fast_jsonparser 0.4.0 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 2b9a3639a83aa2f68468df13dcc8922b244c727c00b2e040a7dd94f9b3832698
-  data.tar.gz: 49e6c154239a25e2ecc695288166914ee60a1500d7e2773ef25d05e39eddfa4d
+  metadata.gz: a696c8dcde3ac2ec599ab0d45bc3ef0f13da2823cbddacd13842337e50cd6200
+  data.tar.gz: 8be657884315230f3010b4713c9646150f95db5468f3fbc12124406df51b69a4
 SHA512:
-  metadata.gz: '08ebbd273a4180261ab78c50874e7a5cf28afa7c1718be05fbc8bec93bd6652097c88ad2b0ac7eba184f4755a209cc0e9256551d04e1350ac1139d4141a50412'
-  data.tar.gz: 6400ba1f5f0f0083b4c2a75f947888ab782695c66c676a1bb35011efde40b7ce2b1ab38d0fbbfc76eda50e80cef6194cc8bf495e3f8cb6b1aef0c8855bf2ed76
+  metadata.gz: 744771f5df6f8259ad6e75d631f8af6ba59ab6d00d94df78abbc0341d995045ba6bf5a32cffc1f476a9b48a5bce45d803c8f91c13c1fe840a2c6f931c8e549be
+  data.tar.gz: d14e73c088cd61d073a03d787075ea461750d7df3e1628c28398bc4f51ada3da4e6c4d036af2b9f81f16ede1ca28948fcd0fe49966b4756ee9c9f72ab63af81c

data/.github/workflows/ruby.yml

@@ -0,0 +1,35 @@
+# This workflow uses actions that are not certified by GitHub.
+# They are provided by a third-party and are governed by
+# separate terms of service, privacy policy, and support
+# documentation.
+# This workflow will download a prebuilt Ruby version, install dependencies and run tests with Rake
+# For more information see: https://github.com/marketplace/actions/setup-ruby-jruby-and-truffleruby
+
+name: Ruby
+
+on:
+  push:
+    branches: [ master ]
+  pull_request:
+    branches: [ master ]
+
+jobs:
+  test:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v2
+    - name: Set up Ruby
+    # To automatically get bug fixes and new Ruby versions for ruby/setup-ruby,
+    # change this to (see https://github.com/ruby/setup-ruby#versioning):
+    # uses: ruby/setup-ruby@v1
+      uses: ruby/setup-ruby@ec106b438a1ff6ff109590de34ddc62c540232e0
+      with:
+        ruby-version: 2.6
+    - name: Install dependencies
+      run: bundle install
+    - name: Compile
+      run: bundle exec rake compile
+    - name: Run tests
+      run: bundle exec rake

data/CHANGELOG.md

@@ -1,3 +1,6 @@
+# 0.5.0
+* Handle concurrent use of the parser in [Issue #15](https://github.com/anilmaurya/fast_jsonparser/pull/15), thanks to [casperisfine](https://github.com/casperisfine)
+
 # 0.4.0
 * load_many accept batch_size parameter to parse documents larger than 1 MB in [PR #5](https://github.com/anilmaurya/fast_jsonparser/pull/5), thanks to [casperisfine](https://github.com/casperisfine)
 * Add option for symbolize_keys, default to true in [PR #9](https://github.com/anilmaurya/fast_jsonparser/pull/9), thanks to [casperisfine](https://github.com/casperisfine)

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    fast_jsonparser (0.3.0)
+    fast_jsonparser (0.4.0)
 
 GEM
   remote: https://rubygems.org/

data/README.md

@@ -128,6 +128,15 @@ FastJsonparser.parse('{"one": 1, "two": 2}', symbolize_keys: false)
 FastJsonparser.parse("123: 1") # FastJsonparser::ParseError (parse error)
 ```
 
+### Known Incompatibilities with stdlib JSON
+
+`FastJsonparser` behaves mostly like stdlib's `JSON`, but there are a few corner cases:
+
+- `FastJsonparser` will use symbols for hash keys by default. You can pass `symbolize_names: false` to have strings instead like `JSON`.
+- `FastJsonparser` will raise on integers outside of the 64bits range (`-9223372036854775808..18446744073709551615`), `JSON` will parse them fine.
+- `FastJsonparser` will raise on invalid string escapings (`"\x15"`), `JSON` will often handle some of them.
+- `FastJsonparser` will raise on `/**/` comments. `JSON` will sometimes ignore them, sometimes raise.
+
 ### Example
 
