puppeteer-ruby 0.37.4 → 0.40.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '000206932f5d3d4626602e984cff4015039644eea2ef89ff139b239fffb9f673'
-  data.tar.gz: 0e560f862a3317a42c1e0e2907c599bc654d05e3f17fd5a650126b73bfd6948e
+  metadata.gz: 189bf68e4da7ec9b5b816ef859469c084697b29237aeaa2e26cdc829620fda22
+  data.tar.gz: 661e878eb399c8513e3a92210949eca200c3853623169b548d0f288b525aa951
 SHA512:
-  metadata.gz: aac4be4a26f3116eee198ea65c0017359589f2ad11cc95125497e3f84ea1277d9bbb8d1071b57fb42812bb12afd4370534c48586fbb79897acafb54d03222159
-  data.tar.gz: de51d1ca30e26e13346532118df225370fad4b1967d205dad0c150574c4fa60a649973aab83e24b2140b8f98f05626eb106a9141f04b94fe5d3927b1231737d4
+  metadata.gz: 4848815b7576739d73ce9f945099508cc2e5122a7fe230f28ae5cb2d78cd77867c97650c7004ab0ce7172938a05de52246e43ce2ecde21c825c3456a971a49d1
+  data.tar.gz: 3acfcec865a1a202dacfb831ca7dfaba5e2228769074e492487208edb7a2e944e6ad9d5202763359b8c36dbddc05e671f1f9fa53176f1e9e832d63c0ab907b8f

data/.rubocop.yml

@@ -149,6 +149,8 @@ Style/MethodCallWithArgsParentheses:
     # utils
     - debug_print
     - debug_puts
+    - exit
+    - puts
 
 Style/MethodCallWithoutArgsParentheses:
   Enabled: true

data/CHANGELOG.md

@@ -1,7 +1,41 @@
-### main [[diff](https://github.com/YusukeIwaki/puppeteer-ruby/compare/0.37.4...main)]
+### main [[diff](https://github.com/YusukeIwaki/puppeteer-ruby/compare/0.40.1...main)]
 
 - xxx
 
+### 0.40.1 [[diff](https://github.com/YusukeIwaki/puppeteer-ruby/compare/0.40.0...0.40.1)]
+
+Bugfix:
+
+- Fix error on reload [#190](https://github.com/YusukeIwaki/puppeteer-ruby/pull/190)
+- Fix error message on failing to find Chrome executable.
+- Fix apidoc
+
+### 0.40.0 [[diff](https://github.com/YusukeIwaki/puppeteer-ruby/compare/0.39.0...0.40.0)]
+
+New features:
+
+- Puppeteer 13.0, 13.1 functionalities
+- Support Ruby 3.1
+
+
+### 0.39.0 [[diff](https://github.com/YusukeIwaki/puppeteer-ruby/compare/0.38.0...0.39.0)]
+
+New features:
+
+- Puppeteer 12.0 functionalities
+
+### 0.38.0 [[diff](https://github.com/YusukeIwaki/puppeteer-ruby/compare/0.37.4...0.38.0)]
+
+New features:
+
+- Puppeteer 11.0 functionalities
+    * OOP iframe support
+    * Customize debugging port
+    * HTTPRequest#initiator
+    * Customize temp directory for user data
+    * Add webp to screenshot quality option allow list
+
+
 ### 0.37.4 [[diff](https://github.com/YusukeIwaki/puppeteer-ruby/compare/0.37.3...0.37.4)]
 
 Bugfix:

data/docs/api_coverage.md

@@ -1,6 +1,6 @@
 # API coverages
-- Puppeteer version: v12.0.0
-- puppeteer-ruby version: 0.37.4
+- Puppeteer version: v13.0.1
+- puppeteer-ruby version: 0.40.1
 
 ## Puppeteer
 
@@ -134,7 +134,7 @@
 * viewport
 * ~~waitFor~~
 * waitForFileChooser => `#wait_for_file_chooser`
-* ~~waitForFrame~~
+* waitForFrame => `#wait_for_frame`
 * waitForFunction => `#wait_for_function`
 * waitForNavigation => `#wait_for_navigation`
 * ~~waitForNetworkIdle~~
@@ -227,7 +227,7 @@
 * goto
 * hover
 * isDetached => `#detached?`
-* ~~isOOPFrame~~
+* isOOPFrame => `#oop_frame?`
 * name
 * parentFrame => `#parent_frame`
 * select
@@ -296,6 +296,7 @@
 * ~~toString~~
 * type => `#type_text`
 * uploadFile => `#upload_file`
+* waitForSelector => `#wait_for_selector`
 
