Updates HTML5::Document.parse with keyword arguments;

lib/nokogiri/html5.rb

@@ -46,16 +46,22 @@ def self.HTML5(...)
   # The document and fragment parsing methods support options that are different from
   # Nokogiri::HTML4::Document or Nokogiri::XML::Document.
   #
-  # - <tt>Nokogiri.HTML5(html, url = nil, encoding = nil, **options)</tt>
-  # - <tt>Nokogiri::HTML5.parse(html, url = nil, encoding = nil, **options)</tt>
-  # - <tt>Nokogiri::HTML5::Document.parse(html, url = nil, encoding = nil, **options)</tt>
-  # - <tt>Nokogiri::HTML5.fragment(html, encoding = nil, **options)</tt>
-  # - <tt>Nokogiri::HTML5::DocumentFragment.parse(html, encoding = nil, **options)</tt>
+  # - <tt>Nokogiri.HTML5(html, url:, encoding:, **parse_options)</tt>
+  # - <tt>Nokogiri::HTML5.parse(html, url:, encoding:, **parse_options)</tt>
+  # - <tt>Nokogiri::HTML5::Document.parse(html, url:, encoding:, **parse_options)</tt>
+  # - <tt>Nokogiri::HTML5.fragment(html, encoding = nil, **parse_options)</tt>
+  # - <tt>Nokogiri::HTML5::DocumentFragment.parse(html, encoding = nil, **parse_options)</tt>
   #
-  # The four currently supported options are +:max_errors+, +:max_tree_depth+, +:max_attributes+,
-  # and +:parse_noscript_content_as_text+ described below.
+  # The four currently supported parse options are
   #
-  # === Error reporting
+  # - +max_errors:+ (Integer, default 0) Maximum number of parse errors to report in HTML5::Document#errors.
+  # - +max_tree_depth:+ (Integer, default +Nokogiri::Gumbo::DEFAULT_MAX_TREE_DEPTH+) Maximum tree depth to parse.
+  # - +max_attributes:+ (Integer, default +Nokogiri::Gumbo::DEFAULT_MAX_ATTRIBUTES+) Maximum number of attributes to parse per element.
+  # - +parse_noscript_content_as_text:+ (Boolean, default false) When enabled, parse +noscript+ tag content as text, mimicking the behavior of web browsers.
+  #
+  # These options are explained in the following sections.
+  #
+  # === Error reporting: +max_errors:+
   #
   # Nokogiri contains an experimental HTML5 parse error reporting facility. By default, no parse
   # errors are reported but this can be configured by passing the +:max_errors+ option to
@@ -112,7 +118,7 @@ def self.HTML5(...)
   # are not part of Nokogiri's public API. That is, these are subject to change without Nokogiri's
   # major version number changing. These may be stabilized in the future.
   #
-  # === Maximum tree depth
+  # === Maximum tree depth: +max_tree_depth:+
   #
   # The maximum depth of the DOM tree parsed by the various parsing methods is configurable by the
   # +:max_tree_depth+ option. If the depth of the tree would exceed this limit, then an
@@ -126,7 +132,7 @@ def self.HTML5(...)
   #   # raises ArgumentError: Document tree depth limit exceeded
   #   doc = Nokogiri.HTML5(html, max_tree_depth: -1)
   #
-  # === Attribute limit per element
+  # === Attribute limit per element: +max_attributes:+
   #
   # The maximum number of attributes per DOM element is configurable by the +:max_attributes+
   # option. If a given element would exceed this limit, then an +ArgumentError+ is thrown.
@@ -142,7 +148,7 @@ def self.HTML5(...)
   #   doc = Nokogiri.HTML5(html, max_attributes: -1)
   #   # parses successfully
   #
-  # === Parse +noscript+ elements' content as text
+  # === Parse +noscript+ elements' content as text: +parse_noscript_content_as_text:+
   #
   # By default, the content of +noscript+ elements is parsed as HTML elements. Browsers that
   # support scripting parse the content of +noscript+ elements as raw text.

