mechanize 2.2.1 → 2.3

data/CHANGELOG.rdoc

@@ -1,5 +1,25 @@
 = Mechanize CHANGELOG
 
+=== 2.3 / 2012-02-20
+
+* Minor enhancement
+  * Add support for the Max-Age attribute in the Set-Cookie header.
+  * Added Mechanize::Download#body for compatibility with Mechanize::File when
+    using Mechanize#get_file with Mechanize::Image or other Download-based
+    pluggable parser.  Issue #202 by angas
+  * Mechanize#max_file_buffer may be set to nil to disable creation of
+    Tempfiles.
+
+* Bug fixes
+  * Applied Mechanize#max_file_buffer to the Content-Encoding handlers as well
+    to prevent extra Tempfiles for small gzip or deflate response
+  * Increased the default Mechanize#max_file_buffer to 100,000 bytes.  This
+    gives ~5MB of response bodies in memory with the default history setting
+    of 50 pages (depending on GC behavior).
+  * Ignore empty path/domain attributes.
+  * Cookies with an empty Expires attribute value were stored as session
+    cookies but cookies without the Expires attribute were not.  Issue #78
+
 === 2.2.1 / 2012-02-13
 
 * Bug fixes

data/README.rdoc

@@ -1,7 +1,7 @@
 = Mechanize
 
 * http://mechanize.rubyforge.org
-* http://github.com/tenderlove/mechanize/tree/master
+* https://github.com/tenderlove/mechanize
 
 == DESCRIPTION
 
@@ -24,7 +24,7 @@ The mechanize mailing list is available here:
 
 The bug tracker is available here:
 
-  * http://github.com/tenderlove/mechanize/issues
+  * https://github.com/tenderlove/mechanize/issues
 
 == Examples
 
@@ -40,7 +40,7 @@ Copyright (c) 2006-2011:
 * {Aaron Patterson}[http://tenderlovemaking.com] (aaronp@rubyforge.org)
 * {Mike Dalessio}[http://mike.daless.io] (mike@csa.net)
 
-Copyright (c) 2011:
+Copyright (c) 2011-2012:
 
 * {Eric Hodel}[http://blog.segment7.net] (drbrain@segment7.net)
 * {Akinori MUSHA}[http://blog.akinori.org] (knu@idaemons.org)

data/lib/mechanize.rb

@@ -73,7 +73,7 @@ class Mechanize
   ##
   # The version of Mechanize you are using.
 
-  VERSION = '2.2.1'
+  VERSION = '2.3'
 
   ##
   # Base mechanize error class
@@ -207,7 +207,9 @@ class Mechanize
   end
 
   ##
-  # Maximum number of items allowed in the history.
+  # Maximum number of items allowed in the history.  The default setting is 50
+  # pages.  Note that the size of the history multiplied by the maximum
+  # response body size 
 
   def max_history
     @agent.history.max_size
@@ -217,9 +219,11 @@ class Mechanize
   # Sets the maximum number of items allowed in the history to +length+.
   #
   # Setting the maximum history length to nil will make the history size
-  # unlimited.  Take care when doing this, mechanize stores page bodies in the
-  # temporary files directory for pages in the history.  For a long-running
-  # mechanize program this can be quite large.
+  # unlimited.  Take care when doing this, mechanize stores response bodies in
+  # memory for pages and in the temporary files directory for other responses.
+  # For a long-running mechanize program this can be quite large.
+  #
+  # See also the discussion under #max_file_buffer=
 
   def max_history= length
     @agent.history.max_size = length
@@ -295,7 +299,7 @@ class Mechanize
       end
       if link.noreferrer?
         href = @agent.resolve(link.href, link.page || current_page)
-        referer = Page.new(nil, {'content-type'=>'text/html'})
+        referer = Page.new
       else
         href = link.href
       end
@@ -382,9 +386,9 @@ class Mechanize
 
     referer ||=
       if uri.to_s =~ %r{\Ahttps?://}
-        Page.new(nil, 'content-type' => 'text/html')
+        Page.new
       else
-        current_page || Page.new(nil, 'content-type' => 'text/html')
+        current_page || Page.new
       end
 
