bidi2pdf 0.1.3 → 0.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 715f3587349e1680386f0c485f45b91a6096143c07cc8b531ca7e284a9516615
-  data.tar.gz: 8914fd60cdb04510070dee6b6d7134a94666584583681e26e830824bbf457059
+  metadata.gz: f132ccc5d9db5bb621f7e1a0adfa4150b77b4618542ba27873a602cf9c342a40
+  data.tar.gz: f3df228dc1de555d14baad4347a1b5b8d0bc8b63f88d6adfbcec86f0da571703
 SHA512:
-  metadata.gz: a8e17273531b0131ef9e8b5454154996dbd340d18cbf16a151478b91234e1e53005e0ffb057e16d540f6870b8915bafbdf03831afc2d9b4e9d861494ca13c38f
-  data.tar.gz: d0a4d3b193fc51b681a2e3b54da5cc7cccf05cf578b908144eb7ca5988327f0ee938b07dce88ca826d37889a469e3f32a7da61c68bbb05b5913731aeb0cc9e38
+  metadata.gz: d74e6e1b505493d47f66c2f1cb4d13bc07aacfa82dbba536439e3b458395bbcd7213103e5fbed167a65945af28bbfb239f6510e93bed2838fa583b95352df8d6
+  data.tar.gz: '096bb337a61db6d5392ba64c38bf4632695ca647b2fe95a227f3d56b189241cae1b19cc65e98115dff73997bb3df1cdcc2821058611683aa449acd11db4fe66f'

data/.rubocop.yml

@@ -20,11 +20,29 @@ RSpec/ExampleLength:
 Layout/MultilineMethodCallIndentation:
   Enabled: false
 
+Layout/FirstArgumentIndentation:
+  Enabled: false
+
+Layout/ClosingParenthesisIndentation:
+  Enabled: false
+
 Layout/FirstHashElementIndentation:
   EnforcedStyle: consistent
 
 Layout/FirstArrayElementIndentation:
-  EnforcedStyle: consistent
+  Enabled: false
+
+Layout/MultilineOperationIndentation:
+  Enabled: false
+
+Layout/BeginEndAlignment:
+  Enabled: false
+
+Layout/ArrayAlignment:
+  Enabled: false
+
+Layout/LineLength:
+  Enabled: false
 
 RSpec/MultipleMemoizedHelpers:
   Max: 10

data/CHANGELOG.md

@@ -6,12 +6,36 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+[unreleased]: https://github.com/dieter-medium/bidi2pdf/compare/v0.1.4..HEAD
 
-## [Unreleased]
+<!-- generated by git-cliff end -->
 
-[unreleased]: https://github.com/dieter-medium/bidi2pdf/compare/v0.1.2..HEAD
+## [0.1.4] - 2025-04-10
 
-<!-- generated by git-cliff end -->
+### 🎨 Refactored
+
+- Modularize browser commands and remove redundancy by @dieter-medium
+- Implement interceptor validation and unit tests by @dieter-medium
+- Connection handling in ConnectionManager by @dieter-medium
+- Interceptors to enforce common interface methods by @dieter-medium
+- Interceptors to simplify client responsibilities by @dieter-medium
+- Enhance error handling and modularize interceptors by @dieter-medium
+
+### 🐛 Fixed
+
+- Resolve race condition between event subscription and emission by @dieter-medium
+- Close custom session on stop function by @dieter-medium
+- Add custom error classes for session management and improve error handling and testability by @dieter-medium
+- Add Chromedriver container support and helper methods for RSpec integration by @dieter-medium
+
+### 🔄 Changed
+
+- Enhance command manager with response handling by @dieter-medium
+
+### 🚀 Added
+
+- Add randomization to user data directory path by @dieter-medium
+- Introduce command abstraction for improved modularity by @dieter-medium
 
 ## [0.1.3] - 2025-04-06
 

data/docker/Dockerfile.chromedriver

@@ -1,5 +1,7 @@
 FROM ruby:3.3
 
+ARG CHROMEDRIVER_PORT=3000
+
 # Install dependencies
 RUN apt-get update && \
     apt-get install -y \
@@ -30,4 +32,7 @@ USER appuser
 
