pincers 0.7.0 → 0.7.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: a19d33e62f9611053e680e501b3bab925c4279fa
-  data.tar.gz: 2266ae44ad8606cc8dd7097c91d2d9ae64a8d80f
+  metadata.gz: 687162c2ca445449ffdacc2890cd9e94182317da
+  data.tar.gz: bf7434146ce2516a3d5b5cae8ac9f9cebd03a4ae
 SHA512:
-  metadata.gz: 7764b044ebdfd024d894273fb511573644be4074415acc24fb46b5fc9f577d7dd5d95cc8cdbb90fbd6ac90de203c4852c4b590f837c64a58cd882c39a8a2f5d6
-  data.tar.gz: 3b35f1c8568797d90d2cb90a87b81997158b76428a3b62a550fa723b2770af8ee05d370d7de1ac4732f4b9feaca2eb4fd73502555d404a2c560f751ed39a82ad
+  metadata.gz: c8d1a5595a9b19c1ed239ed081e101e4df882522752d79c096802978d04fb5b6395ef63d628078d8ef81828f9216e794ba338ae3a73716da21987ba174b561a3
+  data.tar.gz: 2fa86573efa8cb6ccb0e1d26eb2c4e892d3a8e6fae2009a70f3308ec3d3decc29a5c081312067d542de22ca2898f33cbbb6e527c048088f90bca1771a8762e0b

data/lib/pincers/chenso/backend.rb

@@ -2,6 +2,7 @@ require 'pincers/nokogiri/backend'
 require 'pincers/chenso/browsing_manager'
 require 'pincers/chenso/html_page_request'
 require 'pincers/chenso/html_form_request'
+require 'pincers/chenso/html_page_cache'
 require 'pincers/core/helpers/form'
 
 module Pincers::Chenso
@@ -68,7 +69,14 @@ module Pincers::Chenso
     end
 
     def as_http_client
-      @client.fork(false)
+      @client.fork(true)
+    end
+
+    def merge_http_client(_client)
+      @client.join _client
+      if _client.content and /text\/html/ === _client.content_type
+        @browser.push HtmlPageCache.new(_client.uri, _client.content)
+      end
     end
 
     def switch_to_frame(_element)

data/lib/pincers/chenso/browsing_context.rb

@@ -2,14 +2,12 @@ module Pincers::Chenso
 
   class BrowsingContext
 
-    attr_reader :document
-
-    def initialize(_http_client)
+    def initialize(_http_client, _state=nil)
       @client = _http_client
       @history = []
       @pointer = -1
       @childs = {}
-      @document = nil
+      @state = _state
     end
 
     def get_child(_id)
@@ -17,24 +15,32 @@ module Pincers::Chenso
     end
 
     def load_child(_id)
-      @childs[_id] = self.class.new @client
+      @childs[_id] = self.class.new(@client, @state)
     end
 
     def current_url
-      @client.uri.to_s
+      @state ? @state.uri.to_s : nil
+    end
+
+    def document
+      @state ? @state.document : nil
     end
 
     def refresh
       if @pointer >= 0
         navigate @history[@pointer]
-      else nil end
+      end
+      self
     end
 
     def push(_request)
+      _request.fix_uri @state
+
       @history.slice!(@pointer+1..-1)
       @history.push _request
       @pointer += 1
       navigate _request
+      self
     end
 
     def back(_times=1)
@@ -46,6 +52,7 @@ module Pincers::Chenso
       else
         change_pointer @pointer - _times
       end
+      self
     end
 
     def forward(_times=1)
@@ -55,6 +62,7 @@ module Pincers::Chenso
       else
         change_pointer @pointer + _times
       end
+      self
     end
 
   private
@@ -67,9 +75,9 @@ module Pincers::Chenso
     end
 
     def navigate(_request)
-      @document = _request.execute @client
+      @state = _request.execute @client
       @childs.clear
-      @document
+      nil
     end
 
   end

data/lib/pincers/chenso/browsing_state.rb

@@ -0,0 +1,4 @@
+module Pincers::Chenso
+  class BrowsingState < Struct.new(:uri, :document)
+  end
+end

data/lib/pincers/chenso/html_page_cache.rb

@@ -0,0 +1,18 @@
+require 'pincers/chenso/browsing_state'
+
+module Pincers::Chenso
+  class HtmlPageCache
+    def initialize(_uri, _content)
+      @uri =  _uri
+      @content = _content
+    end
+
+    def fix_uri(_current_state)
+      # nothing
+    end
+
+    def execute(_client)
+      BrowsingState.new @uri, ::Nokogiri::HTML(@content)
+    end
+  end
+end

data/lib/pincers/chenso/html_page_request.rb

@@ -1,3 +1,6 @@
+require 'pincers/http/utils'
+require 'pincers/chenso/browsing_state'
+
 module Pincers::Chenso
   class HtmlPageRequest
 