lib/nokogiri/html5/document.rb

@@ -43,41 +43,54 @@ class Document < Nokogiri::HTML4::Document
 
       # Get the parser's quirks mode value. See HTML5::QuirksMode.
       #
-      # This method returns `nil` if the parser was not invoked (e.g., `Nokogiri::HTML5::Document.new`).
+      # This method returns +nil+ if the parser was not invoked (e.g., Nokogiri::HTML5::Document.new).
       #
       # Since v1.14.0
       attr_reader :quirks_mode
 
       class << self
         # :call-seq:
-        #   parse(input)
-        #   parse(input, url=nil, encoding=nil, **options)
-        #   parse(input, url=nil, encoding=nil) { |options| ... }
+        #   parse(input) { |parse_options| ... }
+        #   parse(input, url:, encoding:, **parse_options)
         #
-        # Parse HTML5 input.
+        # Parse \HTML input with a parser compliant with the HTML5 spec. This method uses the
+        # encoding of +input+ if it can be determined, or else falls back to the +encoding:+
+        # parameter.
         #
-        # [Parameters]
-        # - +input+ may be a String, or any object that responds to _read_ and _close_ such as an
-        #   IO, or StringIO.
+        # [Required Parameters]
+        # - +input+ (String | IO) the \HTML content to be parsed.
         #
-        # - +url+ (optional) is a String indicating the canonical URI where this document is located.
+        # [Optional Parameters]
+        # - +url:+ (String) the base URI of the document.
+        # - +encoding+ (Encoding) The encoding that should be used when processing the
+        #   document. This option is only used as a fallback when the encoding of +input+ cannot be
+        #   determined.
+        # - +parse_options+ (Hash) represents keywords arguments that control the behavior of the
+        #   parser. See rdoc-ref:HTML5@Parsing+options for a list of available options.
         #
-        # - +encoding+ (optional) is the encoding that should be used when processing
-        #   the document.
+        # [Yields]
+        #   If present, the block will be passed a Hash object to modify with parse options before the
+        #   input is parsed. See rdoc-ref:HTML5@Parsing+options for a list of available options.
         #
-        # - +options+ (optional) is a configuration Hash (or keyword arguments) to set options
-        #   during parsing. The three currently supported options are +:max_errors+,
-        #   +:max_tree_depth+ and +:max_attributes+, described at Nokogiri::HTML5.
+        #   ⚠ Note that +url:+ and +encoding:+ cannot be set by the configuration block.
         #
-        #   ⚠ Note that these options are different than those made available by
-        #   Nokogiri::XML::Document and Nokogiri::HTML4::Document.
+        # [Returns] Nokogiri::HTML5::Document
         #
-        # - +block+ (optional) is passed a configuration Hash on which parse options may be set. See
-        #   Nokogiri::HTML5 for more information and usage.
+        # *Example:* Parse a string with a specific encoding and custom max errors limit.
         #
-        # [Returns] Nokogiri::HTML5::Document
+        #   Nokogiri::HTML5::Document.parse(socket, encoding: "ISO-8859-1", max_errors: 10)
+        #
+        # *Example:* Parse a string setting the +:parse_noscript_content_as_text+ option using the
+        # configuration block parameter.
+        #
+        #   Nokogiri::HTML5::Document.parse(input) { |c| c[:parse_noscript_content_as_text] = true }
         #
-        def parse(string_or_io, url = nil, encoding = nil, **options, &block)
+        def parse(
+          string_or_io,
+          url_ = nil, encoding_ = nil,
+          url: url_, encoding: encoding_,
+          **options, &block
+        )
           yield options if block
           string_or_io = "" unless string_or_io
 
@@ -98,7 +111,7 @@ def parse(string_or_io, url = nil, encoding = nil, **options, &block)
         # Create a new document from an IO object.
         #
         # 💡 Most users should prefer Document.parse to this method.
-        def read_io(io, url = nil, encoding = nil, **options)
+        def read_io(io, url_ = nil, encoding_ = nil, url: url_, encoding: encoding_, **options)
           raise ArgumentError, "io object doesn't respond to :read" unless io.respond_to?(:read)
 
           do_parse(io, url, encoding, **options)
@@ -107,7 +120,7 @@ def read_io(io, url = nil, encoding = nil, **options)
         # Create a new document from a String.
         #
         # 💡 Most users should prefer Document.parse to this method.
-        def read_memory(string, url = nil, encoding = nil, **options)
+        def read_memory(string, url_ = nil, encoding_ = nil, url: url_, encoding: encoding_, **options)
           raise ArgumentError, "string object doesn't respond to :to_str" unless string.respond_to?(:to_str)
 
