java_bin 0.2.0 → 0.3.0

data/.gitignore

@@ -24,3 +24,8 @@ Makefile
 *.so
 *.log
 *.orig
+*.def
+*.exp
+*.lib
+*.pdb
+*.obj

data/CHANGELOG

@@ -1,3 +1,9 @@
+0.3.0 (2010/01/19)
+  * windows build support
+
+0.2.0 (2010/01/14)
+  * add rsolr/solr-ruby support
+
 0.1.2 (2010/01/12)
   * fix memory leak error (extern string).
 

data/README.rdoc

@@ -1,31 +1,42 @@
 = java_bin
 
-This is a Apache Solr JavaBin format (binary format) implementation for Ruby.
+This is an Apache Solr JavaBin format (binary format) implementation for Ruby.
 
 == Features
 
 * fast parse, and less network traffic.
  MRI 1.8.7
   [data1]
-  ruby eval parse. 5000 times. elapsed time 1.31124
-  javabin parse.   5000 times. elapsed time 0.514837
-  2.54690319460334 times
+  ruby eval parse. 5000 times. elapsed time 1.282744
+  javabin parse.   5000 times. elapsed time 0.4634
+  2.76811394044022 times
 
   [data2]  
-  ruby2 eval parse. 5000 times. elapsed time 6.604737
-  javabin2 parse.   5000 times. elapsed time 2.411031
-  2.73938286152273 times
+  ruby2 eval parse. 5000 times. elapsed time 6.568942
+  javabin2 parse.   5000 times. elapsed time 2.267351
+  2.89718795193157 times
+
+ REE 1.8.7
+  [data1]
+  ruby eval parse. 5000 times. elapsed time 0.940934
+  javabin parse.   5000 times. elapsed time 0.310581
+  3.02959292422911 times
+
+  [data2]
+  ruby2 eval parse. 5000 times. elapsed time 4.708985
+  javabin2 parse.   5000 times. elapsed time 1.482276
+  3.176861124379 times
 
  YARV 1.9.2
   [data1]
-  ruby eval parse. 5000 times. elapsed time 1.35476947
-  javabin parse.   5000 times. elapsed time 0.370740373
-  3.654226970311647 times
+  ruby eval parse. 5000 times. elapsed time 1.348342119
+  javabin parse.   5000 times. elapsed time 0.367371147
+  3.6702450097421506 times
 
   [data2]
-  ruby2 eval parse. 5000 times. elapsed time 6.555066861
-  javabin2 parse.   5000 times. elapsed time 1.752366331
-  3.7406943656919647 times
+  ruby2 eval parse. 5000 times. elapsed time 6.486995777
+  javabin2 parse.   5000 times. elapsed time 1.73077786
+  3.748023317677521 times
 
 * rsolr/solr-ruby support.
 * pure and c extension code.
@@ -34,38 +45,49 @@ This is a Apache Solr JavaBin format (binary format) implementation for Ruby.
 == Requirements
 
 * Ruby1.8.7 or later (include 1.9.x)
-* (recommended) C compiler: you can alos use java_bin without c extension, but 'pure' is 30 times slower than 'ext'.
-* JavaBin has been tested with MRI 1.8.7, REE 1.8.7, YARV 1.9.2 pre1 on Ubuntu Linux 9.10 (32bit), and Apache Solr 1.4  
+* (recommended) C compiler: you can also use java_bin without c extension, but 'pure' is 30 times slower than 'ext'.
+* JavaBin has been tested with MRI 1.8.7, REE 1.8.7, YARV 1.9.2 pre1 on Ubuntu Linux 9.10 (32bit) and MRI 1.8.6 on Windows Vista Business (32bit), and Apache Solr 1.4  
 