     # FIXME: Huge hack so that using a URI as a referer works.  I need to
@@ -392,9 +396,9 @@ class Mechanize
     # Mechanize::Page#base
     unless Mechanize::Parser === referer then
       referer = if referer.is_a?(String) then
-                  Page.new URI(referer), 'content-type' => 'text/html'
+                  Page.new URI(referer)
                 else
-                  Page.new referer, 'content-type' => 'text/html'
+                  Page.new referer
                 end
     end
 
@@ -479,7 +483,7 @@ class Mechanize
   # as the request body, if allowed.
 
   def request_with_entity(verb, uri, entity, headers = {})
-    cur_page = current_page || Page.new(nil, {'content-type'=>'text/html'})
+    cur_page = current_page || Page.new
 
     headers = {
       'Content-Type' => 'application/octet-stream',
@@ -757,7 +761,9 @@ class Mechanize
 
   ##
   # Responses larger than this will be written to a Tempfile instead of stored
-  # in memory.  The default is 10240 bytes
+  # in memory.  The default is 100,000 bytes.
+  #
+  # A value of nil disables creation of Tempfiles.
 
   def max_file_buffer
     @agent.max_file_buffer
@@ -765,7 +771,16 @@ class Mechanize
 
   ##
   # Sets the maximum size of a response body that will be stored in memory to
-  # +bytes+
+  # +bytes+.  A value of nil causes all response bodies to be stored in
+  # memory.
+  #
+  # Note that for Mechanize::Download subclasses, the maximum buffer size
+  # multiplied by the number of pages stored in history (controlled by
+  # #max_history) is an approximate upper limit on the amount of memory
+  # Mechanize will use.  By default, Mechanize can use up to ~5MB to store
+  # response bodies for non-File and non-Page (HTML) responses.
+  #
+  # See also the discussion under #max_history=
 
   def max_file_buffer= bytes
     @agent.max_file_buffer = bytes
@@ -1152,7 +1167,7 @@ class Mechanize
 
   def post_form(uri, form, headers = {})
     cur_page = form.page || current_page ||
-      Page.new(nil, {'content-type'=>'text/html'})
+      Page.new
 
     request_data = form.request_data
 

data/lib/mechanize/cookie.rb

@@ -1,5 +1,5 @@
 require 'time'
-require 'webrick/cookie'
+require 'webrick/httputils'
 require 'domain_name'
 
 # This class is used to represent an HTTP Cookie.
@@ -93,12 +93,13 @@ class Mechanize::Cookie
 
         cookie_elem.each do |pair|
           pair.strip!
-          key, value = pair.split(/\=/, 2)
+          key, value = pair.split(/=/, 2)
           next unless key
           value = WEBrick::HTTPUtils.dequote(value.strip) if value
 
           case key.downcase
           when 'domain'
+            next unless value && !value.empty?
             begin
               cookie.domain = value
               cookie.for_domain = true
@@ -106,27 +107,27 @@ class Mechanize::Cookie
               log.warn("Couldn't parse domain: #{value}") if log
             end
           when 'path'
+            next unless value && !value.empty?
             cookie.path = value
           when 'expires'
-            if value.empty?
-              cookie.session = true
-              next
-            end
-
+            next unless value && !value.empty?
             begin
               cookie.expires = Time::parse(value)
             rescue
               log.warn("Couldn't parse expires: #{value}") if log
             end
           when 'max-age'
+            next unless value && !value.empty?
             begin
               cookie.max_age = Integer(value)
             rescue
               log.warn("Couldn't parse max age '#{value}'") if log
             end
           when 'comment'
+            next unless value
             cookie.comment = value
           when 'version'
+            next unless value
             begin
               cookie.version = Integer(value)
             rescue
@@ -141,6 +142,11 @@ class Mechanize::Cookie
         cookie.path    ||= (uri + './').path
         cookie.secure  ||= false
         cookie.domain  ||= uri.host
+
+        # RFC 6265 4.1.2.2
+        cookie.expires   = Time.now + cookie.max_age if cookie.max_age
+        cookie.session   = !cookie.expires
+
         # Move this in to the cookie jar
         yield cookie if block_given?
 

data/lib/mechanize/download.rb

@@ -40,6 +40,15 @@ class Mechanize::Download
     yield self if block_given?
   end
 