 ## HTTPRequest
 
@@ -308,7 +309,9 @@
 * finalizeInterceptions => `#finalize_interceptions`
 * frame
 * headers
-* ~~initiator~~
+* initiator
+* ~~interceptResolutionState~~
+* ~~isInterceptResolutionHandled~~
 * isNavigationRequest => `#navigation_request?`
 * method
 * postData => `#post_data`
@@ -360,7 +363,7 @@
 
 * connection
 * detach
-* ~~id~~
+* id
 * send
 
 ## Coverage

data/lib/puppeteer/aria_query_handler.rb

@@ -7,10 +7,10 @@ class Puppeteer::AriaQueryHandler
   private def parse_aria_selector(selector)
     known_attributes = %w(name role)
     query_options = {}
-    attribute_regexp = /\[\s*(?<attribute>\w+)\s*=\s*"(?<value>\\.|[^"\\]*)"\s*\]/
+    attribute_regexp = /\[\s*(?<attribute>\w+)\s*=\s*(?<quote>"|')(?<value>\\.|.*?(?=\k<quote>))\k<quote>\s*\]/
     default_name = selector.gsub(attribute_regexp) do
       attribute = $1.strip
-      value = $2
+      value = $3
       unless known_attributes.include?(attribute)
         raise ArgumentError.new("Unkown aria attribute \"#{attribute}\" in selector")
       end
@@ -36,10 +36,11 @@ class Puppeteer::AriaQueryHandler
     end
   end
 
-  def wait_for(dom_world, selector, visible: nil, hidden: nil, timeout: nil)
+  def wait_for(dom_world, selector, visible: nil, hidden: nil, timeout: nil, root: nil)
+    # addHandlerToWorld
     binding_function = Puppeteer::DOMWorld::BindingFunction.new(
       name: 'ariaQuerySelector',
-      proc: -> (selector) { query_one(dom_world.send(:document), selector) },
+      proc: -> (sel) { query_one(root || dom_world.send(:document), sel) },
     )
     dom_world.send(:wait_for_selector_in_page,
       '(_, selector) => globalThis.ariaQuerySelector(selector)',
@@ -47,7 +48,9 @@ class Puppeteer::AriaQueryHandler
       visible: visible,
       hidden: hidden,
       timeout: timeout,
-      binding_function: binding_function)
+      binding_function: binding_function,
+      root: root,
+    )
   end
 
   def query_all(element, selector)

data/lib/puppeteer/browser_runner.rb

@@ -4,13 +4,17 @@ require 'timeout'
 
 # https://github.com/puppeteer/puppeteer/blob/master/lib/Launcher.js
 class Puppeteer::BrowserRunner
+  include Puppeteer::DebugPrint
+
   # @param {string} executablePath
   # @param {!Array<string>} processArguments
   # @param {string=} tempDirectory
-  def initialize(executable_path, process_arguments, temp_directory)
+  def initialize(for_firefox, executable_path, process_arguments, user_data_dir, using_temp_user_data_dir)
+    @for_firefox = for_firefox
     @executable_path = executable_path
     @process_arguments = process_arguments
-    @temp_directory = temp_directory
+    @user_data_dir = user_data_dir
+    @using_temp_user_data_dir = using_temp_user_data_dir
     @proc = nil
     @connection = nil
     @closed = true
@@ -90,8 +94,8 @@ class Puppeteer::BrowserRunner
     @process_closing = -> {
       @proc.dispose
       @closed = true
-      if @temp_directory
-        FileUtils.rm_rf(@temp_directory)
+      if @using_temp_user_data_dir
+        FileUtils.rm_rf(@user_data_dir)
       end
     }
     at_exit do
@@ -122,7 +126,7 @@ class Puppeteer::BrowserRunner
   def close
     return if @closed
 
-    if @temp_directory
+    if @using_temp_user_data_dir && !@for_firefox
       kill
     elsif @connection
       begin
@@ -137,11 +141,18 @@ class Puppeteer::BrowserRunner
 