 == Install
 
+ (Linux)
  gem sources -a http://gemcutter.org
  gem install java_bin
 
+ (Windows)
+ start => visual studio c++ 200x xxxx edition => visual studio tools => visual studio 200x command prompt
+ gem sources -a http://gemcutter.org
+ gem install java_bin
+   or
+ cd (path to)\ext\java_bin\ext
+ ruby extconf.rb
+ nmake
+
 == Simple Usage
 
  require 'rubygems'
  require 'java_bin'
 
  @parser = JavaBin.parser.new
- result = @parser.parse( java bin format string )
+ result = @parser.parse( javabin format string )
 
 === With RSolr/Solr-Ruby
 
 * By RSolr/Solr-Ruby support, a parser automatically uses javabin format instead of ruby format.
 
-  require 'rubygems'
-  require 'rsolr'
-  require 'java_bin'
-
-  solr = RSolr.connect
-  solr.select :q => '*:*'
+ require 'rubygems'
+ require 'rsolr'
+ require 'java_bin'
+ 
+ solr = RSolr.connect
+ solr.select :q => '*:*'
 
 == TODO
 
 * more parse speed
 * license
-* windows build
+* windows build(1.9.x)
+* 64bit build
 * builder(writer)作成
 * JRuby support
 

data/VERSION

@@ -1 +1 @@
-0.2.0
+0.3.0

data/ext/java_bin/ext/extconf.rb

@@ -2,24 +2,24 @@
 require 'mkmf'
 require 'rbconfig'
 
-unless $CFLAGS.gsub!(/ -O[\dsz]?/, ' -O3')
-  $CFLAGS << ' -O3'
-end
-if CONFIG['CC'] =~ /gcc/
-  $CFLAGS << ' -Wall'
-  #unless $CFLAGS.gsub!(/ -O[\dsz]?/, ' -O0 -ggdb')
-  #  $CFLAGS << ' -O0 -ggdb'
-  #end
+unless RUBY_PLATFORM =~ /mswin32/ # Linux
+  unless $CFLAGS.gsub!(/ -O[\dsz]?/, ' -O3')
+    $CFLAGS << ' -O3'
+  end
+  if CONFIG['CC'] =~ /gcc/
+    $CFLAGS << ' -Wall'
+  end
+else # Windows 
+  $CFLAGS.gsub!(/-O2b2xg-/, '/O2b2x')
+  $CFLAGS.gsub!(/-MD/, ' /MT')
+  $CFLAGS.gsub!(/ -G6/, '')
+  $CFLAGS << ' /wd4819' # VC++はUTF-8 BOM無しは駄目なのでここで抑制. gccは逆にBOM付は駄目
 end
 
-if RUBY_VERSION >= '1.9'
-  $CFLAGS << ' -DRUBY_19'
-end
+$CFLAGS << ' -DRUBY_19' if RUBY_VERSION >= '1.9'
 
-have_header("endian.h")
 have_header("byteswap.h")
 have_header("sys/types.h")
-
 have_header("ruby.h")
 have_header("ruby/encoding.h")
 

data/ext/java_bin/ext/parser.c

@@ -1,4 +1,3 @@
-/* utf-8 */
 #include "parser.h"
 
 /*
@@ -94,52 +93,73 @@ static VALUE JavaBinParser_read_string(JAVA_BIN_PARSER* ptr) {
 
 static VALUE JavaBinParser_read_byte(JAVA_BIN_PARSER* ptr) {
   int8_t c;
-  c = _readnumeric(ptr, c);
-  return INT2NUM(*((int8_t*)&c));
+  _readnumeric(ptr, c);
+  return INT2FIX(*((int8_t*)&c));
 }
 
 static VALUE JavaBinParser_read_short(JAVA_BIN_PARSER* ptr) {
   u_int16_t c;
-  c = _readnumeric(ptr, c);
+  _readnumeric(ptr, c);
   c = _swap_16(c);
-  return INT2NUM(*((int16_t*)&c));
+  return INT2FIX(*((int16_t*)&c));
 }
 