 RUN gem install chromedriver-binary && ruby -e 'require "chromedriver/binary"; puts Chromedriver::Binary::ChromedriverDownloader.update'
 
+ENV CHROMEDRIVER_PORT=${CHROMEDRIVER_PORT}
+EXPOSE ${CHROMEDRIVER_PORT}
+
 CMD ["/usr/local/bin/entrypoint.sh"]

data/docker/entrypoint.sh

@@ -1,3 +1,13 @@
 #!/bin/bash
 
-/home/appuser/.webdrivers/chromedriver --port=3000 --headless --whitelisted-ips="" --allowed-origins="*" --disable-dev-shm-usage --verbose
+USER_DATA_DIR=/home/appuser/.cache
+mkdir -p ${USER_DATA_DIR}
+
+/home/appuser/.webdrivers/chromedriver --port=${CHROMEDRIVER_PORT} \
+                                       --headless \
+                                       --whitelisted-ips="" \
+                                       --allowed-origins="*" \
+                                       --disable-dev-shm-usage \
+                                       --disable-gpu \
+                                       --user-data-dir=${USER_DATA_DIR} \
+                                       --verbose

data/lib/bidi2pdf/bidi/add_headers_interceptor.rb

@@ -1,13 +1,21 @@
 # frozen_string_literal: true
 
+require_relative "interceptor"
+
 module Bidi2pdf
   module Bidi
     class AddHeadersInterceptor
-      attr_reader :id, :headers
+      include Interceptor
+
+      class << self
+        def phases = [Bidi2pdf::Bidi::Commands::AddIntercept::BEFORE_REQUEST]
+
+        def events = ["network.beforeRequestSent"]
+      end
+
+      attr_reader :headers, :url_patterns, :context
 
-      def initialize(id, headers:, client:)
-        @id = id
-        @client = client
+      def initialize(headers:, url_patterns:, context:)
         @headers = headers.map do |header|
           {
             name: header[:name],
@@ -17,26 +25,15 @@ module Bidi2pdf
             }
           }
         end
-      end
 
-      def handle_event(response)
-        event_response = response["params"]
-
-        return unless event_response["intercepts"]&.include?(id) && event_response["isBlocked"]
-
-        network_id = event_response["request"]["request"]
-
-        Bidi2pdf.logger.debug "Interceptor #{id} handle event: #{network_id}"
-
-        client.send_cmd "network.continueRequest", {
-          request: network_id,
-          headers: headers
-        }
+        @url_patterns = url_patterns
+        @context = context
       end
 
-      private
-
-      attr_reader :client
+      def process_interception(_event_response, _navigation_id, network_id, _url)
+        cmd = Bidi2pdf::Bidi::Commands::NetworkContinue.new request: network_id, headers: headers
+        client.send_cmd cmd
+      end
     end
   end
 end

data/lib/bidi2pdf/bidi/auth_interceptor.rb

@@ -1,67 +1,60 @@
 # frozen_string_literal: true
 
+require_relative "interceptor"
+
 module Bidi2pdf
   module Bidi
     class AuthInterceptor
-      attr_reader :id, :username, :password, :network_ids
+      include Interceptor
+
+      class << self
+        def phases = [Bidi2pdf::Bidi::Commands::AddIntercept::AUTH_REQUIRED]
+
+        def events = ["network.authRequired"]
+      end
 
-      def initialize(id, username:, password:, client:)
-        @id = id
-        @client = client
+      attr_reader :headers, :url_patterns, :context, :username, :password, :network_ids
+
+      def initialize(username:, password:, url_patterns:, context:)
+        @url_patterns = url_patterns
+        @context = context
         @username = username
         @password = password
         @network_ids = []
       end
 
