mechanize 2.1.pre.1 → 2.1

data/CHANGELOG.rdoc

@@ -1,6 +1,6 @@
 = Mechanize CHANGELOG
 
-=== 2.1 / ??
+=== 2.1 / 2011-12-20
 
 * Deprecations
   * Mechanize#get no longer accepts an options hash.
@@ -9,8 +9,9 @@
 * Minor enhancements
   * Mechanize now depends on net-http-persistent 2.3+.  This new version
     brings idle timeouts to help with the dreaded "too many connection resets"
-    issue when POSTing to a closed connection.  Also, SSL connections will be
-    verified against the system certificate store by default.  Issue #123
+    issue when POSTing to a closed connection.  Issue #123
+  * SSL connections will be verified against the system certificate store by
+    default.
   * Added Mechanize#retry_change_requests to allow mechanize to retry POST and
     other non-idempotent requests when you know it is safe to do so.  Issue
     #123
@@ -39,8 +40,8 @@
     by Jo Hund.
   * Updated 'Mac Safari' User-Agent alias to Safari 5.1.1.  'Mac Safari 4' can
     be used for the old 'Mac Safari' alias.
-  * When given multiple HTTP authentication options mechanize can better pick
-    the strongest method.
+  * When given multiple HTTP authentication options mechanize now picks the
+    strongest method.
   * Improvements to HTTP authorization:
     * mechanize raises Mechanize::UnathorizedError for 401 responses which is
       a sublcass of Mechanize::ResponseCodeError.
@@ -52,10 +53,6 @@
     Issue #167
 
 * Bug fixes
-  * Mechanize takes more care to avoid saving files with certain unsafe names.
-    You should still take care not to use mechanize to save files into your
-    home directory.  Issue #163.
-  * Mechanize#cookie_jar= works again.  Issue #126
   * Mechanize now handles cookies just as most modern browsers do,
     roughly based on RFC 6265.
     * domain=.example.com (which is invalid) is considered identical to
@@ -75,6 +72,10 @@
     * It is recommended that you clear out existing cookie jars for
       regeneration because previously saved cookies may not have been
       parsed correctly.
+  * Mechanize takes more care to avoid saving files with certain unsafe names.
+    You should still take care not to use mechanize to save files directly
+    into your home directory ($HOME).  Issue #163.
+  * Mechanize#cookie_jar= works again.  Issue #126
   * The original Referer value persists on redirection.  Issue #150
   * Do not send a referer on a Refresh header based redirection.
   * Fixed encoding error in tests when LANG=C.  Patch #142 by jinschoi.

data/lib/mechanize.rb

@@ -308,7 +308,7 @@ class Mechanize
     # FIXME: Huge hack so that using a URI as a referer works.  I need to
     # refactor everything to pass around URIs but still support
     # Mechanize::Page#base
-    unless referer.is_a?(Mechanize::File)
+    unless Mechanize::Parser === referer then
       referer = referer.is_a?(String) ?
       Page.new(URI.parse(referer), {'content-type' => 'text/html'}) :
         Page.new(referer, {'content-type' => 'text/html'})
@@ -975,7 +975,7 @@ class Mechanize
     # Find our pluggable parser
     parser_klass = @pluggable_parser.parser content_type
 
-    unless Mechanize::Download === parser_klass then
+    unless parser_klass <= Mechanize::Download then
       body = case body
              when IO, Tempfile, StringIO then
                body.read

data/lib/mechanize/cookie.rb

@@ -74,6 +74,9 @@ class Mechanize::Cookie
   alias for_domain? for_domain
 
   class << self
+    # Parses a Set-Cookie header line +str+ sent from +uri+ into an
+    # array of Cookie objects.  Note that this array may contain
+    # nil's when some of the cookie-pairs are malformed.
     def parse(uri, str, log = Mechanize.log)
       return str.split(/,(?=[^;,]*=)|,$/).map { |c|
         cookie_elem = c.split(/;+/)
@@ -96,8 +99,12 @@ class Mechanize::Cookie
 
           case key.downcase
           when 'domain'
-            cookie.domain = value
-            cookie.for_domain = true
+            begin
+              cookie.domain = value
+              cookie.for_domain = true
+            rescue
+              log.warn("Couldn't parse domain: #{value}") if log
+            end
           when 'path'
             cookie.path = value
           when 'expires'
@@ -147,15 +154,22 @@ class Mechanize::Cookie
   # Sets the domain attribute.  A leading dot in +domain+ implies
   # turning the +for_domain?+ flag on.
   def domain=(domain)
-    if domain.start_with?('.')
-      @for_domain = true
-      domain = domain[1..-1]
-    end
-    # Do we really need to support this?
-    if domain.match(/\A([^:]+):[0-9]+\z/)
-      domain = $1
+    if DomainName === domain
+      @domain_name = domain
+    else
+      domain.is_a?(String) or
+        (domain.respond_to?(:to_str) && (domain = domain.to_str).is_a?(String)) or
+        raise TypeError, "#{domain.class} is not a String"
+      if domain.start_with?('.')
+        @for_domain = true
+        domain = domain[1..-1]
+      end
+      # Do we really need to support this?
+      if domain.match(/\A([^:]+):[0-9]+\z/)
+        domain = $1
+      end
+      @domain_name = DomainName.new(domain)
     end
-    @domain_name = DomainName.new(domain)
     set_domain(@domain_name.hostname)
   end
 