 ```

data/ext/fast_jsonparser/fast_jsonparser.cpp

@@ -6,6 +6,38 @@ VALUE rb_eFastJsonparserUnknownError, rb_eFastJsonparserParseError;
 
 using namespace simdjson;
 
+typedef struct {
+    dom::parser *parser;
+} parser_t;
+
+static void Parser_delete(void *ptr) {
+    parser_t *data = (parser_t*) ptr;
+    delete data->parser;
+}
+
+static size_t Parser_memsize(const void *parser) {
+    return sizeof(dom::parser); // TODO: low priority, figure the real size, e.g. internal buffers etc.
+}
+
+static const rb_data_type_t parser_data_type = {
+    "Parser",
+    { 0, Parser_delete, Parser_memsize, },
+    0, 0, RUBY_TYPED_FREE_IMMEDIATELY
+};
+
+static VALUE parser_allocate(VALUE klass) {
+    parser_t *data;
+    VALUE obj = TypedData_Make_Struct(klass, parser_t, &parser_data_type, data);
+    data->parser = new dom::parser;
+    return obj;
+}
+
+static inline dom::parser * get_parser(VALUE self) {
+    parser_t *data;
+    TypedData_Get_Struct(self, parser_t, &parser_data_type, data);
+    return data->parser;
+}
+
 // Convert tape to Ruby's Object
 static VALUE make_ruby_object(dom::element element, bool symbolize_keys)
 {
@@ -70,9 +102,9 @@ static VALUE make_ruby_object(dom::element element, bool symbolize_keys)
 static VALUE rb_fast_jsonparser_parse(VALUE self, VALUE arg, VALUE symbolize_keys)
 {
     Check_Type(arg, T_STRING);
+    dom::parser *parser = get_parser(self);
 
-    dom::parser parser;
-    auto [doc, error] = parser.parse(RSTRING_PTR(arg), RSTRING_LEN(arg));
+    auto [doc, error] = parser->parse(RSTRING_PTR(arg), RSTRING_LEN(arg));
     if (error != SUCCESS)
     {
         rb_raise(rb_eFastJsonparserParseError, "%s", error_message(error));
@@ -83,9 +115,9 @@ static VALUE rb_fast_jsonparser_parse(VALUE self, VALUE arg, VALUE symbolize_key
 static VALUE rb_fast_jsonparser_load(VALUE self, VALUE arg, VALUE symbolize_keys)
 {
     Check_Type(arg, T_STRING);
+    dom::parser *parser = get_parser(self);
 
-    dom::parser parser;
-    auto [doc, error] = parser.load(RSTRING_PTR(arg));
+    auto [doc, error] = parser->load(RSTRING_PTR(arg));
     if (error != SUCCESS)
     {
         rb_raise(rb_eFastJsonparserParseError, "%s", error_message(error));
@@ -97,11 +129,10 @@ static VALUE rb_fast_jsonparser_load_many(VALUE self, VALUE arg, VALUE symbolize
 {
     Check_Type(arg, T_STRING);
     Check_Type(batch_size, T_FIXNUM);
+    dom::parser *parser = get_parser(self);
 
-    try
-    {
-        dom::parser parser;
-        auto [docs, error] = parser.load_many(RSTRING_PTR(arg), FIX2INT(batch_size));
+    try {
+        auto [docs, error] = parser->load_many(RSTRING_PTR(arg), FIX2INT(batch_size));
         if (error != SUCCESS)
         {
             rb_raise(rb_eFastJsonparserParseError, "%s", error_message(error));
@@ -126,10 +157,12 @@ extern "C"
     void Init_fast_jsonparser(void)
     {
         VALUE rb_mFastJsonparser = rb_const_get(rb_cObject, rb_intern("FastJsonparser"));
+        VALUE rb_cFastJsonparserNative = rb_const_get(rb_mFastJsonparser, rb_intern("Native"));
 