 static VALUE JavaBinParser_read_int(JAVA_BIN_PARSER* ptr) {
   u_int32_t c;
-  c = _readnumeric(ptr, c);
+  _readnumeric(ptr, c);
   c = _swap_32(c);
   return INT2NUM(*((int32_t*)&c));
 }
 
 static VALUE JavaBinParser_read_long(JAVA_BIN_PARSER* ptr) {
   u_int64_t c;
-  c = _readnumeric(ptr, c);
+  _readnumeric(ptr, c);
   c = _swap_64(c);
   return LL2NUM(*((int64_t*)&c));
 }
 
 static VALUE JavaBinParser_read_date(JAVA_BIN_PARSER* ptr) {
   u_int64_t c;
-  c = _readnumeric(ptr, c);
+  _readnumeric(ptr, c);
   c = _swap_64(c);
   return rb_funcall(rb_cTime, i_At, 1, ULL2NUM(*((int64_t*)&c) / 1000));
 }
 
 static VALUE JavaBinParser_read_float(JAVA_BIN_PARSER* ptr) {
   u_int32_t c;
-  c = _readnumeric(ptr, c);
+  _readnumeric(ptr, c);
   c = _swap_32(c);
   return rb_float_new((double)*((float*)&c));
 }
 
 static VALUE JavaBinParser_read_double(JAVA_BIN_PARSER* ptr) {
   u_int64_t c;
-  c = _readnumeric(ptr, c);
+  _readnumeric(ptr, c);
   c = _swap_64(c);
   return rb_float_new(*((double*)&c));
 }
 
+static void JavaBinParser_extend_cache(JAVA_BIN_PARSER* ptr) {
+  _EXTERN_STRING_INFO* newP;
+  int next_size;
+  if (ptr->cache == NULL) {
+    next_size = 64;
+  } else {
+    next_size = ptr->cache_size * 2;
+  }
+
+  newP = (_EXTERN_STRING_INFO*) malloc(next_size * sizeof(_EXTERN_STRING_INFO));
+  if (!newP) {
+    rb_raise(rb_eRuntimeError, "JavaBinParser_extend_cache - allocate error");
+  }
+
+  if (ptr->cache) {
+    memcpy(newP, ptr->cache, sizeof(_EXTERN_STRING_INFO) * ptr->cache_size);
+  }
+  ptr->cache = newP;
+  ptr->cache_size = next_size;
+}
+
 static VALUE JavaBinParser_read_val(JAVA_BIN_PARSER* ptr) {
   int size;
   int i;
@@ -272,27 +292,6 @@ static VALUE JavaBinParser_alloc(VALUE klass) {
   return Data_Wrap_Struct(klass, 0, JavaBinParser_free, NULL);
 }
 
-static void JavaBinParser_extend_cache(JAVA_BIN_PARSER* ptr) {
-  _EXTERN_STRING_INFO* newP;
-  int next_size;
-  if (ptr->cache == NULL) {
-    next_size = 64;
-  } else {
-    next_size = ptr->cache_size * 2;
-  }
-
-  newP = (_EXTERN_STRING_INFO*) malloc(next_size * sizeof(_EXTERN_STRING_INFO));
-  if (!newP) {
-    rb_raise(rb_eRuntimeError, "JavaBinParser_extend_cache - allocate error");
-  }
-
-  if (ptr->cache) {
-    memcpy(newP, ptr->cache, sizeof(_EXTERN_STRING_INFO) * ptr->cache_size);
-  }
-  ptr->cache = newP;
-  ptr->cache_size = next_size;
-}
-
 /*
  * rubyメソッド
  */

data/ext/java_bin/ext/parser.h

@@ -1,23 +1,18 @@
-/* utf-8 */
 #ifndef _PARSER_H_
 #define _PARSER_H_
 
 #include "ruby.h"
 