@@ -7,21 +10,28 @@ module Pincers::Chenso
     }
 
     def initialize(_url, _method=:get, _data=nil, _encoding=nil)
-      @url = _url
+      @url =  _url
       @method = _method
       @data = _data
       @encoding = _encoding
     end
 
-    def execute(_client)
-      @uri = _client.absolute_uri_for @url if @uri.nil?
+    def fix_uri(_current_state)
+      @uri = Pincers::Http::Utils.parse_uri @url
+      if _current_state
+        @uri = URI.join(_current_state.uri, @uri)
+      elsif @uri.relative?
+        raise ArgumentError, 'Absolute uri required'
+      end
+    end
 
+    def execute(_client)
       _client.send(@method, @uri) do |request|
         request.headers.merge DEFAULT_HEADERS
         request.set_form_data(@data, @encoding) unless @data.nil?
       end
 
-      ::Nokogiri::HTML _client.content
+      BrowsingState.new _client.uri, ::Nokogiri::HTML(_client.content)
     end
   end
 end

data/lib/pincers/core/base_backend.rb

@@ -130,6 +130,10 @@ module Pincers::Core
       ensure_implementation :as_http_client
     end
 
+    def merge_http_client(_client)
+      # nothing
+    end
+
     def fetch_resource
       ensure_implementation :fetch_resource
     end

data/lib/pincers/core/root_context.rb

@@ -109,7 +109,7 @@ module Pincers::Core
       http_client = backend.as_http_client
       unless _block.nil?
         r = _block.call http_client
-        # sync_with http_client # TODO :copy cookies and maybe url?
+        backend.merge_http_client http_client
         r
       else
         http_client

data/lib/pincers/http/client.rb

@@ -42,6 +42,10 @@ module Pincers::Http
     end
 
     def set_cookie(_cookie)
+      if _cookie.is_a? Hash
+        _cookie = Cookie.new(_cookie[:name], _cookie[:value], _cookie[:domain], _cookie[:path])
+      end
+
       @session.cookie_jar.set _cookie
     end
 
@@ -72,11 +76,15 @@ module Pincers::Http
       perform_in_session request
     end
 
-    def fork(_keep_session = true)
+    def fork(_keep_session = false)
       fork_session = _keep_session ? @session : @session.clone
       self.class.new fork_session, @document
     end
 
+    def join(_other_client)
+      @session.sync _other_client.session
+    end
+
     def absolute_uri_for(_url)
       uri = _url.is_a?(URI) ? _url : Utils.parse_uri(_url)
       if uri.relative?

data/lib/pincers/http/cookie.rb

@@ -1,3 +1,24 @@
 module Pincers::Http
-  class Cookie < Struct.new(:name, :value, :domain, :path, :expires, :secure); end
+  class Cookie
+
+    attr_reader :name, :value, :domain, :path, :expires, :secure
+
+    def initialize(_name, _value, _domain, _path=nil, _expires=nil, _secure=nil)
+      @name = _name
+      @value = _value
+      @domain = _domain.gsub(/^\./,'').downcase # RFC 6265 5.2.3
+      @path = valid_path?(_path) ? _path.downcase : '/'
+      @expires = _expires
+      @secure = _secure
+    end
+
+  private
+
+    def valid_path?(_path)
+      !_path.nil? && !_path.empty? && _path[0] == '/'
+    end
+
+  end
+
+
 end

data/lib/pincers/http/cookie_jar.rb

@@ -20,38 +20,38 @@ module Pincers::Http
       for_origin(Utils.parse_uri(_url)).find { |c| c.name == _name }
     end
 
-    def set(_parts)
-      return nil unless _parts[:domain]
-      return nil unless _parts[:name]
-      return nil unless _parts[:value]
-
-      cookie = Cookie.new(
-        _parts[:name],
-        _parts[:value],
-        _parts[:domain].gsub(/^\./,''),
-        _parts[:path] || '/',
-        _parts[:expires],
-        _parts[:secure]
-      )
-
-      replace_cookie cookie
-      cookie
-    end
-
-    def set_raw(_uri, _raw)
-      cookie = decode_cookie _raw
+    def set(_cookie)
+      if _cookie.name.nil? or _cookie.value.nil? or _cookie.domain.nil? or _cookie.path.nil?
+        raise ArgumentError, 'Invalid cookie'
+      end
 