-      # rubocop:disable Metrics/AbcSize
-      def handle_event(response)
-        event_response = response["params"]
-
-        return unless event_response["intercepts"]&.include?(id) && event_response["isBlocked"]
-
-        navigation_id = event_response["navigation"]
-        network_id = event_response["request"]["request"]
-        url = event_response["request"]["url"]
-
-        handle_bad_credentials(navigation_id, network_id, url)
+      def process_interception(_event_response, navigation_id, network_id, url)
+        return if handled_bad_credentials(navigation_id, network_id, url)
 
         network_ids << network_id
 
-        Bidi2pdf.logger.debug "Auth-Interceptor #{id} handle event: #{navigation_id}/#{network_id}/#{url}"
-
-        client.send_cmd("network.continueWithAuth", {
-          request: network_id,
-          action: "provideCredentials",
-          credentials: {
-            type: "password",
-            username: username,
-            password: password
-          }
-        })
-      end
+        cmd = Bidi2pdf::Bidi::Commands::ProvideCredentials.new request: network_id, username: username, password: password
 
-      # rubocop:enable Metrics/AbcSize
+        client.send_cmd(cmd)
+      rescue StandardError => e
+        Bidi2pdf.logger.error "Error handling auth event: #{e.message}"
+        Bidi2pdf.logger.error e.backtrace.join("\n")
+        raise e
+      end
 
       private
 
-      def handle_bad_credentials(navigation_id, network_id, url)
-        return unless network_ids.include?(network_id)
+      def handled_bad_credentials(navigation_id, network_id, url)
+        return false unless network_ids.include?(network_id)
 
         network_ids.delete(network_id)
 
-        Bidi2pdf.logger.debug "Auth-Interceptor #{id} already handled event: #{navigation_id}/#{network_id}/#{url}"
+        Bidi2pdf.logger.debug "Auth-Interceptor #{interceptor_id} already handled event: #{navigation_id}/#{network_id}/#{url}"
 
-        # rubocop: disable Layout/LineLength
         Bidi2pdf.logger.error "It seems that the same request is being intercepted multiple times. Check your credentials or the URL you are trying to access. If you are using a proxy, make sure it is configured correctly."
         # rubocop: enable Layout/LineLength
 
-        client.send_cmd("network.continueWithAuth", {
-          request: network_id,
-          action: "cancel"
-        })
-      end
+        cmd = Bidi2pdf::Bidi::Commands::CancelAuth.new request: network_id
+
+        client.send_cmd(cmd)
 
-      attr_reader :client
+        true
+      end
     end
   end
 end

data/lib/bidi2pdf/bidi/browser_tab.rb

@@ -3,7 +3,8 @@
 require "base64"
 
 require_relative "network_events"
-require_relative "print_parameters_validator"
+require_relative "auth_interceptor"
+require_relative "add_headers_interceptor"
 
 module Bidi2pdf
   module Bidi
@@ -20,10 +21,8 @@ module Bidi2pdf
       end
 
       def create_browser_tab
-        client.send_cmd_and_wait("browsingContext.create", {
-          type: "tab",
-          userContext: @user_context_id
-        }) do |response|
+        cmd = Bidi2pdf::Bidi::Commands::CreateTab.new(user_context_id: user_context_id)
+        client.send_cmd_and_wait(cmd) do |response|
           tab_browsing_context_id = response["result"]["context"]
 
           BrowserTab.new(client, tab_browsing_context_id, user_context_id).tap do |tab|
@@ -43,26 +42,18 @@ module Bidi2pdf
         same_site: "strict",
         ttl: 30
       )
-        expiry = Time.now.to_i + ttl
-        client.send_cmd_and_wait("storage.setCookie", {
-          cookie: {
-            name: name,
-            value: {
-              type: "string",
-              value: value
-            },
-            domain: domain,
-            path: path,
-            secure: secure,
-            httpOnly: http_only,
-            sameSite: same_site,
-            expiry: expiry
-          },
-          partition: {
-            type: "context",
-            context: browsing_context_id
-          }
-        }) do |response|
+        cmd = Bidi2pdf::Bidi::Commands::SetTabCookie.new(
+          browsing_context_id: browsing_context_id,
+          name: name,
+          value: value,
+          domain: domain,
+          path: path,
+          secure: secure,
+          http_only: http_only,
+          same_site: same_site,
+          ttl: ttl
+        )
+        client.send_cmd_and_wait(cmd) do |response|
           Bidi2pdf.logger.debug "Cookie set: #{response.inspect}"
         end
       end
@@ -71,43 +62,35 @@ module Bidi2pdf
         headers:,
         url_patterns:
       )
-        client.add_headers_interceptor(
+        AddHeadersInterceptor.new(
           context: browsing_context_id,
           url_patterns: url_patterns,
           headers: headers
-        )
+        ).tap { |interceptor| interceptor.register_with_client(client: client) }
       end
 