-#if HAVE_ENDIAN_H
-#include <endian.h>
-#endif /* HAVE_ENDIAN_H */
-
 #if HAVE_BYTESWAP_H
-#include <byteswap.h>
+  #include <byteswap.h>
 #endif /* HAVE_BYTESWAP_H */
 
 #if HAVE_SYS_TYPES_H
-#include <sys/types.h>
+  #include <sys/types.h>
 #endif /* HAVE_SYS_TYPES_H */
 
 #ifdef HAVE_RUBY_ENCODING_H
-#include <ruby/encoding.h>
+  #include <ruby/encoding.h>
 #endif /* HAVE_RUBY_ENCODING_H */
 
 /*
@@ -86,30 +81,47 @@ typedef struct java_bin_parser {
 } JAVA_BIN_PARSER;
 
 #ifdef HAVE_RUBY_ENCODING_H
-#define _utf8_string(str, len) (rb_enc_str_new(str, len, rb_encUtf8))
+  #define _utf8_string(str, len) (rb_enc_str_new(str, len, rb_encUtf8))
 #else
-#define _utf8_string(str, len) (rb_str_new(str, len))
+  #define _utf8_string(str, len) (rb_str_new(str, len))
 #endif
 
 #define _getbyte(ptr)          (ptr->data[ptr->current++])
 #define _skipbytes(ptr, x)     ((ptr->current) += (x))
 #define _readnumeric(ptr, c) \
-  ({ u_int8_t* p; \
+  { u_int8_t* p; \
      p = (void*)&c; \
      memcpy(p, &ptr->data[ptr->current], sizeof(c)); \
-     _skipbytes(ptr, sizeof(c)); c; \
-   })
-
-#if __BYTE_ORDER == __LITTLE_ENDIAN
-  #define _swap_16(o) bswap_16(o)
-  #define _swap_32(o) bswap_32(o)
-  #define _swap_64(o) bswap_64(o)
+     _skipbytes(ptr, sizeof(c)); \
+  }
+
+#ifndef WORDS_BIGENDIAN
+  #ifdef _WIN32
+    #define _swap_16(o) _byteswap_ushort(o)
+    #define _swap_32(o) _byteswap_ulong(o)
+    #define _swap_64(o) _byteswap_uint64(o)
+  #else
+    #define _swap_16(o) bswap_16(o)
+    #define _swap_32(o) bswap_32(o)
+    #define _swap_64(o) bswap_64(o)
+  #endif
 #else
   #define _swap_16(o) (o)
   #define _swap_32(o) (o)
   #define _swap_64(o) (o)
 #endif
 
+#ifdef _WIN32
+  typedef signed char        int8_t;
+  typedef signed short      int16_t;
+  typedef signed int        int32_t;
+  typedef signed __int64    int64_t;
+  typedef unsigned char     u_int8_t;
+  typedef unsigned short   u_int16_t;
+  typedef unsigned int     u_int32_t;
+  typedef unsigned __int64 u_int64_t;
+#endif
+
 static int32_t JavaBinParser_read_v_int(JAVA_BIN_PARSER* ptr);
 static int64_t JavaBinParser_read_v_long(JAVA_BIN_PARSER* ptr);
 static int JavaBinParser_read_size(JAVA_BIN_PARSER* ptr);

data/java_bin.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = %q{java_bin}
-  s.version = "0.2.0"
+  s.version = "0.3.0"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["kennyj"]
-  s.date = %q{2010-01-14}
+  s.date = %q{2010-01-20}
   s.description = %q{Apache Solr JavaBin format (binary format) implementation for Ruby.}
   s.email = %q{kennyj@gmail.com}
   s.extensions = ["ext/java_bin/ext/extconf.rb"]

data/test/test_java_bin_parser.rb

@@ -1,4 +1,5 @@
 # vim:fileencoding=utf-8
+
 require 'stringio'
 require 'helper'
 
