mechanize 0.9.2 → 0.9.3

data/test/chain/test_header_resolver.rb

@@ -0,0 +1,28 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', "helper"))
+
+class TestHeaderResolver < Test::Unit::TestCase
+  def setup
+    @chain = WWW::Mechanize::Chain.new([
+      WWW::Mechanize::Chain::HeaderResolver.new(
+        true,
+        300,
+        WWW::Mechanize::CookieJar.new,
+        'foobar',
+        {
+          'hello' => 'world',
+          'Content-Type' => 'utf-8'
+        }
+      )
+    ])
+  end
+
+  def test_handle
+    hash = {
+      :request => {},
+      :uri => URI.parse('http://google.com/')
+    }
+    @chain.handle(hash)
+    assert_equal 'world', hash[:request]['hello']
+    assert_equal 'utf-8', hash[:request]['Content-Type']
+  end
+end

data/test/helper.rb

@@ -48,6 +48,8 @@ class Net::HTTP
     '/infinite_redirect'      => InfiniteRedirectTest,
     '/infinite_refresh'       => InfiniteRefreshTest,
     '/redirect'               => RedirectTest,
+    '/refresh_without_url'    => RefreshWithoutUrl,
+    '/refresh_with_empty_url' => RefreshWithEmptyUrl,
     '/digest_auth'            => DigestAuthServlet,
     '/verb'                   => VerbServlet,
   }

data/test/servlets.rb

@@ -164,6 +164,32 @@ class InfiniteRefreshTest < WEBrick::HTTPServlet::AbstractServlet
   end
 end
 
+class RefreshWithoutUrl < WEBrick::HTTPServlet::AbstractServlet
+  @@count = 0
+  def do_GET(req, res)
+    res['Content-Type'] = "text/html"
+    @@count += 1
+    if @@count > 1
+      res['Refresh'] = "0; url=http://localhost/index.html";
+    else
+      res['Refresh'] = "0";
+    end
+  end
+end
+
+class RefreshWithEmptyUrl < WEBrick::HTTPServlet::AbstractServlet
+  @@count = 0
+  def do_GET(req, res)
+    res['Content-Type'] = "text/html"
+    @@count += 1
+    if @@count > 1
+      res['Refresh'] = "0; url=http://localhost/index.html";
+    else
+      res['Refresh'] = "0; url=";
+    end
+  end
+end
+
 class InfiniteRedirectTest < WEBrick::HTTPServlet::AbstractServlet
   def do_GET(req, res)
     res['Content-Type'] = req.query['ct'] || "text/html"

data/test/test_cookie_class.rb

@@ -43,7 +43,7 @@ class CookieClassTest < Test::Unit::TestCase
               "22-AUG-1993 12:59am",
               "22-AUG-1993 12:59 PM",
               #"Friday, August 04, 1995 3:54 PM",
-              "06/21/95 04:24:34 PM",
+              #"06/21/95 04:24:34 PM",
               #"20/06/95 21:07",
               "95-06-08 19:32:48 EDT",
     ]
@@ -62,9 +62,9 @@ class CookieClassTest < Test::Unit::TestCase
   def test_parse_weird_cookie
     cookie = 'n/a, ASPSESSIONIDCSRRQDQR=FBLDGHPBNDJCPCGNCPAENELB; path=/'
     url = URI.parse('http://www.searchinnovation.com/')
-    WWW::Mechanize::Cookie.parse(url, cookie) { |cookie|
-      assert_equal('ASPSESSIONIDCSRRQDQR', cookie.name)
-      assert_equal('FBLDGHPBNDJCPCGNCPAENELB', cookie.value)
+    WWW::Mechanize::Cookie.parse(url, cookie) { |c|
+      assert_equal('ASPSESSIONIDCSRRQDQR', c.name)
+      assert_equal('FBLDGHPBNDJCPCGNCPAENELB', c.value)
     }
   end
 