-        rb_define_module_function(rb_mFastJsonparser, "_parse", reinterpret_cast<VALUE (*)(...)>(rb_fast_jsonparser_parse), 2);
-        rb_define_module_function(rb_mFastJsonparser, "_load", reinterpret_cast<VALUE (*)(...)>(rb_fast_jsonparser_load), 2);
-        rb_define_module_function(rb_mFastJsonparser, "_load_many", reinterpret_cast<VALUE (*)(...)>(rb_fast_jsonparser_load_many), 3);
+        rb_define_alloc_func(rb_cFastJsonparserNative, parser_allocate);
+        rb_define_method(rb_cFastJsonparserNative, "_parse", reinterpret_cast<VALUE (*)(...)>(rb_fast_jsonparser_parse), 2);
+        rb_define_method(rb_cFastJsonparserNative, "_load", reinterpret_cast<VALUE (*)(...)>(rb_fast_jsonparser_load), 2);
+        rb_define_method(rb_cFastJsonparserNative, "_load_many", reinterpret_cast<VALUE (*)(...)>(rb_fast_jsonparser_load_many), 3);
 
         rb_eFastJsonparserParseError = rb_const_get(rb_mFastJsonparser, rb_intern("ParseError"));
         rb_global_variable(&rb_eFastJsonparserParseError);

data/ext/fast_jsonparser/simdjson.h

@@ -2308,7 +2308,7 @@ using ErrorValues [[deprecated("This is an alias and will be removed, use error_
  * @deprecated Error codes should be stored and returned as `error_code`, use `error_message()` instead.
  */
 [[deprecated("Error codes should be stored and returned as `error_code`, use `error_message()` instead.")]]
-inline const std::string &error_message(int error) noexcept;
+inline const std::string error_message(int error) noexcept;
 
 } // namespace simdjson
 
@@ -6367,7 +6367,7 @@ namespace internal {
   // We store the error code so we can validate the error message is associated with the right code
   struct error_code_info {
     error_code code;
-    std::string message;
+    const char* message;
   };
   // These MUST match the codes in error_code. We check this constraint in basictests.
   extern SIMDJSON_DLLIMPORTEXPORT const error_code_info error_codes[];
@@ -6376,10 +6376,10 @@ namespace internal {
 
 inline const char *error_message(error_code error) noexcept {
   // If you're using error_code, we're trusting you got it from the enum.
-  return internal::error_codes[int(error)].message.c_str();
+  return internal::error_codes[int(error)].message;
 }
 
-inline const std::string &error_message(int error) noexcept {
+inline const std::string error_message(int error) noexcept {
   if (error < 0 || error >= error_code::NUM_ERROR_CODES) {
     return internal::error_codes[UNEXPECTED_ERROR].message;
   }

data/lib/fast_jsonparser.rb

@@ -12,15 +12,15 @@ module FastJsonparser
 
   class << self
     def parse(source, symbolize_keys: true)
-      _parse(source, symbolize_keys)
+      parser._parse(source, symbolize_keys)
     end
 
     def load(source, symbolize_keys: true)
-      _load(source, symbolize_keys)
+      parser._load(source, symbolize_keys)
     end
 
     def load_many(source, symbolize_keys: true, batch_size: DEFAULT_BATCH_SIZE, &block)
-      _load_many(source, symbolize_keys, batch_size, &block)
+      Native.new._load_many(source, symbolize_keys, batch_size, &block)
     rescue UnknownError => error
       case error.message
       when "This parser can't support a document that big"
@@ -30,7 +30,17 @@ module FastJsonparser
       end
     end
 
-    require "fast_jsonparser/fast_jsonparser" # loads cpp extension
-    private :_parse, :_load, :_load_many
+    private
+
+    def parser
+      @parser ||= Native.new
+    end
+  end
+
+  class Native
   end
+
+  require "fast_jsonparser/fast_jsonparser" # loads cpp extension
+
+  private_constant :Native
 end

data/lib/fast_jsonparser/version.rb

@@ -1,3 +1,3 @@
 module FastJsonparser
-  VERSION = "0.4.0"
+  VERSION = "0.5.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: fast_jsonparser
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.5.0
 platform: ruby
 authors:
 - Anil Maurya
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2020-07-22 00:00:00.000000000 Z
+date: 2020-08-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -102,6 +102,7 @@ extensions:
 - ext/fast_jsonparser/extconf.rb
 extra_rdoc_files: []
 files:
+- ".github/workflows/ruby.yml"
 - ".gitignore"
 - ".travis.yml"
 - CHANGELOG.md