ferrum_pdf 2.1.0 → 3.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e10a66bff6f06cfd7388c12247ca55e92a79bc48ff741e889e0ac0bfde0286e0
-  data.tar.gz: 2a92a79c5a7ca34091644fcd5795d1d1c84d49f7d834f02afd6d9ec462ba892b
+  metadata.gz: a0c9bdbd12753b4efc5b29bb208908735218fc2614e45eb74e57b6623ca61688
+  data.tar.gz: cb96c86e8c45a34037db93555c9ac5d692f7ccc4d0f37c88e211e6dcb17f7aa1
 SHA512:
-  metadata.gz: c6d4472cb10faebb29ab64634f16063c7aac7927baf76e3e58dc72478d3d21819935138e6595a872f9e0766ff51a17303cad51a1857d62110b2e13fd4877234f
-  data.tar.gz: 9915cadbed613d26c113a57d5a23268ec3758ef9f566b5f6f387469ed4cce948d5348173a840cf9d36a44d935fda1a3e5f7cb06af5c5f042d73f2fb06707a892
+  metadata.gz: b3e3efc16d9c2601e6ef725d9d1aa6f73d631c8dcfaff8a4aa76a2254e933d7205e33f00a98ac780c7cf0b99fa074e2d820abee42de372fb68966a00fb533faa
+  data.tar.gz: 292b03680170140e5e901f8e84128970cb88a17e17147b57142ad686edca9c31da24c327e8c733505e6188a50310cac7d112a3ce1f92f89ed50b301ec7a0cd9b

data/README.md

@@ -1,6 +1,6 @@
 # 📄 FerrumPdf
 
-PDFs & screentshots for Rails using [Ferrum](https://github.com/rubycdp/ferrum) & headless Chrome.
+PDFs & screenshots for Rails using [Ferrum](https://github.com/rubycdp/ferrum) & headless Chrome.
 
 Inspired by [Grover](https://github.com/Studiosity/grover), but without the Node.js and puppeteer dependencies. 🎉
 
@@ -70,10 +70,11 @@ FerrumPdf.render_pdf(
   url: "https://example.com/page", # Provide a URL to the content
 
   html: content, # or provide HTML
-  base_url: request.base_url, # Preprocesses `html` to convert relative paths and protocols. Example: "https://example.org"
+  display_url: request.original_url, # When supplying content via :html its best to give Chrome a hint of the current url so that it can process relative paths in the document. If you don't provide this, http://example.com will be used instead.
 
   authorize: { user: "username", password: "password" }, # Used for authenticating with basic auth
   wait_for_idle_options: { connections: 0, duration: 0.05, timeout: 5 }, # Used for setting network wait_for_idle options
+  timeout_if_open_connections: true,
 
   pdf_options: {
     landscape: false, # paper orientation
@@ -166,7 +167,7 @@ FerrumPdf.render_screenshot(
   url: "https://example.com/page", # Provide a URL to the content
 
   html: content, # or provide HTML
-  base_url: request.base_url, # Preprocesses `html` to convert relative paths and protocols. Example: "https://example.org"
+  display_url: request.original_url, # When supplying content via :html its best to give Chrome a hint of the current url so that it can process relative paths in the document. If you don't provide this, http://example.com will be used instead.
 