+  ##
+  # The body of this response as a String.
+  #
+  # Take care, this may use lots of memory if the response body is large.
+
+  def body
+    @body_io.read.tap { @body_io.rewind }
+  end
+
   ##
   # Saves a copy of the body_io to +filename+
 

data/lib/mechanize/http/agent.rb

@@ -98,7 +98,7 @@ class Mechanize::HTTP::Agent
   attr_accessor :cookie_jar
 
   # Responses larger than this will be written to a Tempfile instead of stored
-  # in memory.
+  # in memory.  Setting this to nil disables creation of Tempfiles.
   attr_accessor :max_file_buffer
 
   # :section: Utility
@@ -126,7 +126,7 @@ class Mechanize::HTTP::Agent
     @gzip_enabled             = true
     @history                  = Mechanize::History.new
     @keep_alive               = true
-    @max_file_buffer          = 10240
+    @max_file_buffer          = 100_000 # 5MB for response bodies
     @open_timeout             = nil
     @post_connect_hooks       = []
     @pre_connect_hooks        = []
@@ -375,14 +375,7 @@ class Mechanize::HTTP::Agent
     log.debug('gzip response') if log
 
     zio = Zlib::GzipReader.new body_io
-    out_io = Tempfile.new 'mechanize-decode'
-    out_io.unlink
-    out_io.binmode
-
-    until zio.eof? do
-      out_io.write zio.read 16384
-    end
-
+    out_io = auto_io 'mechanize-gunzip', 16384, zio
     zio.finish
 
     return out_io
@@ -400,6 +393,7 @@ class Mechanize::HTTP::Agent
     end
   ensure
     zio.close if zio and not zio.closed?
+    body_io.close unless body_io.closed?
   end
 
   ##
@@ -421,6 +415,8 @@ class Mechanize::HTTP::Agent
       log.error("unable to inflate response: #{e}") if log
       raise
     end
+  ensure
+    body_io.close
   end
 
   def disable_keep_alive request
@@ -739,7 +735,10 @@ class Mechanize::HTTP::Agent
     raise Mechanize::Error, message
   ensure
     begin
-      body_io.close! if Tempfile === body_io and out_io.path != body_io.path
+      if Tempfile === body_io and
+         (StringIO === out_io or out_io.path != body_io.path) then
+        body_io.close! 
+      end
     rescue IOError
       # HACK ruby 1.8 raises IOError when closing the stream
     end
@@ -783,7 +782,7 @@ class Mechanize::HTTP::Agent
     sleep delay
     @history.push(page, page.uri)
     fetch new_url, :get, {}, [],
-          Mechanize::Page.new(nil, {'content-type'=>'text/html'}), redirects
+          Mechanize::Page.new, redirects
   end
 
   def response_log response
@@ -804,7 +803,7 @@ class Mechanize::HTTP::Agent
   def response_read response, request, uri
     content_length = response.content_length
 
-    if content_length and content_length > @max_file_buffer then
+    if use_tempfile? content_length then
       body_io = Tempfile.new 'mechanize-raw'
       body_io.unlink
       body_io.binmode if defined? body_io.binmode
@@ -819,7 +818,7 @@ class Mechanize::HTTP::Agent
       response.read_body { |part|
         total += part.length
 
-        if StringIO === body_io and total > @max_file_buffer then
+        if StringIO === body_io and use_tempfile? total then
           new_io = Tempfile.new 'mechanize-raw'
           new_io.unlink
           new_io.binmode
@@ -1034,17 +1033,52 @@ class Mechanize::HTTP::Agent
 
   # :section: Utility
 
+  ##
+  # Creates a new output IO by reading +input_io+ in +read_size+ chunks.  If
+  # the output is over the max_file_buffer size a Tempfile with +name+ is
+  # created.
+  #
+  # If a block is provided, each chunk of +input_io+ is yielded for further
+  # processing.
+
+  def auto_io name, read_size, input_io
+    out_io = StringIO.new
+
+    out_io.set_encoding Encoding::BINARY if out_io.respond_to? :set_encoding
+
+    until input_io.eof? do
+      if StringIO === out_io and use_tempfile? out_io.size then
+        new_io = Tempfile.new name
+        new_io.unlink
+        new_io.binmode
+
+        new_io.write out_io.string
+        out_io = new_io
+      end
+
+      chunk = input_io.read read_size
+      chunk = yield chunk if block_given?
+
+      out_io.write chunk
+    end
+
+    out_io.rewind
+
+    out_io
+  end
+
   def inflate compressed, window_bits = nil
     inflate = Zlib::Inflate.new window_bits
-    out_io = Tempfile.new 'mechanize-decode'
 
