nokolexbor 0.6.0 → 0.6.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: eddfccec75e5089344c81814159f243d5c46e1d747d7cb2351a8efcdc554e46e
-  data.tar.gz: 1112c8ee0d902e2ef382f582a39e306a5a908ca03bb5b5bc6837e361663552fe
+  metadata.gz: 5095dc8ed38a170bab48cc78e7a7e4475657b3066cb114a46eab1d9f4ed25c36
+  data.tar.gz: c2a00a4ebc644dda81e0b010166063a5ad975ad7e2410739c69af2246cce560f
 SHA512:
-  metadata.gz: 66ad9e8663b0f072a308b01e98314622fc2b920d0c3523c8fb8faa81555bf86cb45bd2c7dc230aeaf6333c6964c59d508eb97162ab88db4130eab07412297db4
-  data.tar.gz: 60255a5d4f8beefa0b51e37480138f6c52fa0fa4f414585409f3a71cfb00ee489a734ed93a6d34689c9601d70b1101af2ab831e4c6c207e069a441c66491d668
+  metadata.gz: 07df8df17e08dbe9293b6f0e1f878ee70d787f3370c33a00bcf0ca71dee007a708ecf4a7b069039cb05d95e93cd741654a3bdb8a11833804f2b9f63dd5d9fb4b
+  data.tar.gz: 31addbc3877af1b1d814cb720c17e8130d98280d35831794e135e259b790de0c719236004c202649092366323a7e82a9175003be0116ac0267639d15fd186081

data/ext/nokolexbor/extconf.rb

@@ -24,7 +24,20 @@ def which(cmd)
   return nil
 end
 
+if !find_executable('cmake')
+  abort "ERROR: CMake is required to build Lexbor."
+end
+
 cmake_flags = [ ENV["CMAKE_FLAGS"] ]
+
+cmake_version_full = `cmake --version`
+unless cmake_version_full.empty?
+  cmake_version = cmake_version_full[/cmake version ([0-9\.]+)/, 1]
+  if Gem::Version.new(cmake_version) >= Gem::Version.new('4.0')
+    cmake_flags << "-DCMAKE_POLICY_VERSION_MINIMUM=3.5"
+  end
+end
+
 cmake_flags << "-DCMAKE_TRY_COMPILE_TARGET_TYPE=STATIC_LIBRARY"
 # Set system name explicitly when cross-compiling
 cmake_flags << "-DCMAKE_SYSTEM_NAME=Windows -DWIN32=1" if windows?
@@ -116,10 +129,6 @@ LEXBOR_DIR = File.join(CWD, '..', '..', 'vendor', 'lexbor')
 EXT_DIR = File.join(CWD, '..', '..', 'ext', 'nokolexbor')
 INSTALL_DIR = File.join(LEXBOR_DIR, 'dist')
 
-if !find_executable('cmake')
-  abort "ERROR: CMake is required to build Lexbor."
-end
-
 Dir.chdir(LEXBOR_DIR) do
   # Patch lexbor
   Dir['../../patches/*-lexbor-*.patch'].each do |patch_file|

data/ext/nokolexbor/nl_document.c

@@ -22,26 +22,9 @@ const rb_data_type_t nl_document_type = {
     RUBY_TYPED_FREE_IMMEDIATELY,
 };
 
-/**
- * call-seq:
- *   parse(string_or_io) -> Document
- *
- * Parse HTML into a {Document}.
- *
- * @param string_or_io [String, #read]
- *   The HTML to be parsed. It may be a String, or any object that
- *   responds to #read such as an IO, or StringIO.
- */
 static VALUE
-nl_document_parse(VALUE self, VALUE rb_string_or_io)
+nl_document_parse_native(VALUE self, VALUE rb_html)
 {
-  VALUE id_read = rb_intern("read");
-  VALUE rb_html;
-  if (rb_respond_to(rb_string_or_io, id_read)) {
-    rb_html = rb_funcall(rb_string_or_io, id_read, 0);
-  } else {
-    rb_html = rb_string_or_io;
-  }
   const char *html_c = StringValuePtr(rb_html);
   size_t html_len = RSTRING_LEN(rb_html);
 
@@ -74,7 +57,7 @@ nl_document_parse(VALUE self, VALUE rb_string_or_io)
 static VALUE
 nl_document_new(VALUE self)
 {
-  return nl_document_parse(self, rb_str_new("", 0));
+  return nl_document_parse_native(self, rb_str_new("", 0));
 }
 
 lxb_dom_document_t *
@@ -136,7 +119,7 @@ void Init_nl_document(void)
 {
   cNokolexborDocument = rb_define_class_under(mNokolexbor, "Document", cNokolexborNode);
   rb_define_singleton_method(cNokolexborDocument, "new", nl_document_new, 0);
-  rb_define_singleton_method(cNokolexborDocument, "parse", nl_document_parse, 1);
+  rb_define_singleton_method(cNokolexborDocument, "parse_native", nl_document_parse_native, 1);
   rb_define_method(cNokolexborDocument, "title", nl_document_get_title, 0);
   rb_define_method(cNokolexborDocument, "title=", nl_document_set_title, 1);
   rb_define_method(cNokolexborDocument, "root", nl_document_root, 0);

data/lib/nokolexbor/document.rb

@@ -141,6 +141,26 @@ module Nokolexbor
       end
     end
 
+    # Parse HTML into a {Document}.
+    #
+    # @param string_or_io [String, #read]
+    #   The HTML to be parsed. It may be a String, or any object that
+    #   responds to #read such as an IO, or StringIO.
+    #
+    # @return [Document]
+    def self.parse(string_or_io)
+      html = string_or_io
+      if string_or_io.respond_to?(:read)
+        html = string_or_io.read
+      end
+
+      if html.respond_to?(:encoding) && html.encoding != Encoding::UTF_8
+        html = html.encode(Encoding::UTF_8, invalid: :replace, undef: :replace)
+      end
+
+      parse_native(html)
+    end
+
     private
 
     IMPLIED_XPATH_CONTEXTS = ["//"].freeze # :nodoc:

data/lib/nokolexbor/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Nokolexbor
-  VERSION = '0.6.0'
+  VERSION = '0.6.2'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: nokolexbor
 version: !ruby/object:Gem::Version
-  version: 0.6.0
+  version: 0.6.2
 platform: ruby
 authors:
 - Yicheng Zhou
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-12-31 00:00:00.000000000 Z
+date: 2025-05-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake-compiler
@@ -523,7 +523,7 @@ licenses:
 - MIT
 metadata:
   msys2_mingw_dependencies: cmake
-post_install_message: 
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -539,7 +539,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubygems_version: 3.0.3.1
-signing_key: 
+signing_key:
 specification_version: 4
 summary: High-performance HTML5 parser, with support for both CSS selectors and XPath.
 test_files: []