data/test/test_cookie_jar.rb

@@ -10,6 +10,25 @@ class CookieJarTest < Test::Unit::TestCase
     c
   end
 
+  def test_two_cookies_same_domain_and_name_different_paths
+    values = {  :name     => 'Foo',
+                :value    => 'Bar',
+                :path     => '/',
+                :expires  => Time.now + (10 * 86400),
+                :domain   => 'rubyforge.org'
+             }
+
+    url = URI.parse('http://rubyforge.org/')
+
+    jar = WWW::Mechanize::CookieJar.new
+    cookie = cookie_from_hash(values)
+    jar.add(url, cookie)
+    jar.add(url, cookie_from_hash(values.merge(:path => '/onetwo')))
+
+    assert_equal(1, jar.cookies(url).length)
+    assert_equal 2, jar.cookies(URI.parse('http://rubyforge.org/onetwo')).length
+  end
+
   def test_domain_case
     values = {  :name     => 'Foo',
                 :value    => 'Bar',
@@ -30,7 +49,7 @@ class CookieJarTest < Test::Unit::TestCase
     jar.add(url, cookie_from_hash( values.merge(  :domain => 'RuByForge.Org',
                                                   :name   => 'aaron'
                                                ) ) )
-                  
+
     assert_equal(2, jar.cookies(url).length)
 
     url2 = URI.parse('http://RuByFoRgE.oRg/')
@@ -57,7 +76,7 @@ class CookieJarTest < Test::Unit::TestCase
     jar.add(url, cookie_from_hash( values.merge(  :domain => 'RuByForge.Org',
                                                   :name   => 'aaron'
                                                ) ) )
-                  
+
     assert_equal(2, jar.cookies(url).length)
 
     url2 = URI.parse('http://RuByFoRgE.oRg/')
@@ -240,11 +259,11 @@ class CookieJarTest < Test::Unit::TestCase
     # When given a URI with a blank path, CookieJar#cookies should return
     # cookies with the path '/':
     url = URI.parse('http://rubyforge.org')
-    assert_equal '', url.path    
-    assert_equal(0, jar.cookies(url).length)    
+    assert_equal '', url.path
+    assert_equal(0, jar.cookies(url).length)
     # Now add a cookie with the path set to '/':
-    jar.add(url, cookie_from_hash(values.merge( :name => 'has_root_path', 
-                                          :path => '/')))    
+    jar.add(url, cookie_from_hash(values.merge( :name => 'has_root_path',
+                                          :path => '/')))
     assert_equal(1, jar.cookies(url).length)
   end
 
@@ -282,8 +301,8 @@ class CookieJarTest < Test::Unit::TestCase
                                           :expires => Time.now - (10 * 86400))))
     assert_equal(0, jar.cookies(url).length)
   end
-  
-  
+
+
   def test_save_and_read_cookiestxt
     values = {  :name     => 'Foo',
                 :value    => 'Bar',
@@ -301,7 +320,7 @@ class CookieJarTest < Test::Unit::TestCase
     jar.add(url, cookie)
     jar.add(url, cookie_from_hash(values.merge( :name => 'Baz' )))
     assert_equal(2, jar.cookies(url).length)
-    
+
     jar.save_as("cookies.txt", :cookiestxt)
     jar.clear!
     assert_equal(0, jar.cookies(url).length)

data/test/test_follow_meta.rb

@@ -11,6 +11,17 @@ class FollowMetaTest < Test::Unit::TestCase
     assert_equal(1, page.meta.length)
   end
 
+  def test_meta_refresh_does_not_send_referer
+    @agent.follow_meta_refresh = true
+    requests = []
+    @agent.pre_connect_hooks << lambda { |params|
+      requests << params[:request]
+    }
+
+    page = @agent.get('http://localhost/tc_follow_meta.html')
+    assert_nil requests[1]['referer']
+  end
+
   def test_follow_meta_if_set
     @agent.follow_meta_refresh = true
 
