nokogiri-html-ext 1.3.0 → 1.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4639298a3f7cf742a7a82c93c615e93c09120d1029889f7994e954edade6fe8e
-  data.tar.gz: 1f3a16c7057aa11b6ec51ed28a6f7d22f64c41bac121ab1765a1a77c7275a65e
+  metadata.gz: 81844b7536086078b24d1600282bb324c1325cc62b527b45bf81a13090ff413a
+  data.tar.gz: fdda5725873079052a6315c6d19a388dce12586c54ed88cb42321f215f2826fe
 SHA512:
-  metadata.gz: afdba65269d468d672cad6559772b0afeac2e7d1a9201c0052e460ad9acb897fb50c7e6fcd8a85551a3343ab4306dc05d59c2403750d3e91fed6dca23bc3bb2a
-  data.tar.gz: 3e85242200eb798f68903a247a992980fb3ce6142077b83b6008f0cc6297684dcfd47a4773464d45c5800022a90cf355354c6d65194c32ebb9667d683c5ebe28
+  metadata.gz: 307a4c317c840578338d253d1c6c8586ab863ec97f32cde0339069c8a096220c6863d718d305f000357cee68fdfb4e5b36e7cc8a91eb27e7bcd55045b7cf7a77
+  data.tar.gz: d1701b55fe58d9b80378b2a9b09a5c73a0c528895869aa99ecd01ce6d4b291c6108708b5eb6ea55c1201e545c975e85af14489b58272282473451b27a4c2e929

data/README.md

@@ -79,11 +79,12 @@ doc.at_css("base").to_s
 
 nokogiri-html-ext will resolve a document's relative URLs against a provided source URL. The source URL _should_ be an absolute URL (e.g. `https://jgarber.example`) representing the location of the document being parsed. The source URL _may_ be any `String` (or any Ruby object that responds to `#to_s`).
 
-nokogiri-html-ext takes advantage of [the `Nokogiri::XML::Document.parse` method](https://github.com/sparklemotion/nokogiri/blob/main/lib/nokogiri/xml/document.rb#L48)'s second positional argument to set the parsed document's URL.Nokogiri's source code is _very_ complex, but in short: [the `Nokogiri::HTML` method](https://github.com/sparklemotion/nokogiri/blob/main/lib/nokogiri/html.rb#L7-L8) is an alias to [the `Nokogiri::HTML4` method](https://github.com/sparklemotion/nokogiri/blob/main/lib/nokogiri/html4.rb#L10-L12) which eventually winds its way to the aforementioned `Nokogiri::XML::Document.parse` method. _Phew._ 🥵
+nokogiri-html-ext takes advantage of [the `Nokogiri::XML::Document.parse` method](https://github.com/sparklemotion/nokogiri/blob/main/lib/nokogiri/xml/document.rb#L48)'s second positional argument to set the parsed document's URL. Nokogiri's source code is _very_ complex, but in short: [the `Nokogiri::HTML` method](https://github.com/sparklemotion/nokogiri/blob/main/lib/nokogiri/html.rb#L7-L8) is an alias to [the `Nokogiri::HTML4` method](https://github.com/sparklemotion/nokogiri/blob/main/lib/nokogiri/html4.rb#L6-L9) which eventually winds its way to the aforementioned `Nokogiri::XML::Document.parse` method. _Phew._ 🥵
 
-URL resolution uses Ruby's built-in URL parsing and normalizing capabilities. Absolute URLs will remain unmodified.
+URL resolution uses [the Addressable gem](https://rubygems.org/gems/addressable)'s URL parsing and normalizing capabilities. Absolute URLs will remain unmodified.
 
-**Note:** If the document's markup includes a `<base>` element whose `href` attribute is an absolute URL, _that_ URL will take precedence when performing URL resolution.
+> [!NOTE]
+If the document's markup includes a `<base>` element whose `href` attribute is an absolute URL, _that_ URL will take precedence when performing URL resolution.
 
 An abbreviated example:
 