   screenshot_options: {
     format: "png" # or "jpeg"
@@ -186,7 +187,6 @@ You can set default values for page loads, PDF renders, and screenshot renders w
 
 ```ruby
 FerrumPdf.configure do |config|
-  config.page_options.base_url = "https://example.com/"
   config.page_options.authorize = { user: "username", password: "password" }
   config.page_options.wait_for_idle_options = { timeout: 90 }
   config.page_options.retries = 3

data/lib/ferrum_pdf/railtie.rb

@@ -1,11 +1,5 @@
 module FerrumPdf
   class Railtie < ::Rails::Railtie
-    initializer "ferrum_pdf.assets_helper" do
-      ActiveSupport.on_load(:action_view) do
-        include FerrumPdf::AssetsHelper
-      end
-    end
-
     initializer "ferrum_pdf.controller" do
       ActiveSupport.on_load(:action_controller) do
         # render ferrum_pdf: { pdf options }, template: "whatever", disposition: :inline, filename: "example.pdf"
@@ -13,7 +7,7 @@ module FerrumPdf
           send_data_options = options.extract!(:disposition, :filename, :status)
           url = pdf_options.delete(:url)
           html = render_to_string(**options.with_defaults(formats: [ :html ])) if url.blank?
-          pdf = FerrumPdf.render_pdf(html: html, base_url: request.base_url, url: url, pdf_options: pdf_options)
+          pdf = FerrumPdf.render_pdf(html: html, display_url: request.original_url, url: url, pdf_options: pdf_options)
           send_data(pdf, **send_data_options.with_defaults(type: :pdf))
         end
 
@@ -22,7 +16,7 @@ module FerrumPdf
           send_data_options = options.extract!(:disposition, :filename, :status)
           url = screenshot_options.delete(:url)
           html = render_to_string(**options.with_defaults(formats: [ :html ])) if url.blank?
-          screenshot = FerrumPdf.render_screenshot(url: url, html: html, base_url: request.base_url, screenshot_options: screenshot_options)
+          screenshot = FerrumPdf.render_screenshot(url: url, html: html, display_url: request.original_url, screenshot_options: screenshot_options)
           send_data(screenshot, **send_data_options.with_defaults(type: screenshot_options.fetch(:format, :png)))
         end
       end

data/lib/ferrum_pdf/version.rb

@@ -1,3 +1,3 @@
 module FerrumPdf
-  VERSION = "2.1.0"
+  VERSION = "3.0.0"
 end

data/lib/ferrum_pdf.rb

@@ -9,9 +9,6 @@ module FerrumPdf
     <div class='text right'><span class='pageNumber'></span>/<span class='totalPages'></span></div>
   HTML
 
-  autoload :AssetsHelper, "ferrum_pdf/assets_helper"
-  autoload :HTMLPreprocessor, "ferrum_pdf/html_preprocessor"
-
   mattr_accessor :browser_mutex, default: Mutex.new
   mattr_accessor :config, default: ActiveSupport::OrderedOptions.new.merge(
     window_size: [ 1920, 1080 ],
@@ -57,9 +54,9 @@ module FerrumPdf
     #   render_pdf(url: "https://example.org/receipts/example.pdf")
     #   render_pdf(html: "<h1>Hello world</h1>")
     #
-    # For rendering HTML, we also need the base_url for preprocessing URLs with relative paths & protocols
+    # For rendering HTML, we also need display_url for so that Chrome can interpret URLs with relative paths & protocols
     #
-    #   render_pdf(html: "<h1>Hello world</h1>", base_url: "https://example.org/")
+    #   render_pdf(html: "<h1>Hello world</h1>", display_url: "https://example.org/hello_world")
     #
     def render_pdf(pdf_options: {}, **load_page_args)
       load_page(**load_page_args) do |browser, page|
@@ -73,9 +70,9 @@ module FerrumPdf
     # render_screenshot(url: "https://example.org/receipts/example.pdf")
     # render_screenshot(html: "<h1>Hello world</h1>")
     #
-    # For rendering HTML, we also need the base_url for preprocessing URLs with relative paths & protocols
+    # For rendering HTML, we also need display_url for so that Chrome can interpret URLs with relative paths & protocols
     #
-    #   render_screenshot(html: "<h1>Hello world</h1>", base_url: "https://example.org/")
+    #   render_screenshot(html: "<h1>Hello world</h1>", display_url: "https://example.org/hello_world")
     #
     def render_screenshot(screenshot_options: {}, **load_page_args)
       load_page(**load_page_args) do |browser, page|
@@ -86,14 +83,12 @@ module FerrumPdf
 
     # Loads page into the browser to be used for rendering PDFs or screenshots
     #
-    # This automatically applies HTML preprocessing if `html:` is present
-    #
-    def load_page(url: nil, html: nil, base_url: nil, authorize: nil, wait_for_idle_options: nil, browser: nil, retries: nil)
+    def load_page(url: nil, html: nil, display_url: nil, authorize: nil, wait_for_idle_options: nil, timeout_if_open_connections: true, browser: nil, retries: nil)
       try ||= 0
       authorize ||= config.dig(:page_options, :authorize)
-      base_url ||= config.dig(:page_options, :base_url)
       retries ||= config.page_options.fetch(:retries, 1)