@@ -12,13 +13,15 @@ end
 
 class TestJavaBinParser < Test::Unit::TestCase
 
+  READ_FLAG = (RUBY_VERSION >= '1.9' ? 'r:utf-8' : 'rb')
+
   private
   def write_v_int(i, output)
-    while ((i & ~0x7F) != 0) 
+    while ((i & ~0x7F) != 0)
       output.putc(((i & 0x7f) | 0x80))
       # i >>>= 7
       i = (i >> 7) # TODO 論理シフト
-    end 
+    end
     output.putc(i)
   end
 
@@ -27,9 +30,9 @@ class TestJavaBinParser < Test::Unit::TestCase
       if (size < 0x1f)
         output.putc(tag | size)
       else
- 	output.putc(tag | 0x1f)
- 	write_v_int(size - 0x1f, output)
-      end 
+        output.putc(tag | 0x1f)
+        write_v_int(size - 0x1f, output)
+      end
     else
       output.putc(tag)
       write_v_int(size, output)
@@ -60,22 +63,22 @@ class TestJavaBinParser < Test::Unit::TestCase
   end
 
   def test_javabin_dat
-    result = @parser.parse(open("fixtures/javabin.dat", "r:utf-8").read)
+    result = @parser.parse(open("fixtures/javabin.dat", READ_FLAG).read)
     assert result['response']['docs'][0]['features'].include?('eaiou with umlauts: ëäïöü')
     assert_equal result['response']['docs'][1]['incubationdate_dt'], Time.local(2006, 1, 17, 9, 0, 0)
-    assert_equal result['response']['docs'][1]['score'], 0.5030758380889893 
+    assert_equal result['response']['docs'][1]['score'], 0.5030758380889893
   end
 
   def test_javabin2_dat
-    result = @parser.parse(open("fixtures/javabin2.dat", "r:utf-8").read)
+    result = @parser.parse(open("fixtures/javabin2.dat", READ_FLAG).read)
     assert_equal 19, result['response']['docs'].size
   end
 
 
   TIMES = 5000
   def test_javabin_parse_and_ruby_eval
-    r  = open("fixtures/ruby.dat", "r:utf-8").read
-    jb = open("fixtures/javabin.dat", "r:utf-8").read
+    r  = open("fixtures/ruby.dat", READ_FLAG).read
+    jb = open("fixtures/javabin.dat", READ_FLAG).read
     puts ""
     r_et  = elapsed_time("ruby eval parse. ", TIMES) { eval(r) }
     jb_et = elapsed_time("javabin parse.   ", TIMES) { @parser.parse(jb) }
@@ -84,8 +87,8 @@ class TestJavaBinParser < Test::Unit::TestCase
   end
 
   def test_javabin2_parse_and_ruby2_eval
-    r  = open("fixtures/ruby2.dat", "r:utf-8").read
-    jb = open("fixtures/javabin2.dat", "r:utf-8").read
+    r  = open("fixtures/ruby2.dat", READ_FLAG).read
+    jb = open("fixtures/javabin2.dat", READ_FLAG).read
     puts ""
     r_et  = elapsed_time("ruby2 eval parse. ", TIMES) { eval(r) }
     jb_et = elapsed_time("javabin2 parse.   ", TIMES) { @parser.parse(jb) }
@@ -185,7 +188,7 @@ class TestJavaBinParser < Test::Unit::TestCase
     write_v_int(array.size, sio)
     array.each { |e| sio.putc e }
     sio.pos = 0
-    assert_equal array, @parser.parse(sio.read) 
+    assert_equal array, @parser.parse(sio.read)
   end
 
   def test_iterator

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: java_bin
 version: !ruby/object:Gem::Version 
-  version: 0.2.0
+  version: 0.3.0
 platform: ruby
 authors: 
 - kennyj
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-01-14 00:00:00 +09:00
+date: 2010-01-20 00:00:00 +09:00
 default_executable: 
 dependencies: []