@@ -22,6 +33,34 @@ class FollowMetaTest < Test::Unit::TestCase
     assert_equal('http://localhost/index.html', page.uri.to_s)
     assert_equal('http://localhost/index.html', @agent.history.last.uri.to_s)
   end
+  
+  def test_follow_meta_with_empty_url
+    @agent.follow_meta_refresh = true
+  
+    page = @agent.get('http://localhost/refresh_with_empty_url')
+  
+    assert_equal(3, @agent.history.length)
+    assert_equal('http://localhost/refresh_with_empty_url',
+                 @agent.history[0].uri.to_s)
+    assert_equal('http://localhost/refresh_with_empty_url',
+                 @agent.history[1].uri.to_s)
+    assert_equal('http://localhost/index.html', page.uri.to_s)
+    assert_equal('http://localhost/index.html', @agent.history.last.uri.to_s)
+  end
+  
+  def test_follow_meta_without_url
+    @agent.follow_meta_refresh = true
+  
+    page = @agent.get('http://localhost/refresh_without_url')
+  
+    assert_equal(3, @agent.history.length)
+    assert_equal('http://localhost/refresh_without_url',
+                 @agent.history[0].uri.to_s)
+    assert_equal('http://localhost/refresh_without_url',
+                 @agent.history[1].uri.to_s)
+    assert_equal('http://localhost/index.html', page.uri.to_s)
+    assert_equal('http://localhost/index.html', @agent.history.last.uri.to_s)
+  end
 
   def test_always_follow_302
     @agent.follow_meta_refresh = false

data/test/test_forms.rb

@@ -23,17 +23,6 @@ class FormsMechTest < Test::Unit::TestCase
     assert_equal 'bar', headers['foo']
   end
 
-  def test_submit_takes_arbirary_headers
-    page = @agent.get('http://localhost:2000/form_no_action.html')
-    assert form = page.forms.first
-    form.action = '/http_headers'
-    page = form.submit(nil, { 'foo' => 'bar' })
-    headers = Hash[*(
-      page.body.split("\n").map { |x| x.split('|') }.flatten
-    )]
-    assert_equal 'bar', headers['foo']
-  end
-
   # Test submitting form with two fields of the same name
   def test_post_multival
     page = @agent.get("http://localhost/form_multival.html")

data/test/test_mech.rb

@@ -10,6 +10,13 @@ class TestMechMethods < Test::Unit::TestCase
     assert_equal('http://localhost/?foo=~2', page.uri.to_s)
   end
 
+  def test_parser_can_be_set
+    @agent.html_parser = {}
+    assert_raises(NoMethodError) {
+      @agent.get('http://localhost/?foo=~2').links
+    }
+  end
+
   def test_submit_takes_arbirary_headers
     page = @agent.get('http://localhost:2000/form_no_action.html')
     assert form = page.forms.first
@@ -91,12 +98,14 @@ class TestMechMethods < Test::Unit::TestCase
     }
   end
 
-  def test_kcode_url
-    $KCODE = 'u'
-    page = @agent.get("http://localhost/?a=#{[0xd6].pack('U')}")
-    assert_not_nil(page)
-    assert_equal('http://localhost/?a=%D6', page.uri.to_s)
-    $KCODE = 'NONE'
+  unless RUBY_VERSION >= '1.9.0'
+    def test_kcode_url
+      $KCODE = 'u'
+      page = @agent.get("http://localhost/?a=#{[0xd6].pack('U')}")
+      assert_not_nil(page)
+      assert_equal('http://localhost/?a=%D6', page.uri.to_s)
+      $KCODE = 'NONE'
+    end
   end
 
