nokogumbo 1.4.9 → 1.4.10

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: ecb515ec9c492a01dce23eed6eb7336695f166d6
-  data.tar.gz: c1ac4272ee113e56ecd72302847886042165a1fb
+  metadata.gz: c97bf9e43a0be85f20cf07ec15901052a1f9dbe1
+  data.tar.gz: 4c195ae6b3fc161648b7fc0c7eea06d14e8bd451
 SHA512:
-  metadata.gz: b979502577d944afa23cf0db4307494c6e53c320c3efc87c8066174d8ae3eee8902f60164375869dd6bd154abb1f6f69c42db9fbb0bc7fbd86d1b2f3e33bafd3
-  data.tar.gz: abd153cd24e504dc82c80d38ad91c8dd3dae25e913d650d0c92a8823cb1638dcf99baa43eb40a6a36573f9de7d14d78e2d77921062e341f0e82d49290a2f7ad7
+  metadata.gz: f49c427bdb96bce746e541c7864c755b0ece284f0f7fff8d0f3913403b8c9e9941b1d9c06ddec25142be4ee7ead2c97e96b9eac26eeee119cbb6cfc1879d5f7b
+  data.tar.gz: bf2b358e4f496f5f8e1c3b308e8c7befc308b9dda1af7399d4baa73cc59a395f630d315a5791565a088d7dfdf9c9a0eaaea5638798da6a2dfc6cea559df1fa25

data/ext/nokogumboc/nokogumbo.c

@@ -19,10 +19,13 @@
 //
 
 #include <ruby.h>
-#include <gumbo.h>
+#include "gumbo.h"
+#include "error.h"
+#include "parser.h"
 
 // class constants
 static VALUE Document;
+static VALUE XMLSyntaxError;
 
 #ifdef NGLIB
 #include <nokogiri.h>
@@ -182,10 +185,10 @@ static xmlNodePtr walk_tree(xmlDocPtr document, GumboNode *node) {
 
 // Parse a string using gumbo_parse into a Nokogiri document
 static VALUE parse(VALUE self, VALUE string) {
-  GumboOutput *output = gumbo_parse_with_options(
-    &kGumboDefaultOptions, RSTRING_PTR(string),
-    (size_t) RSTRING_LEN(string)
-  );
+  const GumboOptions *options = &kGumboDefaultOptions;
+  const char *input = RSTRING_PTR(string);
+  size_t input_len = RSTRING_LEN(string);
+  GumboOutput *output = gumbo_parse_with_options(options, input, input_len);
   xmlDocPtr doc = xmlNewDoc(CONST_CAST "1.0");
 #ifdef NGLIB
   doc->type = XML_HTML_DOCUMENT_NODE;
@@ -210,9 +213,42 @@ static VALUE parse(VALUE self, VALUE string) {
         xmlAddChild((xmlNodePtr)doc, node);
     }
   }
-  gumbo_destroy_output(&kGumboDefaultOptions, output);
 
-  return Nokogiri_wrap_xml_document(Document, doc);
+  VALUE rdoc = Nokogiri_wrap_xml_document(Document, doc);
+
+  // Add parse errors to rdoc.
+  if (output->errors.length) {
+    GumboVector *errors = &output->errors;
+    GumboParser parser = { ._options = options };
+    GumboStringBuffer msg;
+    VALUE rerrors = rb_ary_new2(errors->length);
+
+    gumbo_string_buffer_init(&parser, &msg);
+    for (int i=0; i < errors->length; i++) {
+      GumboError *err = errors->data[i];
+      gumbo_string_buffer_clear(&parser, &msg);
+      gumbo_caret_diagnostic_to_string(&parser, err, input, &msg);
+      VALUE err_str = rb_str_new(msg.data, msg.length);
+      VALUE syntax_error = rb_class_new_instance(1, &err_str, XMLSyntaxError);
+      rb_iv_set(syntax_error, "@domain", INT2NUM(1)); // XML_FROM_PARSER
+      rb_iv_set(syntax_error, "@code", INT2NUM(1));   // XML_ERR_INTERNAL_ERROR
+      rb_iv_set(syntax_error, "@level", INT2NUM(2));  // XML_ERR_ERROR
+      rb_iv_set(syntax_error, "@file", Qnil);
+      rb_iv_set(syntax_error, "@line", INT2NUM(err->position.line));
+      rb_iv_set(syntax_error, "@str1", Qnil);
+      rb_iv_set(syntax_error, "@str2", Qnil);
+      rb_iv_set(syntax_error, "@str3", Qnil);
+      rb_iv_set(syntax_error, "@int1", INT2NUM(err->type));
+      rb_iv_set(syntax_error, "@column", INT2NUM(err->position.column));
+      rb_ary_push(rerrors, syntax_error);
+    }
+    rb_iv_set(rdoc, "@errors", rerrors);
+    gumbo_string_buffer_destroy(&parser, &msg);
+  }
+
+  gumbo_destroy_output(options, output);
+
+  return rdoc;
 }
 