       wait_for_idle_options = config.page_options.fetch(:wait_for_idle_options, {}).merge(wait_for_idle_options || {})
+      timeout_if_open_connections ||= config.page_options.fetch(:timeout_if_open_connections, true)
 
       with_browser(browser) do |browser|
         # Closes page automatically after block finishes
@@ -103,13 +98,24 @@ module FerrumPdf
 
           # Load content
           if html
-            page.content = FerrumPdf::HTMLPreprocessor.process(html, base_url)
+            html_intercepted = false
+            page.network.intercept
+            page.on(:request) do |request|
+              if html_intercepted
+                request.continue
+              else
+                html_intercepted = true
+                request.respond(body: html.blank? ? " " : html)
+              end
+            end
+            page.go_to(display_url || "http://example.com")
           else
             page.go_to(url)
           end
 
           # Wait for everything to load
-          page.network.wait_for_idle!(**wait_for_idle_options)
+          idle = page.network.wait_for_idle(**wait_for_idle_options)
+          raise Ferrum::TimeoutError if timeout_if_open_connections && !idle
 
           yield browser, page
         end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: ferrum_pdf
 version: !ruby/object:Gem::Version
-  version: 2.1.0
+  version: 3.0.0
 platform: ruby
 authors:
 - Chris Oliver
@@ -49,8 +49,6 @@ files:
 - README.md
 - Rakefile
 - lib/ferrum_pdf.rb
-- lib/ferrum_pdf/assets_helper.rb
-- lib/ferrum_pdf/html_preprocessor.rb
 - lib/ferrum_pdf/railtie.rb
 - lib/ferrum_pdf/version.rb
 homepage: https://github.com/excid3/ferrum_pdf
@@ -74,7 +72,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.7.1
+rubygems_version: 3.7.2
 specification_version: 4
 summary: PDFs & screenshots for Rails using Ferrum & headless Chrome
 test_files: []

data/lib/ferrum_pdf/assets_helper.rb

@@ -1,65 +0,0 @@
-module FerrumPdf
-  class BaseAsset
-    def initialize(asset)
-      @asset = asset
-    end
-  end
-
-  class PropshaftAsset < BaseAsset
-    def content_type
-      @asset.content_type.to_s
-    end
-
-    def content
-      @asset.content
-    end
-  end
-
-  class SprocketsAsset < BaseAsset
-    def content_type
-      @asset.content_type
-    end
-
-    def content
-      @asset.source
-    end
-  end
-
-  class AssetFinder
-    class << self
-      def find(path)
-        if Rails.application.assets.respond_to?(:load_path)
-          propshaft_asset(path)
-        elsif Rails.application.assets.respond_to?(:find_asset)
-          sprockets_asset(path)
-        else
-          nil
-        end
-      end
-
-      def propshaft_asset(path)
-        (asset = Rails.application.assets.load_path.find(path)) ? PropshaftAsset.new(asset) : nil
-      end
-
-      def sprockets_asset(path)
-        (asset = Rails.application.assets.find_asset(path)) ? SprocketsAsset.new(asset) : nil
-      end
-    end
-  end
-
-  module AssetsHelper
-    def ferrum_pdf_inline_stylesheet(path)
-      (asset = AssetFinder.find(path)) ? "<style>#{asset.content}</style>".html_safe : nil
-    end
-
-    def ferrum_pdf_inline_javascript(path)
-      (asset = AssetFinder.find(path)) ? "<script>#{asset.content}</script>".html_safe : nil
-    end
-
-    def ferrum_pdf_base64_asset(path)
-      return nil unless (asset = AssetFinder.find(path))
-
-      "data:#{asset.content_type};base64,#{Base64.encode64(asset.content).gsub(/\s+/, '')}"
-    end
-  end
-end

data/lib/ferrum_pdf/html_preprocessor.rb

@@ -1,33 +0,0 @@
-module FerrumPdf
-  # Helper module for preparing HTML for conversion
-  #
-  # Sourced from the PDFKit project
-  # @see https://github.com/pdfkit/pdfkit
-  module HTMLPreprocessor
-    # Change relative paths to absolute, and relative protocols to absolute protocols
-    #
-    #   process("Some HTML", "https://example.org")
-    #
-    def self.process(html, base_url)
-      return html if base_url.blank?
-
-      base_url += "/" unless base_url.end_with? "/"
-      protocol = base_url.split("://").first
-      html = translate_relative_paths(html, base_url) if base_url
-      html = translate_relative_protocols(html, protocol) if protocol
-      html
-    end
-
-    def self.translate_relative_paths(html, base_url)
-      # Try out this regexp using rubular http://rubular.com/r/hiAxBNX7KE
-      html.gsub(%r{(href|src)=(['"])/([^/"']([^"']*|[^"']*))?['"]}, "\\1=\\2#{base_url}\\3\\2")
-    end
-    private_class_method :translate_relative_paths
-
-    def self.translate_relative_protocols(body, protocol)
-      # Try out this regexp using rubular http://rubular.com/r/0Ohk0wFYxV
-      body.gsub(%r{(href|src)=(['"])//([^"']*|[^"']*)['"]}, "\\1=\\2#{protocol}://\\3\\2")
-    end
-    private_class_method :translate_relative_protocols
-  end
-end