   # @return {Promise}
   def kill
-    if @temp_directory
-      FileUtils.rm_rf(@temp_directory)
-    end
-    unless @closed
-      @proc.kill
+    # If the process failed to launch (for example if the browser executable path
+    # is invalid), then the process does not get a pid assigned. A call to
+    # `proc.kill` would error, as the `pid` to-be-killed can not be found.
+    @proc&.kill
+
+    # Attempt to remove temporary profile directory to avoid littering.
+    begin
+      if @using_temp_user_data_dir
+        FileUtils.rm_rf(@user_data_dir)
+      end
+    rescue => err
+      debug_puts(err)
     end
   end
 

data/lib/puppeteer/cdp_session.rb

@@ -15,6 +15,11 @@ class Puppeteer::CDPSession
     @session_id = session_id
   end
 
+  # @internal
+  def id
+    @session_id
+  end
+
   attr_reader :connection
 
   # @param method [String]

data/lib/puppeteer/custom_query_handler.rb

@@ -21,12 +21,12 @@ class Puppeteer::CustomQueryHandler
     nil
   end
 
-  def wait_for(dom_world, selector, visible: nil, hidden: nil, timeout: nil)
+  def wait_for(dom_world, selector, visible: nil, hidden: nil, timeout: nil, root: nil)
     unless @query_one
       raise NotImplementedError.new("#{self.class}##{__method__} is not implemented.")
     end
 
-    dom_world.send(:wait_for_selector_in_page, @query_one, selector, visible: visible, hidden: hidden, timeout: timeout)
+    dom_world.send(:wait_for_selector_in_page, @query_one, selector, visible: visible, hidden: hidden, timeout: timeout, root: root)
   end
 
   def query_all(element, selector)

data/lib/puppeteer/dom_world.rb

@@ -45,10 +45,14 @@ class Puppeteer::DOMWorld
     end
   end
 
-  # @param {!Puppeteer.FrameManager} frameManager
-  # @param {!Puppeteer.Frame} frame
-  # @param {!Puppeteer.TimeoutSettings} timeoutSettings
-  def initialize(frame_manager, frame, timeout_settings)
+  # @param client [Puppeteer::CDPSession]
+  # @param frame_manager [Puppeteer::FrameManager]
+  # @param frame [Puppeteer::Frame]
+  # @param timeout_settings [Puppeteer::TimeoutSettings]
+  def initialize(client, frame_manager, frame, timeout_settings)
+    # Keep own reference to client because it might differ from the FrameManager's
+    # client for OOP iframes.
+    @client = client
     @frame_manager = frame_manager
     @frame = frame
     @timeout_settings = timeout_settings
@@ -58,7 +62,7 @@ class Puppeteer::DOMWorld
     @ctx_bindings = Set.new
     @detached = false
 
-    frame_manager.client.on_event('Runtime.bindingCalled', &method(:handle_binding_called))
+    @client.on_event('Runtime.bindingCalled', &method(:handle_binding_called))
   end
 
   attr_reader :frame
@@ -416,10 +420,10 @@ class Puppeteer::DOMWorld
   # @param visible [Boolean] Wait for element visible (not 'display: none' nor 'visibility: hidden') on true. default to false.
   # @param hidden [Boolean] Wait for element invisible ('display: none' nor 'visibility: hidden') on true. default to false.
   # @param timeout [Integer]
-  def wait_for_selector(selector, visible: nil, hidden: nil, timeout: nil)
+  def wait_for_selector(selector, visible: nil, hidden: nil, timeout: nil, root: nil)
     # call wait_for_selector_in_page with custom query selector.
     query_selector_manager = Puppeteer::QueryHandlerManager.instance
-    query_selector_manager.detect_query_handler(selector).wait_for(self, visible: visible, hidden: hidden, timeout: timeout)
+    query_selector_manager.detect_query_handler(selector).wait_for(self, visible: visible, hidden: hidden, timeout: timeout, root: root)
   end
 
   private def binding_identifier(name, context)
@@ -493,10 +497,11 @@ class Puppeteer::DOMWorld
   # @param visible [Boolean] Wait for element visible (not 'display: none' nor 'visibility: hidden') on true. default to false.
   # @param hidden [Boolean] Wait for element invisible ('display: none' nor 'visibility: hidden') on true. default to false.
   # @param timeout [Integer]
-  private def wait_for_selector_in_page(query_one, selector, visible: nil, hidden: nil, timeout: nil, binding_function: nil)
+  private def wait_for_selector_in_page(query_one, selector, visible: nil, hidden: nil, timeout: nil, root: nil, binding_function: nil)
     option_wait_for_visible = visible || false
     option_wait_for_hidden = hidden || false
     option_timeout = timeout || @timeout_settings.timeout
+    option_root = root
 