-    until compressed.eof? do
-      out_io.write inflate.inflate compressed.read 1024
+    out_io = auto_io 'mechanize-inflate', 1024, compressed do |chunk|
+      inflate.inflate chunk
     end
 
     out_io.write inflate.finish
 
     out_io
+  ensure
+    inflate.close
   end
 
   def log
@@ -1083,5 +1117,12 @@ class Mechanize::HTTP::Agent
     @http.proxy = proxy_uri
   end
 
+  def use_tempfile? size
+    return false unless @max_file_buffer
+    return false unless size
+
+    size >= @max_file_buffer
+  end
+
 end
 

data/lib/mechanize/page.rb

@@ -13,6 +13,10 @@ class Mechanize::Page < Mechanize::File
   extend Forwardable
   extend Mechanize::ElementMatcher
 
+  DEFAULT_RESPONSE = {
+    'content-type' => 'text/html',
+  }.freeze
+
   attr_accessor :mech
 
   ##
@@ -21,8 +25,9 @@ class Mechanize::Page < Mechanize::File
   attr_reader :encodings
 
   def initialize(uri=nil, response=nil, body=nil, code=nil, mech=nil)
+    response ||= DEFAULT_RESPONSE
     raise Mechanize::ContentTypeError, response['content-type'] unless
-      response['content-type'] =~ /^(text\/html)|(application\/xhtml\+xml)/i
+      response['content-type'] =~ %r{\A(?:text/html|application/xhtml\+xml)(?:$|\s*[\s;,])}i
 
     @meta_content_type = nil
     @encoding = nil

data/lib/mechanize/pluggable_parsers.rb

@@ -16,6 +16,10 @@ require 'mechanize/page'
 # Mechanize::File.  For large files you should subclass Mechanize::Download as
 # the content is only loaded into memory in small chunks.
 #
+# When writing your own pluggable parser, be sure to provide a method #body
+# that returns a String containing the response body for compatibility with
+# Mechanize#get_file.
+#
 # == Example
 #
 # To create your own parser, just create a class that takes four parameters in

data/lib/mechanize/test_case.rb

@@ -38,9 +38,7 @@ class Mechanize::TestCase < MiniTest::Unit::TestCase
 </html>
     END
 
-    response = { 'content-type' => 'text/html' }
-
-    Mechanize::Page.new uri, response, html, 200, agent
+    Mechanize::Page.new uri, nil, html, 200, agent
   end
 
   def have_encoding?
@@ -49,7 +47,7 @@ class Mechanize::TestCase < MiniTest::Unit::TestCase
 
   def html_page body
     uri = URI 'http://example/'
-    Mechanize::Page.new uri, { 'content-type' => 'text/html' }, body, 200, @mech
+    Mechanize::Page.new uri, nil, body, 200, @mech
   end
 
   def in_tmpdir

data/test/test_mechanize.rb

@@ -388,13 +388,6 @@ but not <a href="/" rel="me nofollow">this</a>!
     assert block_called
   end
 
-  def test_get_file
-    page = @mech.get("http://localhost/frame_test.html")
-    content_length = page.header['Content-Length']
-    page_as_string = @mech.get_file("http://localhost/frame_test.html")
-    assert_equal(content_length.to_i, page_as_string.length.to_i)
-  end
-
   def test_get_follow_meta_refresh
     @mech.follow_meta_refresh = true
 
@@ -696,6 +689,21 @@ but not <a href="/" rel="me nofollow">this</a>!
     assert_equal('File Upload Form', pages.title)
   end
 
+  def test_get_file
+    body = @mech.get_file 'http://localhost/frame_test.html'
+
+    assert_kind_of String, body
+    refute_empty body
+  end
+
+  def test_get_file_download
+    # non-Mechanize::File
+    body = @mech.get_file 'http://localhost/button.jpg'
+
+    assert_kind_of String, body
+    refute_empty body
+  end
+
   def test_head
     page = @mech.head('http://localhost/verb', { 'q' => 'foo' })
     assert_equal 0, @mech.history.length

data/test/test_mechanize_cookie.rb

@@ -179,14 +179,46 @@ class TestMechanizeCookie < Mechanize::TestCase
     assert !cookie.for_domain?
   end
 