   def test_history

data/test/test_meta.rb

@@ -0,0 +1,65 @@
+require File.expand_path(File.join(File.dirname(__FILE__), "helper"))
+
+class MetaTest < Test::Unit::TestCase
+  Meta = WWW::Mechanize::Page::Meta
+  
+  #
+  # CONTENT_REGEXP test
+  #
+  
+  def test_content_regexp
+    r = Meta::CONTENT_REGEXP
+    
+    assert r =~ "0; url=http://localhost:8080/path"
+    assert_equal "0", $1
+    assert_equal "http://localhost:8080/path", $3
+    
+    assert r =~ "100.001; url=http://localhost:8080/path"
+    assert_equal "100.001", $1
+    assert_equal "http://localhost:8080/path", $3
+    
+    assert r =~ "0; url='http://localhost:8080/path'"
+    assert_equal "0", $1
+    assert_equal "http://localhost:8080/path", $3
+    
+    assert r =~ "0; url=\"http://localhost:8080/path\""
+    assert_equal "0", $1
+    assert_equal "http://localhost:8080/path", $3
+    
+    assert r =~ "0; url="
+    assert_equal "0", $1
+    assert_equal "", $3
+    
+    assert r =~ "0"
+    assert_equal "0", $1
+    assert_equal nil, $3
+    
+    assert r =~ "   0;   "
+    assert_equal "0", $1
+    assert_equal nil, $3
+    
+    assert r =~ "0; UrL=http://localhost:8080/path"
+    assert_equal "0", $1
+    assert_equal "http://localhost:8080/path", $3
+  end
+  
+  #
+  # parse test
+  #
+  
+  def test_parse_documentation
+    uri = URI.parse('http://current.com/')
+  
+    assert_equal ['5', 'http://example.com/'], Meta.parse("5;url=http://example.com/", uri)
+    assert_equal ['5', 'http://current.com/'], Meta.parse("5;url=", uri) 
+    assert_equal ['5', 'http://current.com/'], Meta.parse("5", uri) 
+    assert_equal nil, Meta.parse("invalid content", uri)
+  end
+  
+  def test_parse_returns_nil_if_no_delay_and_url_can_be_parsed
+    uri = URI.parse('http://current.com/')
+    
+    assert_equal nil, Meta.parse("invalid content", uri)
+    assert_equal nil, Meta.parse("invalid content", uri) {|delay, url| 'not nil' }
+  end
+end

data/test/test_page.rb

@@ -7,17 +7,57 @@ class TestPage < Test::Unit::TestCase
     @agent = WWW::Mechanize.new
   end
 
+  def test_page_gets_charset_from_page
+    page = @agent.get("http://localhost/tc_charset.html")
+    assert_equal 'windows-1255', page.encoding
+  end
+
+  def test_double_semicolon
+    page = @agent.get("http://localhost/http_headers?content-disposition=#{CGI.escape('attachment;; filename=fooooo')}")
+    assert page.parser
+  end
+
   def test_broken_charset
     page = @agent.get("http://localhost/http_headers?content-type=#{CGI.escape('text/html; charset=akldsjfhaldjfksh')}")
     assert page.parser
   end
 
+  def test_mostly_broken_charset
+    page = @agent.get("http://localhost/http_headers?content-type=#{CGI.escape('text/html; charset=ISO_8859-1')}")
+    assert_equal 'ISO_8859-1', page.encoding
+  end
+
+  def test_another_mostly_broken_charset
+    page = @agent.get("http://localhost/http_headers?content-type=#{CGI.escape('text/html; charset=UTF8')}")
+    assert_equal 'UTF8', page.parser.encoding
+    assert_equal 'UTF8', page.encoding
+  end
+
   def test_upper_case_content_type
     page = @agent.get("http://localhost/http_headers?content-type=#{CGI.escape('text/HTML')}")
     assert_instance_of WWW::Mechanize::Page, page
     assert_equal 'text/HTML', page.content_type
   end
 