     polling =
       if option_wait_for_visible || option_wait_for_hidden
@@ -507,11 +512,11 @@ class Puppeteer::DOMWorld
     title = "selector #{selector}#{option_wait_for_hidden ? 'to be hidden' : ''}"
 
     selector_predicate = make_predicate_string(
-      predicate_arg_def: '(selector, waitForVisible, waitForHidden)',
+      predicate_arg_def: '(root, selector, waitForVisible, waitForHidden)',
       predicate_query_handler: query_one,
       async: true,
       predicate_body: <<~JAVASCRIPT
-        const node = await predicateQueryHandler(document, selector)
+        const node = await predicateQueryHandler(root, selector)
         return checkWaitForOptions(node, waitForVisible, waitForHidden);
       JAVASCRIPT
     )
@@ -523,6 +528,7 @@ class Puppeteer::DOMWorld
       polling: polling,
       timeout: option_timeout,
       args: [selector, option_wait_for_visible, option_wait_for_hidden],
+      root: option_root,
       binding_function: binding_function,
     )
     handle = wait_task.await_promise
@@ -537,10 +543,11 @@ class Puppeteer::DOMWorld
   # @param visible [Boolean] Wait for element visible (not 'display: none' nor 'visibility: hidden') on true. default to false.
   # @param hidden [Boolean] Wait for element invisible ('display: none' nor 'visibility: hidden') on true. default to false.
   # @param timeout [Integer]
-  def wait_for_xpath(xpath, visible: nil, hidden: nil, timeout: nil)
+  def wait_for_xpath(xpath, visible: nil, hidden: nil, timeout: nil, root: nil)
     option_wait_for_visible = visible || false
     option_wait_for_hidden = hidden || false
     option_timeout = timeout || @timeout_settings.timeout
+    option_root = root
 
     polling =
       if option_wait_for_visible || option_wait_for_hidden
@@ -551,9 +558,9 @@ class Puppeteer::DOMWorld
     title = "XPath #{xpath}#{option_wait_for_hidden ? 'to be hidden' : ''}"
 
     xpath_predicate = make_predicate_string(
-      predicate_arg_def: '(selector, waitForVisible, waitForHidden)',
+      predicate_arg_def: '(root, selector, waitForVisible, waitForHidden)',
       predicate_body: <<~JAVASCRIPT
-        const node = document.evaluate(selector, document, null, XPathResult.FIRST_ORDERED_NODE_TYPE, null).singleNodeValue;
+        const node = document.evaluate(selector, root, null, XPathResult.FIRST_ORDERED_NODE_TYPE, null).singleNodeValue;
         return checkWaitForOptions(node, waitForVisible, waitForHidden);
       JAVASCRIPT
     )
@@ -565,6 +572,7 @@ class Puppeteer::DOMWorld
       polling: polling,
       timeout: option_timeout,
       args: [xpath, option_wait_for_visible, option_wait_for_hidden],
+      root: option_root,
     )
     handle = wait_task.await_promise
     unless handle.as_element

data/lib/puppeteer/element_handle.rb

@@ -20,6 +20,60 @@ class Puppeteer::ElementHandle < Puppeteer::JSHandle
     @disposed = false
   end
 