           do_parse(string, url, encoding, **options)
@@ -144,7 +157,8 @@ def initialize(*args) # :nodoc:
       # - +markup+ (String) The HTML5 markup fragment to be parsed
       #
       # [Returns]
-      #   Nokogiri::HTML5::DocumentFragment. This object's children will be empty if `markup` is not passed, is empty, or is `nil`.
+      #   Nokogiri::HTML5::DocumentFragment. This object's children will be empty if +markup+ is not
+      #   passed, is empty, or is +nil+.
       #
       def fragment(markup = nil)
         DocumentFragment.new(self, markup)

test/html5/test_api.rb

@@ -42,6 +42,22 @@ def test_url
 
     doc = Nokogiri::HTML5(html, url, max_errors: 1)
     assert_equal(url, doc.errors[0].file)
+
+    # with keyword args
+    doc = Nokogiri::HTML5::Document.parse(html, url: nil)
+    assert_nil(doc.url)
+
+    doc = Nokogiri::HTML5::Document.parse(html, url: url)
+    assert_equal(url, doc.url)
+
+    doc = Nokogiri::HTML5::Document.parse(html, url: url, max_errors: 1)
+    assert_equal(url, doc.errors[0].file)
+
+    doc = Nokogiri::HTML5.parse(html, url: url, max_errors: 1)
+    assert_equal(url, doc.errors[0].file)
+
+    doc = Nokogiri::HTML5(html, url: url, max_errors: 1)
+    assert_equal(url, doc.errors[0].file)
   end
 
   def test_parse_encoding
@@ -57,6 +73,11 @@ def test_parse_encoding
     assert_match(/おはようございます/, Nokogiri::HTML5(raw, nil, Encoding::SHIFT_JIS).to_s)
     assert_match(/おはようございます/, Nokogiri::HTML5.parse(raw, nil, Encoding::SHIFT_JIS).to_s)
     assert_match(/おはようございます/, Nokogiri::HTML5::Document.parse(raw, nil, Encoding::SHIFT_JIS).to_s)
+
+    # with kwargs
+    assert_match(/おはようございます/, Nokogiri::HTML5(raw, encoding: Encoding::SHIFT_JIS).to_s)
+    assert_match(/おはようございます/, Nokogiri::HTML5.parse(raw, encoding: Encoding::SHIFT_JIS).to_s)
+    assert_match(/おはようございます/, Nokogiri::HTML5::Document.parse(raw, encoding: Encoding::SHIFT_JIS).to_s)
   end
 
   def test_fragment_encoding

test/html5/test_nokogumbo.rb

@@ -205,21 +205,29 @@ def test_fragment_default_max_attributes
     assert_raises(ArgumentError) { Nokogiri::HTML5.fragment(html) }
   end
 
+  TWO_ERROR_DOC = "<!DOCTYPE html><html><!-- <!-- --></a>"
+
   def test_parse_errors
-    doc = Nokogiri::HTML5("<!DOCTYPE html><html><!-- <!-- --></a>", max_errors: 10)
+    doc = Nokogiri::HTML5(TWO_ERROR_DOC, max_errors: 10)
     assert_equal(2, doc.errors.length)
     doc = Nokogiri::HTML5("<!DOCTYPE html><html>", max_errors: 10)
     assert_empty(doc.errors)
   end
 
   def test_max_errors
     # This document contains 2 parse errors, but we force limit to 1.
-    doc = Nokogiri::HTML5("<!DOCTYPE html><html><!-- -- --></a>", max_errors: 1)
+    doc = Nokogiri::HTML5(TWO_ERROR_DOC, max_errors: 1)
     assert_equal(1, doc.errors.length)
     doc = Nokogiri::HTML5("<!DOCTYPE html><html>", max_errors: 1)
     assert_empty(doc.errors)
   end
 
+  def test_max_errors_with_config_block
+    # This document contains 2 parse errors, but we force limit to 1.
+    doc = Nokogiri::HTML5(TWO_ERROR_DOC) { |c| c[:max_errors] = 1 }
+    assert_equal(1, doc.errors.length)
+  end
+
   def test_default_max_errors
     # This document contains 200 parse errors, but default limit is 0.
     doc = Nokogiri::HTML5("<!DOCTYPE html><html>" + "</p>" * 200)