       def basic_auth(username:, password:, url_patterns:)
-        client.add_auth_interceptor(
+        AuthInterceptor.new(
           context: browsing_context_id,
           url_patterns: url_patterns,
-          username: username,
-          password: password
-        )
+          username: username, password: password
+        ).tap { |interceptor| interceptor.register_with_client(client: client) }
       end
 
       def open_page(url)
         client.on_event("network.responseStarted", "network.responseCompleted", "network.fetchError",
                         &network_events.method(:handle_event))
 
-        client.send_cmd_and_wait("browsingContext.navigate", {
-          url: url,
-          context: browsing_context_id,
-          wait: "complete"
-        }) do |response|
+        cmd = Bidi2pdf::Bidi::Commands::BrowsingContextNavigate.new url: url, context: browsing_context_id
+
+        client.send_cmd_and_wait(cmd) do |response|
           Bidi2pdf.logger.debug "Navigated to page url: #{url} response: #{response}"
         end
       end
 
       def execute_script(script)
-        client.send_cmd_and_wait("script.evaluate", {
-          expression: script,
-          target: {
-            context: browsing_context_id
-          },
-          awaitPromise: true
-        }) do |response|
+        cmd = Bidi2pdf::Bidi::Commands::ScriptEvaluate.new context: browsing_context_id, expression: script
+        client.send_cmd_and_wait(cmd) do |response|
           Bidi2pdf.logger.debug "Script Result: #{response.inspect}"
 
           response["result"]
@@ -128,13 +111,10 @@ module Bidi2pdf
         @open = false
       end
 
-      # rubocop:disable Metrics/AbcSize
       def print(outputfile, print_options: { background: true })
-        PrintParametersValidator.validate!(print_options)
-
-        cmd_params = (print_options || {}).merge(context: browsing_context_id)
+        cmd = Bidi2pdf::Bidi::Commands::BrowsingContextPrint.new context: browsing_context_id, print_options: print_options
 
-        client.send_cmd_and_wait("browsingContext.print", cmd_params) do |response|
+        client.send_cmd_and_wait(cmd) do |response|
           if response["result"]
             pdf_base64 = response["result"]["data"]
 
@@ -152,13 +132,13 @@ module Bidi2pdf
         end
       end
 
-      # rubocop:enable Metrics/AbcSize
-
       private
 
       def close_context
-        client.send_cmd_and_wait("browsingContext.close", { context: browsing_context_id }) do |response|
-          Bidi2pdf.logger.debug "Browsing context closed: #{response}"
+        that = self
+        cmd = Bidi2pdf::Bidi::Commands::BrowsingContextClose.new context: browsing_context_id
+        client.send_cmd_and_wait(cmd) do |response|
+          Bidi2pdf.logger.info "Browsing context closed: #{that.browsing_context_id} #{response}"
         end
       end
 

data/lib/bidi2pdf/bidi/client.rb

@@ -4,10 +4,9 @@ require "json"
 require "websocket-client-simple"
 
 require_relative "web_socket_dispatcher"
-require_relative "add_headers_interceptor"
-require_relative "auth_interceptor"
 require_relative "command_manager"
 require_relative "connection_manager"
+require_relative "commands"
 
 module Bidi2pdf
   module Bidi
@@ -19,20 +18,21 @@ module Bidi2pdf
       def initialize(ws_url)
         @ws_url = ws_url
         @started = false
+        @connection_manager = ConnectionManager.new(logger: Bidi2pdf.logger)
       end
 
       def start
         return @socket if started?
 