 // Initialize the Nokogumbo class and fetch constants we will use later
@@ -224,10 +260,11 @@ void Init_nokogumboc() {
   VALUE Nokogiri = rb_const_get(rb_cObject, rb_intern("Nokogiri"));
   VALUE HTML = rb_const_get(Nokogiri, rb_intern("HTML"));
   Document = rb_const_get(HTML, rb_intern("Document"));
+  VALUE XML = rb_const_get(Nokogiri, rb_intern("XML"));
+  XMLSyntaxError = rb_const_get(XML, rb_intern("SyntaxError"));
 
 #ifndef NGLIB
   // more class constants
-  VALUE XML = rb_const_get(Nokogiri, rb_intern("XML"));
   Element = rb_const_get(XML, rb_intern("Element"));
   Text = rb_const_get(XML, rb_intern("Text"));
   CDATA = rb_const_get(XML, rb_intern("CDATA"));

data/test-nokogumbo.rb

@@ -125,6 +125,13 @@ class TestNokogumbo < Minitest::Test
     assert_equal ["html", "comment", "html", "comment"], doc.children.map(&:name)
   end
 
+  def test_parse_errors
+    doc = Nokogiri::HTML5("<!DOCTYPE html><html><!-- -- --></a>")
+    assert_equal doc.errors.length, 2
+    doc = Nokogiri::HTML5("<!DOCTYPE html><html>")
+    assert_empty doc.errors
+  end
+
 private
 
   def buffer

metadata

@@ -1,27 +1,27 @@
 --- !ruby/object:Gem::Specification
 name: nokogumbo
 version: !ruby/object:Gem::Version
-  version: 1.4.9
+  version: 1.4.10
 platform: ruby
 authors:
 - Sam Ruby
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-08-01 00:00:00.000000000 Z
+date: 2016-11-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
 description: Nokogumbo allows a Ruby program to invoke the Gumbo HTML5 parser and
@@ -32,11 +32,10 @@ extensions:
 - ext/nokogumboc/extconf.rb
 extra_rdoc_files: []
 files:
-- ext/nokogumboc/extconf.rb
-- ext/nokogumboc/nokogumbo.c
-- lib/nokogumbo.rb
 - LICENSE.txt
 - README.md
+- ext/nokogumboc/extconf.rb
+- ext/nokogumboc/nokogumbo.c
 - gumbo-parser/src/attribute.c
 - gumbo-parser/src/attribute.h
 - gumbo-parser/src/char_ref.c
@@ -69,10 +68,11 @@ files:
 - gumbo-parser/src/vector.c
 - gumbo-parser/src/vector.h
 - gumbo-parser/visualc/include/strings.h
+- lib/nokogumbo.rb
 - test-nokogumbo.rb
 homepage: https://github.com/rubys/nokogumbo/#readme
 licenses:
-- Apache 2.0
+- Apache-2.0
 metadata: {}
 post_install_message: 
 rdoc_options: []
@@ -80,17 +80,17 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.14.1
+rubygems_version: 2.5.1
 signing_key: 
 specification_version: 4
 summary: Nokogiri interface to the Gumbo HTML5 parser