@@ -133,6 +134,9 @@ doc.resolve_relative_url("biz/baz")
 #=> "https://jgarber.example/foo/biz/baz"
 ```
 
+> [!NOTE]
+> Nokogiri's default `Nokogiri::HTML` method returns a `Nokogiri::HTML4::Document` which will encode URLs with non-ASCII characters. For example, `☠️.example` will be encoded as `%25E2%2598%25A0%25EF%25B8%258F.example`. For a more consistence experience, use the `Nokogiri::HTML5` method which does not encode URLs in this manner.
+
 ## Acknowledgments
 
 nokogiri-html-ext wouldn't exist without the [Nokogiri](https://nokogiri.org) project and its [community](https://github.com/sparklemotion/nokogiri).

data/lib/nokogiri/html-ext.rb

@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 
+require "addressable/uri"
 require "nokogiri"
-require "uri"
 
 require_relative "html_ext/document"

data/lib/nokogiri/html_ext/document.rb

@@ -28,11 +28,9 @@ module Nokogiri
 
       ATTRIBUTES_XPATHS =
         URL_ATTRIBUTES_MAP.merge(SRCSET_ATTRIBUTES_MAP).flat_map do |attribute, names|
-          names.map { |name| "//#{name} / @#{attribute}" }
+          names.map { |name| ".//#{name} / @#{attribute}" }
         end
 
-      URI_PARSER = defined?(URI::RFC2396_PARSER) ? URI::RFC2396_PARSER : URI::Generic::DEFAULT_PARSER
-
       # Get the +<base>+ element's HREF attribute value.
       #
       # @return [String, nil]
@@ -64,15 +62,10 @@ module Nokogiri
       #
       # @return [String]
       def resolve_relative_url(url)
-        strs = [doc_url_str, base_href, url]
-
-        strs.compact!
-        strs.map! { |str| URI_PARSER.escape(str) }
+        strs = [document.url, base_href, url].compact
 
-        # Escape each component before joining (Ruby's +URI.parse+ only likes
-        # ASCII) and subsequently unescaping.
-        URI_PARSER.unescape(URI_PARSER.join(*strs).normalize.to_s)
-      rescue URI::InvalidComponentError, URI::InvalidURIError
+        Addressable::URI.join(*strs).to_s
+      rescue Addressable::URI::InvalidURIError
         url
       end
 
@@ -96,15 +89,12 @@ module Nokogiri
 
       private
 
-      # +Nokogiri::HTML4::Document#url+ may be double-escaped if the parser
-      # detects non-ASCII characters. For example, +https://[skull emoji].example+
-      # is returned as +"https%3A//%25E2%2598%25A0%25EF%25B8%258F.example+.
+      # Resolve a set of +String+s that represent +srcset+ attribute image
+      # candidate strings.
+      #
+      # @param srcset_attributes [Array<String>]
       #
       # @return [String]
-      def doc_url_str
-        @doc_url_str ||= URI_PARSER.unescape(URI_PARSER.unescape(document.url)).strip
-      end
-
       def resolve_srcset_attributes(srcset_attributes)
         srcset_attributes.map! do |candidate_string|
           # rubocop:disable Style/PerlBackrefs

data/nokogiri-html-ext.gemspec

@@ -4,7 +4,7 @@ Gem::Specification.new do |spec|
   spec.required_ruby_version = ">= 2.7"
 
   spec.name          = "nokogiri-html-ext"
-  spec.version       = "1.3.0"
+  spec.version       = "1.4.0"
   spec.authors       = ["Jason Garber"]
   spec.email         = ["jason@sixtwothree.org"]
 
@@ -28,5 +28,6 @@ Gem::Specification.new do |spec|
     "source_code_uri" => "#{spec.homepage}/src/tag/v#{spec.version}",
   }
 
+  spec.add_dependency "addressable", "~> 2.8.7"
   spec.add_dependency "nokogiri", "~> 1.14"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nokogiri-html-ext
 version: !ruby/object:Gem::Version
-  version: 1.3.0
+  version: 1.4.0
 platform: ruby
 authors:
 - Jason Garber
@@ -9,6 +9,20 @@ bindir: bin
 cert_chain: []
 date: 1980-01-02 00:00:00.000000000 Z
 dependencies:
+- !ruby/object:Gem::Dependency
+  name: addressable
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.8.7
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.8.7
 - !ruby/object:Gem::Dependency
   name: nokogiri
   requirement: !ruby/object:Gem::Requirement
@@ -41,11 +55,11 @@ licenses:
 - MIT
 metadata:
   bug_tracker_uri: https://codeberg.org/jgarber/nokogiri-html-ext/issues
-  changelog_uri: https://codeberg.org/jgarber/nokogiri-html-ext/releases/tag/v1.3.0
-  documentation_uri: https://rubydoc.info/gems/nokogiri-html-ext/1.3.0
+  changelog_uri: https://codeberg.org/jgarber/nokogiri-html-ext/releases/tag/v1.4.0
+  documentation_uri: https://rubydoc.info/gems/nokogiri-html-ext/1.4.0
   homepage_uri: https://codeberg.org/jgarber/nokogiri-html-ext
   rubygems_mfa_required: 'true'
-  source_code_uri: https://codeberg.org/jgarber/nokogiri-html-ext/src/tag/v1.3.0
+  source_code_uri: https://codeberg.org/jgarber/nokogiri-html-ext/src/tag/v1.4.0
 rdoc_options: []
 require_paths:
 - lib