data/lib/mechanize/download.rb

@@ -51,7 +51,7 @@ class Mechanize::Download
         end
       end
     else
-      FileUtils.cp @body_io.path, filename
+      FileUtils.mv @body_io.path, filename
     end
   end
 

data/lib/mechanize/http/agent.rb

@@ -695,6 +695,7 @@ class Mechanize::HTTP::Agent
       begin
         zio = Zlib::GzipReader.new body_io
         out_io = Tempfile.new 'mechanize-decode'
+        out_io.binmode
 
         until zio.eof? do
           out_io.write zio.read 16384
@@ -723,16 +724,9 @@ class Mechanize::HTTP::Agent
   end
 
   def response_cookies response, uri, page
-    log = log()	 # reduce method calls
     if Mechanize::Page === page and page.body =~ /Set-Cookie/n
       page.search('//head/meta[@http-equiv="Set-Cookie"]').each do |meta|
-        Mechanize::Cookie.parse(uri, meta['content'], log) { |c|
-          if @cookie_jar.add(uri, c)
-            log.debug("saved cookie: #{c}") if log
-          else
-            log.debug("rejected cookie: #{c}") if log
-          end
-        }
+        save_cookies(uri, meta['content'])
       end
     end
 
@@ -740,17 +734,22 @@ class Mechanize::HTTP::Agent
 
     return unless header_cookies
 
-    header_cookies.each do |cookie|
-      Mechanize::Cookie.parse(uri, cookie, log) { |c|
-        if @cookie_jar.add(uri, c)
-          log.debug("saved cookie: #{c}") if log
-        else
-          log.debug("rejected cookie: #{c}") if log
-        end
-      }
+    header_cookies.each do |set_cookie|
+      save_cookies(uri, set_cookie)
     end
   end
 
+  def save_cookies(uri, set_cookie)
+    log = log()	 # reduce method calls
+    Mechanize::Cookie.parse(uri, set_cookie, log) { |c|
+      if @cookie_jar.add(uri, c)
+        log.debug("saved cookie: #{c}") if log
+      else
+        log.debug("rejected cookie: #{c}") if log
+      end
+    }
+  end
+
   def response_follow_meta_refresh response, uri, page, redirects
     delay, new_url = get_meta_refresh(response, uri, page)
     return nil unless new_url

data/lib/mechanize/test_case.rb

@@ -43,6 +43,10 @@ class Mechanize::TestCase < MiniTest::Unit::TestCase
     Mechanize::Page.new uri, response, html, 200, agent
   end
 
+  def have_encoding?
+    Object.const_defined? :Encoding
+  end
+
   def html_page body
     uri = URI 'http://example/'
     Mechanize::Page.new uri, { 'content-type' => 'text/html' }, body, 200, @mech

data/test/test_mechanize.rb

@@ -452,6 +452,19 @@ but not <a href="/" rel="me nofollow">this</a>!
     assert_nil requests.last['referer']
   end
 
+  def test_get_referer_download
+    download = Mechanize::Download.new URI 'http://example/prev'
+
+    uri = URI 'http://example'
+
+    page = @mech.get uri, { :q => 'h' }, download, { 'X-H' => 'v' }
+
+    assert_equal URI('http://example/?q=h'), page.uri
+    assert_equal URI('http://example'), uri
+
+    assert_equal 'http://example/prev', requests.first['referer']
+  end
+
   def test_get_robots
     @mech.robots = true
 