-        @socket = WebSocket::Client::Simple.connect(ws_url)
-
-        @connection_manager = ConnectionManager.new(logger: Bidi2pdf.logger)
-        @command_manager = CommandManager.new(@socket, logger: Bidi2pdf.logger)
+        WebSocket::Client::Simple.connect(ws_url) do |socket|
+          @socket = socket
+          @command_manager = CommandManager.new(@socket, logger: Bidi2pdf.logger)
 
-        dispatcher.on_open { @connection_manager.mark_connected }
-        dispatcher.on_message { |data| handle_response_to_cmd(data) }
+          dispatcher.on_open { @connection_manager.mark_connected }
+          dispatcher.on_message { |data| handle_response_to_cmd(data) }
+          dispatcher.start_listening
+        end
 
-        dispatcher.start_listening
         @started = true
 
         @socket
@@ -42,15 +42,23 @@ module Bidi2pdf
 
       def wait_until_open(timeout: Bidi2pdf.default_timeout)
         @connection_manager.wait_until_open(timeout: timeout)
+      rescue Bidi2pdf::WebsocketError => e
+        raise Bidi2pdf::WebsocketError, "Client#start must be called within #{timeout} sec." unless started?
+
+        raise e
       end
 
-      def send_cmd(method, params = {})
-        @command_manager.send_cmd(method, params)
+      def send_cmd(cmd)
+        raise Bidi2pdf::ClientError, "Client#start must be called before" unless started?
+
+        @command_manager.send_cmd(cmd)
       end
 
-      def send_cmd_and_wait(method, params = {}, timeout: Bidi2pdf.default_timeout, &block)
-        timed("Command #{method}") do
-          @command_manager.send_cmd_and_wait(method, params, timeout: timeout, &block)
+      def send_cmd_and_wait(cmd, timeout: Bidi2pdf.default_timeout, &block)
+        raise Bidi2pdf::ClientError, "Client#start must be called before" unless started?
+
+        timed("Command #{cmd.inspect}") do
+          @command_manager.send_cmd_and_wait(cmd, timeout: timeout, &block)
         end
       end
 
@@ -64,7 +72,8 @@ module Bidi2pdf
 
       def on_event(*names, &block)
         names.each { |name| dispatcher.on_event(name, &block) }
-        send_cmd("session.subscribe", { events: names }) if names.any?
+        cmd = Bidi2pdf::Bidi::Commands::SessionSubscribe.new(events: names)
+        send_cmd(cmd) if names.any?
       end
 
       def remove_message_listener(block) = dispatcher.remove_message_listener(block)
@@ -73,28 +82,6 @@ module Bidi2pdf
         names.each { |event_name| dispatcher.remove_event_listener(event_name, block) }
       end
 
-      def add_headers_interceptor(context:, url_patterns:, headers:)
-        add_interceptor(
-          context: context,
-          url_patterns: url_patterns,
-          phase: "beforeRequestSent",
-          event: "network.beforeRequestSent",
-          interceptor_class: AddHeadersInterceptor,
-          extra_args: { headers: headers }
-        )
-      end
-
-      def add_auth_interceptor(context:, url_patterns:, username:, password:)
-        add_interceptor(
-          context: context,
-          url_patterns: url_patterns,
-          phase: "authRequired",
-          event: "network.authRequired",
-          interceptor_class: AuthInterceptor,
-          extra_args: { username: username, password: password }
-        )
-      end
-
       def close
         return unless @socket
 
@@ -120,21 +107,6 @@ module Bidi2pdf
           Bidi2pdf.logger.warn "Unknown response: #{data.inspect}"
         end
       end
-
-      def add_interceptor(context:, url_patterns:, phase:, event:, interceptor_class:, extra_args: {})
-        send_cmd_and_wait("network.addIntercept", {
-          context: context,
-          phases: [phase],
-          urlPatterns: url_patterns
-        }) do |response|
-          id = response["result"]["intercept"]
-          Bidi2pdf.logger.debug "Interceptor added: #{id}"
-
-          interceptor_class.new(id, **extra_args, client: self).tap do |interceptor|
-            on_event(event, &interceptor.method(:handle_event))
-          end
-        end
-      end
     end
   end
 end

data/lib/bidi2pdf/bidi/command_manager.rb

@@ -3,18 +3,35 @@
 module Bidi2pdf
   module Bidi
     class CommandManager
+      class << self
+        def initialize_counter
+          @id = 0
+          @id_mutex = Mutex.new
+        end
+
+        def next_id = @id_mutex.synchronize { @id += 1 }
+      end
+
+      initialize_counter
+
       def initialize(socket, logger:)
         @socket = socket
         @logger = logger
 