-  def test_parse_expires_session
-    cookie = 'PRETANET=TGIAqbFXtt; Name=/PRETANET; Path=/; Max-Age=1.2; Content-type=text/html; Domain=192.168.6.196; expires=;'
+  def test_parse_max_age
+    url = URI.parse('http://localhost/')
+
+    date = 'Mon, 19 Feb 2012 19:26:04 GMT'
+
+    cookie = Mechanize::Cookie.parse(url, "name=Akinori; expires=#{date}").first
+    assert_equal Time.at(1329679564), cookie.expires
+
+    cookie = Mechanize::Cookie.parse(url, 'name=Akinori; max-age=3600').first
+    assert_in_delta Time.now + 3600, cookie.expires, 1
+
+    # Max-Age has precedence over Expires
+    cookie = Mechanize::Cookie.parse(url, "name=Akinori; max-age=3600; expires=#{date}").first
+    assert_in_delta Time.now + 3600, cookie.expires, 1
+
+    cookie = Mechanize::Cookie.parse(url, "name=Akinori; expires=#{date}; max-age=3600").first
+    assert_in_delta Time.now + 3600, cookie.expires, 1
+  end
 
+  def test_parse_expires_session
     url = URI.parse('http://localhost/')
 
-    cookie = Mechanize::Cookie.parse(url, cookie).first
+    [
+      'name=Akinori',
+      'name=Akinori; expires',
+      'name=Akinori; max-age',
+      'name=Akinori; expires=',
+      'name=Akinori; max-age=',
+    ].each { |str|
+      cookie = Mechanize::Cookie.parse(url, str).first
+      assert cookie.session, str
+    }
 
-    assert cookie.session
+    [
+      'name=Akinori; expires=Mon, 19 Feb 2012 19:26:04 GMT',
+      'name=Akinori; max-age=3600',
+    ].each { |str|
+      cookie = Mechanize::Cookie.parse(url, str).first
+      assert !cookie.session, str
+    }
   end
 
   def test_parse_many
@@ -198,10 +230,15 @@ class TestMechanizeCookie < Mechanize::TestCase
       "name=Aaron; Domain=localhost; Expires=Sun, 06 Nov 2011 00:29:51 GMT; Path=/; HttpOnly, " \
       "expired=doh; Expires=Fri, 04 Nov 2011 00:29:51 GMT; Path=/, " \
       "a_path=some_path; Expires=Sun, 06 Nov 2011 00:29:51 GMT; Path=/some_path, " \
-      "no_path=no_path; Expires=Sun, 06 Nov 2011 00:29:51 GMT, no_expires=nope; Path=/"
+      "no_path1=no_path; Expires=Sun, 06 Nov 2011 00:29:52 GMT, no_expires=nope; Path=/, " \
+      "no_path2=no_path; Expires=Sun, 06 Nov 2011 00:29:52 GMT; no_expires=nope; Path, " \
+      "no_path3=no_path; Expires=Sun, 06 Nov 2011 00:29:52 GMT; no_expires=nope; Path=, " \
+      "no_domain1=no_domain; Expires=Sun, 06 Nov 2011 00:29:53 GMT; no_expires=nope, " \
+      "no_domain2=no_domain; Expires=Sun, 06 Nov 2011 00:29:53 GMT; no_expires=nope; Domain, " \
+      "no_domain3=no_domain; Expires=Sun, 06 Nov 2011 00:29:53 GMT; no_expires=nope; Domain="
 
     cookies = Mechanize::Cookie.parse url, cookie_str
-    assert_equal 8, cookies.length
+    assert_equal 13, cookies.length
 
     name = cookies.find { |c| c.name == 'name' }
     assert_equal "Aaron",             name.value
@@ -218,10 +255,20 @@ class TestMechanizeCookie < Mechanize::TestCase
     assert_equal "/",    no_expires.path
     assert_nil           no_expires.expires
 