-      cookie.domain = compute_domain cookie.domain, _uri.host
-      cookie.path = compute_path cookie.path, _uri.path
+      @cookies.each_with_index do |cookie, i|
+        if equivalent(cookie, _cookie)
+          @cookies[i] = _cookie
+          return _cookie
+        end
+      end
 
-      return nil if cookie.domain.nil? or cookie.path.nil?
+      @cookies << _cookie
+      _cookie
+    end
 
-      replace_cookie(cookie)
+    def set_raw(_request_uri, _raw)
+      cookie = decode_cookie _request_uri, _raw
+      set cookie unless cookie.nil?
       cookie
     end
 
+    def set_from_header(_uri, _header)
+      _header.split(/, (?=\w+=)/).map do |raw_cookie|
+        set_raw _uri, raw_cookie.strip
+      end
+    end
+
     def for_origin(_uri)
+      # RFC 6265 5.4.1
       @cookies.select do |c|
+        # TODO: add scheme and host only checks
         domains_match c.domain, _uri.host and paths_match c.path, _uri.path
       end
     end
@@ -62,38 +62,56 @@ module Pincers::Http
 
   private
 
-    def decode_cookie(_raw)
+    def decode_cookie(_request, _raw)
       # taken from WEBrick implementation
       cookie_elem = _raw.split(/;/)
       first_elem = cookie_elem.shift
       first_elem.strip!
-      key, value = first_elem.split(/\=/, 2)
 
-      cookie = Cookie.new(key, dequote(value))
+      name, value = first_elem.split(/\=/, 2)
+      domain = nil
+      path = nil
+      expires = nil
+      secure = nil
+      # TODO: host_only = true
+
       cookie_elem.each do |pair|
         pair.strip!
-        key, value = pair.split(/\=/, 2)
-        value = dequote(value.strip) if value
-
-        case key.downcase
-        when "domain"  then cookie.domain  = value.downcase
-        when "path"    then cookie.path    = value.downcase
-        when "expires" then cookie.expires = value
-        # when "max-age" then cookie.max_age = Integer(value)
-        # when "comment" then cookie.comment = value
-        # when "version" then cookie.version = Integer(value)
-        when "secure"  then cookie.secure = true
+        opt_key, opt_value = pair.split(/\=/, 2)
+        opt_value = dequote(opt_value.strip) if opt_value
+
+        case opt_key.downcase
+        when "domain"
+          domain = opt_value.downcase
+          # TODO: host_only = false
+          return nil unless domains_match(domain, _request.host) # RFC 6265 5.3.6
+        when "path"
+          path = opt_value.downcase if opt_value[0] == '/' # RFC 6265 5.2.4
+        when "expires" then expires = opt_value
+        # when "max-age" then max_age = Integer(value)
+        # when "comment" then comment = value
+        # when "version" then version = Integer(value)
+        when "secure"  then secure = true
         end
       end
 
-      cookie
+      Cookie.new(
+        name,
+        dequote(value),
+        domain || _request.host,
+        path || default_path(_request.path),
+        expires,
+        secure
+      )
     end
 
     def domains_match(_cookie_domain, _request_domain)
       # RFC 6265 - 5.1.3
       # TODO: ensure request domain is not an IP
       return true if _cookie_domain == _request_domain
-      return true if _request_domain.end_with? ".#{_cookie_domain}"
+      if _request_domain.end_with? "#{_cookie_domain}"
+        return true if _cookie_domain[0] == '.' or _request_domain.end_with? ".#{_cookie_domain}"
+      end
       return false
     end
 
@@ -101,26 +119,13 @@ module Pincers::Http
       # RFC 6265 - 5.1.4
       _request_path = '/' if _request_path.empty?
       return true if _cookie_path == _request_path
-      return true if _cookie_path[-1] == '/' and _request_path.start_with? _cookie_path
-      return true if _request_path.start_with? "#{_cookie_path}/"
+      if _request_path.start_with? _cookie_path
+        return true if _cookie_path[-1] == '/' or _request_path.start_with? "#{_cookie_path}/"
+      end
       return false
     end
 
-    def compute_domain(_cookie_domain, _request_domain)
-      return _request_domain if _cookie_domain.nil?
-      # cookies with different domain are discarded
-      return nil unless _cookie_domain.end_with? _request_domain
-      return _cookie_domain.gsub(/^\./,'') # remove leading dot
-    end
-
-    def compute_path(_cookie_path, _request_path)
-      default_path = compute_default_path(_request_path)
-      return default_path if _cookie_path.nil?
-      return nil unless _cookie_path.start_with? default_path
-      return _cookie_path
-    end
-
-    def compute_default_path(_request_path)
+    def default_path(_request_path)
       # RFC 6265 - 5.1.4
       return '/' unless _request_path[0] === '/'
       ls_idx = _request_path.rindex('/')
@@ -128,17 +133,6 @@ module Pincers::Http
       _request_path[0..ls_idx]
     end
 