@@ -791,6 +804,7 @@ but not <a href="/" rel="me nofollow">this</a>!
     download = @mech.parse @uri, response, StringIO.new('raw')
 
     assert_kind_of Mechanize::Download, download
+    assert_kind_of StringIO, download.content
   end
 
   def test_parse_html

data/test/test_mechanize_cookie.rb

@@ -406,5 +406,46 @@ class TestMechanizeCookie < Mechanize::TestCase
       end
     end
   end
+
+  def test_new
+    cookie = Mechanize::Cookie.new('key', 'value')
+    assert_equal 'key', cookie.name
+    assert_equal 'value', cookie.value
+    assert_equal nil, cookie.expires
+
+    # Minimum unit for the expires attribute is second
+    expires = Time.at((Time.now + 3600).to_i)
+
+    cookie = Mechanize::Cookie.new('key', 'value', :expires => expires.dup)
+    assert_equal 'key', cookie.name
+    assert_equal 'value', cookie.value
+    assert_equal expires, cookie.expires
+
+    cookie = Mechanize::Cookie.new(:value => 'value', :name => 'key', :expires => expires.dup)
+    assert_equal 'key', cookie.name
+    assert_equal 'value', cookie.value
+    assert_equal expires, cookie.expires
+  end
+
+  def test_domain=
+    url = URI.parse('http://host.dom.example.com:8080/')
+
+    cookie_str = 'a=b; domain=Example.Com'
+    cookie = Mechanize::Cookie.parse(url, cookie_str).first
+    assert 'example.com', cookie.domain
+
+    cookie.domain = DomainName(url.host)
+    assert 'host.dom.example.com', cookie.domain
+
+    cookie.domain = 'Dom.example.com'
+    assert 'dom.example.com', cookie.domain
+
+    cookie.domain = Object.new.tap { |o|
+      def o.to_str
+        'Example.com'
+      end
+    }
+    assert 'example.com', cookie.domain
+  end
 end
 

data/test/test_mechanize_cookie_jar.rb

@@ -47,13 +47,19 @@ class TestMechanizeCookieJar < Mechanize::TestCase
     assert_equal(2, @jar.cookies(url2).length)
   end
 
-  def test_no_domain_case
-    url = URI.parse('http://www.rubyforge.org/')
+  def test_host_only
+    url = URI.parse('http://rubyforge.org/')
 
     @jar.add(url, Mechanize::Cookie.new(
         cookie_values(:domain => 'rubyforge.org', :for_domain => false)))
 
-    assert_equal(0, @jar.cookies(url).length)
+    assert_equal(1, @jar.cookies(url).length)
+
+    assert_equal(1, @jar.cookies(URI('http://RubyForge.org/')).length)
+
+    assert_equal(1, @jar.cookies(URI('https://RubyForge.org/')).length)
+
+    assert_equal(0, @jar.cookies(URI('http://www.rubyforge.org/')).length)
   end
 
   def test_empty_value

data/test/test_mechanize_http_agent.rb

@@ -601,6 +601,26 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
     assert_equal 'part', body.read
   end
 
+  def test_response_content_encoding_gzip_encoding_bad
+    def @res.content_length() 24 end
+    @res.instance_variable_set(:@header,
+                               'content-encoding' => %w[gzip],
+                               'content-type' => 'text/html; charset=UTF-8')
+
+    # "test\xB2"
+    body_io = StringIO.new \
+      "\037\213\b\000*+\314N\000\003+I-.\331\004\000x\016\003\376\005\000\000\000"
+
+    body = @agent.response_content_encoding @res, body_io
+
+    expected = "test\xB2"
+    expected.force_encoding Encoding::BINARY if have_encoding?
+
+    content = body.read
+    assert_equal expected, content
+    assert_equal Encoding::BINARY, content.encoding if have_encoding?
+  end
+
   def test_response_content_encoding_none
     def @res.content_length() 4 end
     @res.instance_variable_set :@header, 'content-encoding' => %w[none]

data/test/test_mechanize_page_encoding.rb

@@ -168,8 +168,8 @@ class TestMechanizePageEncoding < Mechanize::TestCase
   end
 
   def test_parser_encoding_when_searching_elements
-    skip "Encoding not implemented" unless Object.const_defined? :Encoding
-    
+    skip "Encoding not implemented" unless have_encoding?
+
     body = '<span id="latin1">hi</span>'
     page = util_page body, 'content-type' => 'text/html,charset=ISO-8859-1'
 

metadata

@@ -1,14 +1,12 @@
 --- !ruby/object:Gem::Specification 
 name: mechanize
 version: !ruby/object:Gem::Version 
-  hash: 1923831923
+  hash: 1
   prerelease: 
   segments: 
   - 2
   - 1
-  - pre
-  - 1
-  version: 2.1.pre.1
+  version: "2.1"
 platform: ruby
 authors: 
 - Eric Hodel
@@ -40,7 +38,7 @@ cert_chain:
   x52qPcexcYZR7w==
   -----END CERTIFICATE-----
 
-date: 2011-11-09 00:00:00 Z
+date: 2011-12-20 00:00:00 Z
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: net-http-digest_auth
@@ -418,7 +416,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: mechanize
-rubygems_version: 1.8.10
+rubygems_version: 1.8.12
 signing_key: 
 specification_version: 3
 summary: The Mechanize library is used for automating interaction with websites