-    no_path = cookies.find { |c| c.name == 'no_path' }
-    assert_equal "no_path",           no_path.value
-    assert_equal "/",                 no_path.path
-    assert_equal Time.at(1320539391), no_path.expires
+    no_path_cookies = cookies.select { |c| c.value == 'no_path' }
+    assert_equal 3, no_path_cookies.size
+    no_path_cookies.each { |c|
+      assert_equal "/",                 c.path,    c.name
+      assert_equal Time.at(1320539392), c.expires, c.name
+    }
+
+    no_domain_cookies = cookies.select { |c| c.value == 'no_domain' }
+    assert_equal 3, no_domain_cookies.size
+    no_domain_cookies.each { |c|
+      assert !c.for_domain?, c.name
+      assert_equal c.domain, url.host, c.name
+      assert_equal Time.at(1320539393), c.expires, c.name
+    }
 
     assert cookies.find { |c| c.name == 'expired' }
   end

data/test/test_mechanize_download.rb

@@ -8,6 +8,16 @@ class TestMechanizeDownload < Mechanize::TestCase
     @parser = Mechanize::Download
   end
 
+  def test_body
+    uri = URI.parse 'http://example/foo.html'
+    body_io = StringIO.new '0123456789'
+
+    download = @parser.new uri, nil, body_io
+
+    assert_equal '0123456789', download.body
+    assert_equal 0, download.body_io.pos
+  end
+
   def test_save_string_io
     uri = URI.parse 'http://example/foo.html'
     body_io = StringIO.new '0123456789'

data/test/test_mechanize_http_agent.rb

@@ -31,6 +31,69 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
     realm
   end
 
+  def test_auto_io
+    Tempfile.open 'input' do |input_io|
+      input_io.binmode
+      input_io.write '12345'
+      input_io.rewind
+
+      out_io = @agent.auto_io __name__, 1024, input_io
+
+      assert_equal '12345', out_io.string
+
+      assert_equal Encoding::BINARY, out_io.string.encoding if
+        Object.const_defined? :Encoding
+    end
+  end
+
+  def test_auto_io_chunk
+    Tempfile.open 'input' do |input_io|
+      chunks = []
+
+      input_io.binmode
+      input_io.write '12345'
+      input_io.rewind
+
+      @agent.auto_io __name__, 1, input_io do |chunk|
+        chunks << chunk
+      end
+
+      assert_equal %w[1 2 3 4 5], chunks
+    end
+  end
+
+  def test_auto_io_tempfile
+    @agent.max_file_buffer = 3
+
+    Tempfile.open 'input' do |input_io|
+      input_io.binmode
+      input_io.write '12345'
+      input_io.rewind
+
+      out_io = @agent.auto_io __name__, 1, input_io
+
+      result = out_io.read
+      assert_equal '12345', result
+
+      assert_equal Encoding::BINARY, result.encoding if
+        Object.const_defined? :Encoding
+    end
+  end
+
+  def test_auto_io_yield
+    Tempfile.open 'input' do |input_io|
+      input_io.binmode
+      input_io.write '12345'
+      input_io.rewind
+
+      out_io = @agent.auto_io __name__, 1024, input_io do |chunk|
+        "x#{chunk}"
+      end
+
+      assert_equal 'x12345', out_io.string
+    end
+  end
+
   def test_certificate_equals
     cert_path = File.expand_path '../data/server.crt', __FILE__
     cert = OpenSSL::X509::Certificate.new File.read cert_path
@@ -176,8 +239,7 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
     @agent.follow_meta_refresh = true
     @agent.follow_meta_refresh_self = true
 
-    page = Mechanize::Page.new(@uri, {'content-type' => 'text/html'}, '',
-                               200, @mech)
+    page = Mechanize::Page.new(@uri, nil, '', 200, @mech)
     @res.instance_variable_set :@header, 'refresh' => ['0']
 
     refresh = @agent.get_meta_refresh @res, @uri, page
@@ -186,8 +248,7 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
   end
 
   def test_get_meta_refresh_header_no_follow
-    page = Mechanize::Page.new(@uri, {'content-type' => 'text/html'}, '',
-                               200, @mech)
+    page = Mechanize::Page.new(@uri, nil, '', 200, @mech)
     @res.instance_variable_set :@header, 'refresh' => ['0']
 
     refresh = @agent.get_meta_refresh @res, @uri, page
@@ -198,8 +259,7 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
   def test_get_meta_refresh_header_no_follow_self
     @agent.follow_meta_refresh = true
 
-    page = Mechanize::Page.new(@uri, {'content-type' => 'text/html'}, '',
-                               200, @mech)
+    page = Mechanize::Page.new(@uri, nil, '', 200, @mech)
     @res.instance_variable_set :@header, 'refresh' => ['0']
 
     refresh = @agent.get_meta_refresh @res, @uri, page
@@ -216,8 +276,7 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
 <meta http-equiv="refresh" content="0">
     BODY
 