-    def replace_cookie(_cookie)
-      @cookies.each_with_index do |cookie, i|
-        if equivalent(cookie, _cookie)
-          @cookies[i] = _cookie
-          return
-        end
-      end
-
-      @cookies << _cookie
-    end
-
     def dequote(_str)
       # taken from WEBrick implementation
       ret = (/\A"(.*)"\Z/ =~ _str) ? $1 : _str.dup

data/lib/pincers/http/session.rb

@@ -38,6 +38,11 @@ module Pincers::Http
       self.class.new self
     end
 
+    def sync(_other)
+      @headers.merge! _other.headers
+      _other.cookie_jar.cookies.each { |c| cookie_jar.set c }
+    end
+
     def perform(_request)
       perform_recursive _request, @redirect_limit, nil
     end
@@ -57,16 +62,16 @@ module Pincers::Http
       copy_headers http_request, _request.headers
       set_cookies http_request, uri
 
-      http_response = connect(uri).request http_request
+      http_response = connect(uri).request(http_request)
+      update_cookies(uri, http_response)
 
       case http_response
       when Net::HTTPSuccess then
-        update_cookies(uri, http_response)
         http_response.uri = uri # uri is not always set by net/http
         http_response
       when Net::HTTPRedirection then
         location = Utils.parse_uri(http_response['location'])
-        perform(_request, _limit - 1, location)
+        perform_recursive(_request, _limit - 1, location)
       else
         handle_error_response http_response
       end
@@ -92,8 +97,8 @@ module Pincers::Http
     end
 
     def update_cookies(_uri, _response)
-      cookies = _response.get_fields('set-cookie')
-      cookies.each { |raw| @cookie_jar.set_raw _uri, raw } if cookies
+      fields = _response.get_fields('set-cookie')
+      fields.each { |field| cookie_jar.set_from_header _uri, field } if fields
     end
   end
 end

data/lib/pincers/version.rb

@@ -1,3 +1,3 @@
 module Pincers
-  VERSION = "0.7.0"
+  VERSION = "0.7.1"
 end

data/lib/pincers/webdriver/backend.rb

@@ -1,7 +1,8 @@
 require 'selenium-webdriver'
 require 'pincers/core/base_backend'
-require 'pincers/http/client'
 require 'pincers/webdriver/http_document'
+require 'pincers/http/client'
+require 'pincers/http/cookie'
 
 module Pincers::Webdriver
   class Backend < Pincers::Core::BaseBackend
@@ -161,7 +162,7 @@ module Pincers::Webdriver
       session = Pincers::Http::Session.new
       session.headers['User-Agent'] = user_agent
       session.proxy = proxy_address
-      driver.manage.all_cookies.each { |c| session.cookie_jar.set c }
+      load_cookies_in_session session
 
       Pincers::Http::Client.new session, HttpDocument.new(self)
     end
@@ -217,5 +218,20 @@ module Pincers::Webdriver
       proxy = driver.capabilities.proxy
       proxy.nil? ? nil : (proxy.http || proxy.ssl)
     end
+
+    def load_cookies_in_session(_session)
+      driver.manage.all_cookies.each do |wd_cookie|
+        if wd_cookie[:domain] and wd_cookie[:name] and wd_cookie[:value]
+          _session.cookie_jar.set Pincers::Http::Cookie.new(
+            wd_cookie[:name],
+            wd_cookie[:value],
+            wd_cookie[:domain],
+            wd_cookie[:path],
+            wd_cookie[:expires],
+            wd_cookie[:secure]
+          )
+        end
+      end
+    end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pincers
 version: !ruby/object:Gem::Version
-  version: 0.7.0
+  version: 0.7.1
 platform: ruby
 authors:
 - Ignacio Baixas
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-10-21 00:00:00.000000000 Z
+date: 2015-10-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
@@ -231,8 +231,10 @@ files:
 - lib/pincers/chenso/backend.rb
 - lib/pincers/chenso/browsing_context.rb
 - lib/pincers/chenso/browsing_manager.rb
+- lib/pincers/chenso/browsing_state.rb
 - lib/pincers/chenso/factory.rb
 - lib/pincers/chenso/html_form_request.rb
+- lib/pincers/chenso/html_page_cache.rb
 - lib/pincers/chenso/html_page_request.rb
 - lib/pincers/core/base_backend.rb
 - lib/pincers/core/base_factory.rb