 'text/html; charset=EUC-JP'
+
+    assert_equal 'EUC-JP', page.encoding
+    assert_equal 'EUC-JP', page.parser.encoding
+  end
+
+  def test_form
+    page  = @agent.get("http://localhost/tc_form_action.html")
+
+    form = page.form(:name => 'post_form1')
+    assert form
+    yielded = false
+
+    form = page.form(:name => 'post_form1') { |f|
+      yielded = true
+      assert f
+      assert_equal(form, f)
+    }
+
+    assert yielded
+
+    form_by_action = page.form(:action => '/form_post?a=b&b=c')
+    assert form_by_action
+    assert_equal(form, form_by_action)
+  end
+
+end
+