-    page = Mechanize::Page.new(@uri, {'content-type' => 'text/html'}, body,
-                               200, @mech)
+    page = Mechanize::Page.new(@uri, nil, body, 200, @mech)
 
     refresh = @agent.get_meta_refresh @res, @uri, page
 
@@ -230,8 +289,7 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
 <meta http-equiv="refresh" content="0">
     BODY
 
-    page = Mechanize::Page.new(@uri, {'content-type' => 'text/html'}, body,
-                               200, @mech)
+    page = Mechanize::Page.new(@uri, nil, body, 200, @mech)
 
     refresh = @agent.get_meta_refresh @res, @uri, page
 
@@ -246,8 +304,7 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
 <meta http-equiv="refresh" content="0">
     BODY
 
-    page = Mechanize::Page.new(@uri, {'content-type' => 'text/html'}, body,
-                               200, @mech)
+    page = Mechanize::Page.new(@uri, nil, body, 200, @mech)
 
     refresh = @agent.get_meta_refresh @res, @uri, page
 
@@ -301,6 +358,14 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
     assert_equal 1, @agent.http.idle_timeout
   end
 
+  def test_inflate
+    body_io = StringIO.new "x\x9C+H,*\x01\x00\x04?\x01\xB8"
+
+    result = @agent.inflate body_io
+
+    assert_equal 'part', result.read
+  end
+
   def test_post_connect
     @agent.post_connect_hooks << proc { |agent, uri, response, body|
       assert_equal @agent, agent
@@ -716,6 +781,8 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
     body = @agent.response_content_encoding @res, body_io
 
     assert_equal 'part', body.read
+
+    assert body_io.closed?
   end
 
   def test_response_content_encoding_deflate_chunked
@@ -737,6 +804,8 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
 
     assert_match %r%error handling content-encoding deflate:%, e.message
     assert_match %r%Zlib%, e.message
+
+    assert body_io.closed?
   end
 
   def test_response_content_encoding_deflate_empty
@@ -764,6 +833,8 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
     body = @agent.response_content_encoding @res, body_io
 
     assert_equal 'part', body.read
+
+    assert body_io.closed?
   end
 
   def test_response_content_encoding_gzip_chunked
@@ -795,6 +866,8 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
 
     assert_match %r%unable to gunzip response, trying raw inflate%, log.string
     assert_match %r%unable to gunzip response:%, log.string
+
+    assert body_io.closed?
   end
 
   def test_response_content_encoding_gzip_corrupt_checksum
@@ -958,8 +1031,7 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
 <meta http-equiv="refresh" content="0">
     BODY
 
-    page = Mechanize::Page.new(uri, {'content-type' => 'text/html'}, body,
-                               200, @mech)
+    page = Mechanize::Page.new(uri, nil, body, 200, @mech)
 
     @agent.follow_meta_refresh = true
     @agent.follow_meta_refresh_self = true
@@ -977,8 +1049,7 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
 <meta http-equiv="refresh" content="0">
     BODY
 
-    page = Mechanize::Page.new(uri, {'content-type' => 'text/html'}, body,
-                               200, @mech)
+    page = Mechanize::Page.new(uri, nil, body, 200, @mech)
 
     @agent.follow_meta_refresh = true
     @agent.follow_meta_refresh_self = true
@@ -1159,6 +1230,8 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
   end
 
   def test_response_read_large
+    @agent.max_file_buffer = 10240
+
     def @res.read_body() yield 'a' * 10241 end
     def @res.content_length() 10241 end
 
@@ -1169,6 +1242,8 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
   end
 
   def test_response_read_large_chunked
+    @agent.max_file_buffer = 10240
+
     def @res.read_body
       11.times do yield 'a' * 1024 end
     end
@@ -1363,6 +1438,19 @@ class TestMechanizeHttpAgent < Mechanize::TestCase
     end
   end
 
+  def test_use_tempfile_eh
+    refute @agent.use_tempfile? nil
+
+    @agent.max_file_buffer = 1
+
+    refute @agent.use_tempfile? 0
+    assert @agent.use_tempfile? 1
+
+    @agent.max_file_buffer = nil
+
+    refute @agent.use_tempfile? 1
+  end
+
   def test_verify_none_equals
     @agent.verify_mode = OpenSSL::SSL::VERIFY_NONE
 

data/test/test_mechanize_page.rb

@@ -8,13 +8,43 @@ class TestMechanizePage < Mechanize::TestCase
     @uri = URI 'http://example/'
   end
 