+  def test_encoding_override_before_parser_initialized
+    # document has a bad encoding information - windows-1255
+    page = @agent.get("http://localhost/tc_bad_charset.html")
+    # encoding is wrong, so user wants to force ISO-8859-2
+    page.encoding = 'ISO-8859-2'
+    assert_equal 'ISO-8859-2', page.encoding
+  end
+
+  def test_encoding_override_after_parser_was_initialized
+    # document has a bad encoding information - windows-1255
+    page = @agent.get("http://localhost/tc_bad_charset.html")
+    page.parser
+    # autodetection sets encoding to windows-1255
+    assert_equal 'windows-1255', page.encoding
+    # encoding is wrong, so user wants to force ISO-8859-2
+    page.encoding = 'ISO-8859-2'
+    assert_equal 'ISO-8859-2', page.encoding
+  end
+
   def test_page_gets_charset_sent_by_server
     page = @agent.get("http://localhost/http_headers?content-type=#{CGI.escape('text/html; charset=UTF-8')}")
     assert_equal 'UTF-8', page.encoding
@@ -48,6 +88,17 @@ class TestPage < Test::Unit::TestCase
     assert_equal(nil, page.title)
   end
 
+  def test_page_decoded_with_charset
+    page = WWW::Mechanize::Page.new(
+      URI.parse('http://tenderlovemaking.com/'),
+      { 'content-type' => 'text/html; charset=EUC-JP' },
+      '<html><body>hello</body></html>',
+      400,
+      @agent
+    )
+    assert_equal 'EUC-JP', page.parser.encoding
+  end
+
   def test_find_form_with_hash
     page  = @agent.get("http://localhost/tc_form_action.html")
     form = page.form(:name => 'post_form1')
@@ -65,3 +116,4 @@ class TestPage < Test::Unit::TestCase
     assert_equal(form, form_by_action)
   end
 end
+

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: mechanize
 version: !ruby/object:Gem::Version 
-  version: 0.9.2
+  version: 0.9.3
 platform: ruby
 authors: 
 - Aaron Patterson
@@ -10,7 +10,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-03-05 00:00:00 -08:00
+date: 2009-06-08 00:00:00 -07:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -31,9 +31,14 @@ dependencies:
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
-        version: 1.9.0
+        version: 1.12.2
     version: 
-description: The Mechanize library is used for automating interaction with websites.  Mechanize automatically stores and sends cookies, follows redirects, can follow links, and submit forms.  Form fields can be populated and submitted.  Mechanize also keeps track of the sites that you have visited as a history.
+description: |-
+  The Mechanize library is used for automating interaction with websites. 
+  Mechanize automatically stores and sends cookies, follows redirects,
+  can follow links, and submit forms.  Form fields can be populated and
+  submitted.  Mechanize also keeps track of the sites that you have visited as
+  a history.
 email: 
 - aaronp@rubyforge.org
 - mike.dalessio@gmail.com
@@ -220,7 +225,9 @@ files:
 - test/test_upload.rb
 - test/test_verbs.rb
 has_rdoc: true
-homepage: "  http://mechanize.rubyforge.org/"
+homepage: http://mechanize.rubyforge.org/
+licenses: []
+
 post_install_message: 
 rdoc_options: 
 - --main
@@ -242,13 +249,14 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: mechanize
-rubygems_version: 1.3.1
+rubygems_version: 1.3.3
 signing_key: 
-specification_version: 2
+specification_version: 3
 summary: Mechanize provides automated web-browsing
 test_files: 
 - test/chain/test_argument_validator.rb
 - test/chain/test_custom_headers.rb
+- test/chain/test_header_resolver.rb
 - test/chain/test_parameter_resolver.rb
 - test/chain/test_request_resolver.rb
 - test/chain/test_response_reader.rb
@@ -280,6 +288,7 @@ test_files:
 - test/test_links.rb
 - test/test_mech.rb
 - test/test_mechanize_file.rb
+- test/test_meta.rb
 - test/test_multi_select.rb
 - test/test_no_attributes.rb
 - test/test_option.rb