+  def inspect
+    values = %i[context remote_object page disposed].map do |sym|
+      value = instance_variable_get(:"@#{sym}")
+      "@#{sym}=#{value}"
+    end
+    "#<Puppeteer::ElementHandle #{values.join(' ')}>"
+  end
+
+  #
+  # Wait for the `selector` to appear within the element. If at the moment of calling the
+  # method the `selector` already exists, the method will return immediately. If
+  # the `selector` doesn't appear after the `timeout` milliseconds of waiting, the
+  # function will throw.
+  #
+  # This method does not work across navigations or if the element is detached from DOM.
+  #
+  # @param selector - A
+  # {@link https://developer.mozilla.org/en-US/docs/Web/CSS/CSS_Selectors | selector}
+  # of an element to wait for
+  # @param options - Optional waiting parameters
+  # @returns Promise which resolves when element specified by selector string
+  # is added to DOM. Resolves to `null` if waiting for hidden: `true` and
+  # selector is not found in DOM.
+  # @remarks
+  # The optional parameters in `options` are:
+  #
+  # - `visible`: wait for the selected element to be present in DOM and to be
+  # visible, i.e. to not have `display: none` or `visibility: hidden` CSS
+  # properties. Defaults to `false`.
+  #
+  # - `hidden`: wait for the selected element to not be found in the DOM or to be hidden,
+  # i.e. have `display: none` or `visibility: hidden` CSS properties. Defaults to
+  # `false`.
+  #
+  # - `timeout`: maximum time to wait in milliseconds. Defaults to `30000`
+  # (30 seconds). Pass `0` to disable timeout. The default value can be changed
+  # by using the {@link Page.setDefaultTimeout} method.
+  def wait_for_selector(selector, visible: nil, hidden: nil, timeout: nil)
+    frame = @context.frame
+
+    secondary_world = frame.secondary_world
+    adopted_root = secondary_world.execution_context.adopt_element_handle(self)
+    handle = secondary_world.wait_for_selector(selector, visible: visible, hidden: hidden, timeout: timeout, root: adopted_root)
+    adopted_root.dispose
+    return nil unless handle
+
+    main_world = frame.main_world
+    result = main_world.execution_context.adopt_element_handle(handle)
+    handle.dispose
+    result
+  end
+
+  define_async_method :async_wait_for_selector
+
   def as_element
     self
   end

data/lib/puppeteer/events.rb

@@ -81,6 +81,7 @@ module NetworkManagerEmittedEvents ; end
 
 {
   Request: EventsDefinitionUtils.symbol('NetworkManager.Request'),
+  RequestServedFromCache: EventsDefinitionUtils.symbol('NetworkManager.RequestServedFromCache'),
   Response: EventsDefinitionUtils.symbol('NetworkManager.Response'),
   RequestFailed: EventsDefinitionUtils.symbol('NetworkManager.RequestFailed'),
   RequestFinished: EventsDefinitionUtils.symbol('NetworkManager.RequestFinished'),

data/lib/puppeteer/frame.rb

@@ -1,25 +1,47 @@
 class Puppeteer::Frame
   using Puppeteer::DefineAsyncMethod
 
-  # @param {!FrameManager} frameManager
-  # @param {!Puppeteer.CDPSession} client
-  # @param {?Frame} parentFrame
-  # @param {string} frameId
-  def initialize(frame_manager, client, parent_frame, frame_id)
+  # @param frame_manager [Puppeteer::FrameManager]
+  # @param parent_frame [Puppeteer::Frame|nil]
+  # @param frame_id [String]
+  # @param client [Puppeteer::CDPSession]
+  def initialize(frame_manager, parent_frame, frame_id, client)
     @frame_manager = frame_manager
-    @client = client
     @parent_frame = parent_frame
     @id = frame_id
     @detached = false
 
     @loader_id = ''
     @lifecycle_events = Set.new
-    @main_world = Puppeteer::DOMWorld.new(frame_manager, self, frame_manager.timeout_settings)
-    @secondary_world = Puppeteer::DOMWorld.new(frame_manager, self, frame_manager.timeout_settings)
     @child_frames = Set.new
     if parent_frame
       parent_frame._child_frames << self
     end
+
+    update_client(client)
+  end
+
+  def inspect
+    values = %i[id parent_frame detached loader_id lifecycle_events child_frames].map do |sym|
+      value = instance_variable_get(:"@#{sym}")
+      "@#{sym}=#{value}"
+    end
+    "#<Puppeteer::Frame #{values.join(' ')}>"
+  end
+
+  def _client
+    @client
+  end
+
+  # @param client [Puppeteer::CDPSession]
+  private def update_client(client)
+    @client = client
+    @main_world = Puppeteer::DOMWorld.new(@client, @frame_manager, self, @frame_manager.timeout_settings)
+    @secondary_world = Puppeteer::DOMWorld.new(@client, @frame_manager, self, @frame_manager.timeout_settings)
+  end
+
+  def oop_frame?
+    @client != @frame_manager.client
   end
 
   attr_accessor :frame_manager, :id, :loader_id, :lifecycle_events, :main_world, :secondary_world