-  def test_initialize_bad_content_type
-    e = assert_raises Mechanize::ContentTypeError do
-      Mechanize::Page.new(URI('http://example/'),
-                          { 'content-type' => 'text/xml' }, 'hello', '200')
-    end
+  def test_initialize_good_content_type
+    page = Mechanize::Page.new
+    assert_equal('text/html', page.content_type)
+
+    [
+      'text/html',
+      'Text/HTML',
+      'text/html; charset=UTF-8',
+      'text/html ; charset=US-ASCII',
+      'application/xhtml+xml',
+      'Application/XHTML+XML',
+      'application/xhtml+xml; charset=UTF-8',
+      'application/xhtml+xml ; charset=US-ASCII',
+    ].each { |content_type|
+      page = Mechanize::Page.new(URI('http://example/'),
+        { 'content-type' => content_type }, 'hello', '200')
+
+      assert_equal(content_type, page.content_type, content_type)
+    }
+  end
 
-    assert_equal('text/xml', e.content_type)
+  def test_initialize_bad_content_type
+    [
+      'text/xml',
+      'text/xhtml',
+      'text/htmlfu',
+      'footext/html',
+      'application/xhtml+xmlfu',
+      'fooapplication/xhtml+xml',
+    ].each { |content_type|
+      e = assert_raises(Mechanize::ContentTypeError, content_type) do
+        Mechanize::Page.new(URI('http://example/'),
+          { 'content-type' => content_type }, 'hello', '200')
+      end
+
+      assert_equal(content_type, e.content_type, content_type)
+    }
   end
 
   def test_frames

data/test/test_mechanize_page_image.rb

@@ -8,8 +8,7 @@ class TestMechanizePageImage < Mechanize::TestCase
     @uri = URI 'http://example/'
     @src = (@uri + 'a.jpg').to_s
 
-    @empty_page = Mechanize::Page.new(@uri, { 'content-type' => 'text/html' },
-                                      '', 200, @mech)
+    @empty_page = Mechanize::Page.new(@uri, nil, '', 200, @mech)
   end
 
   def img attributes

data/test/test_mechanize_page_meta_refresh.rb

@@ -15,8 +15,7 @@ class TestMechanizePageMetaRefresh < Mechanize::TestCase
 <head><meta http-equiv="refresh" content="#{delay};url=#{uri}"></head>
     BODY
 
-    Mechanize::Page.new(@uri, { 'content-type' => 'text/html' }, body, 200,
-                        @mech)
+    Mechanize::Page.new(@uri, nil, body, 200, @mech)
   end
 
   def util_meta_refresh page
@@ -126,8 +125,7 @@ class TestMechanizePageMetaRefresh < Mechanize::TestCase
 <head><meta http-equiv="refresh"></head>
     BODY
 
-    page = Mechanize::Page.new(@uri, { 'content-type' => 'text/html' }, body,
-                               200, @mech)
+    page = Mechanize::Page.new(@uri, nil, body, 200, @mech)
 
     assert_nil util_meta_refresh page
   end
@@ -137,8 +135,7 @@ class TestMechanizePageMetaRefresh < Mechanize::TestCase
 <head><meta http-equiv="other-thing" content="0;"></head>
     BODY
 
-    page = Mechanize::Page.new(@uri, { 'content-type' => 'text/html' }, body,
-                               200, @mech)
+    page = Mechanize::Page.new(@uri, nil, body, 200, @mech)
 
     assert_nil util_meta_refresh page
   end

metadata

@@ -5,9 +5,8 @@ version: !ruby/object:Gem::Version
   prerelease: 
   segments: 
   - 2
-  - 2
-  - 1
-  version: 2.2.1
+  - 3
+  version: "2.3"
 platform: ruby
 authors: 
 - Eric Hodel
@@ -39,7 +38,7 @@ cert_chain:
   YJY7T/W2n+eWy8WuPhzVUkyzguj0bQe27NDeabgCh2mHd4Hynk2AkYh8MQ==
   -----END CERTIFICATE-----
 
-date: 2012-02-14 00:00:00 Z
+date: 2012-02-21 00:00:00 Z
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: net-http-digest_auth