-        @id = 0
-        @next_id_mutex = Mutex.new
         @pending_responses = {}
+        @initiated_cmds = {}
       end
 
-      def send_cmd(method, params = {})
+      def send_cmd(cmd, store_response: false)
         id = next_id
-        payload = { id: id, method: method, params: params }
+
+        if store_response
+          init_queue_for id
+        else
+          @initiated_cmds[id] = true
+        end
+
+        payload = cmd.as_payload(id)
 
         @logger.debug "Sending command: #{redact_sensitive_fields(payload).inspect}"
         @socket.send(payload.to_json)
@@ -22,40 +39,46 @@ module Bidi2pdf
         id
       end
 
-      def send_cmd_and_wait(method, params = {}, timeout: Bidi2pdf.default_timeout)
-        id = send_cmd(method, params)
-        queue = @pending_responses[id]
+      def send_cmd_and_wait(cmd, timeout: Bidi2pdf.default_timeout)
+        id = send_cmd(cmd, store_response: true)
+        response = pop_response id, timeout: timeout
 
-        response = queue.pop(timeout: timeout)
-        raise_timeout_error(id, method, params) if response.nil?
-        raise "Error response: #{response["error"]}" if response["error"]
+        raise_timeout_error(id, cmd) if response.nil?
+        raise CmdError, "Error response: #{response["error"]} #{cmd.inspect}" if response["error"]
 
         block_given? ? yield(response) : response
       ensure
         @pending_responses.delete(id)
       end
 
-      def queue_for(id)
-        @pending_responses[id]
+      def pop_response(id, timeout:)
+        raise CmdResponseNotStoredError, "No response stored for command ID #{id} or already popped or this command was not send" unless @pending_responses.key?(id)
+
+        @pending_responses[id].pop(timeout: timeout)
+      ensure
+        @pending_responses.delete(id)
       end
 
       def handle_response(data)
-        if (id = data["id"]) && @pending_responses.key?(id)
-          @pending_responses[id]&.push(data)
-        else
-          false
+        if (id = data["id"])
+          if @pending_responses.key?(id)
+            @pending_responses[id]&.push(data)
+            return true
+          elsif @initiated_cmds.key?(id)
+            @logger.error "Received error: #{data["error"]} for cmd: #{id}" if data["error"]
+
+            return @initiated_cmds.delete(id)
+          end
         end
+
+        false
       end
 
       private
 
-      def next_id
-        @next_id_mutex.synchronize do
-          @id += 1
-          @pending_responses[@id] = Thread::Queue.new
-          @id
-        end
-      end
+      def init_queue_for(id) = @pending_responses[id] = Thread::Queue.new
+
+      def next_id = self.class.next_id
 
       def redact_sensitive_fields(obj, sensitive_keys = %w[value token password authorization username])
         case obj
@@ -71,11 +94,10 @@ module Bidi2pdf
         end
       end
 
-      def raise_timeout_error(id, method, params)
-        # rubocop:disable Layout/LineLength
-        @logger.error "Timeout waiting for response to command #{id}, cmd: #{method}, params: #{redact_sensitive_fields(params).inspect}"
-        # rubocop:enable Layout/LineLength
-        raise "Timeout waiting for response to command ID #{id}"
+      def raise_timeout_error(id, cmd)
+        @logger.error "Timeout waiting for response to command #{id}, cmd: #{cmd.inspect}"
+
+        raise CmdTimeoutError, "Timeout waiting for response to command ID #{id}"
       end
     end
   end