jieba-php 0.25

fukuball · Feb 16, 2016 · 40c0834 · 40c0834
1 parent 3e3d0f2
commit 40c0834
Show file tree

Hide file tree

Showing 16 changed files with 638,723 additions and 639 deletions.
diff --git a/README.md b/README.md
@@ -6,7 +6,9 @@ jieba-php
 [![Codacy Badge](https://api.codacy.com/project/badge/grade/9360ebe8fc2d47d8a64f49f57d2f016f)](https://www.codacy.com/app/fukuball/jieba-php)
 [![Made with Love](https://img.shields.io/badge/made%20with-%e2%9d%a4-ff69b4.svg)](http://www.fukuball.com)
 
-"結巴"中文分詞：做最好的 PHP 中文分詞、中文斷詞組件，目前翻譯版本為 jieba-0.20 版本，未來再慢慢往上升級，效能也需要再改善，請有興趣的開發者一起加入開發！若想使用 Python 版本請前往 [fxsjy/jieba](https://github.com/fxsjy/jieba)
+"結巴"中文分詞：做最好的 PHP 中文分詞、中文斷詞組件，目前翻譯版本為 jieba-0.25 版本，未來再慢慢往上升級，效能也需要再改善，請有興趣的開發者一起加入開發！若想使用 Python 版本請前往 [fxsjy/jieba](https://github.com/fxsjy/jieba)
+
+現在已經可以支援繁體中文！只要將字典切換為 big 模式即可！
 
 "Jieba" (Chinese for "to stutter") Chinese text segmentation: built to be the best PHP Chinese word segmentation module.
 
@@ -476,6 +478,67 @@ array(21) {
 }
 ```
 
+功能 5)：切換成繁體字典
+==============
+
+代碼示例 (Tutorial)
+
+```php
+ini_set('memory_limit', '1024M');
+
+require_once dirname(dirname(__FILE__))."/vendor/multi-array/MultiArray.php";
+require_once dirname(dirname(__FILE__))."/vendor/multi-array/Factory/MultiArrayFactory.php";
+require_once dirname(dirname(__FILE__))."/class/Jieba.php";
+require_once dirname(dirname(__FILE__))."/class/Finalseg.php";
+use Fukuball\Jieba\Jieba;
+use Fukuball\Jieba\Finalseg;
+Jieba::init(array('mode'=>'default','dict'=>'big'));
+Finalseg::init();
+
+$seg_list = Jieba::cut("怜香惜玉也得要看对象啊！");
+var_dump($seg_list);
+
+$seg_list = Jieba::cut("憐香惜玉也得要看對象啊！");
+var_dump($seg_list);
+```
+
+Output:
+
+```php
+array(7) {
+  [0]=>
+  string(12) "怜香惜玉"
+  [1]=>
+  string(3) "也"
+  [2]=>
+  string(3) "得"
+  [3]=>
+  string(3) "要"
+  [4]=>
+  string(3) "看"
+  [5]=>
+  string(6) "对象"
+  [6]=>
+  string(3) "啊"
+}
+array(7) {
+  [0]=>
+  string(12) "憐香惜玉"
+  [1]=>
+  string(3) "也"
+  [2]=>
+  string(3) "得"
+  [3]=>
+  string(3) "要"
+  [4]=>
+  string(3) "看"
+  [5]=>
+  string(6) "對象"
+  [6]=>
+  string(3) "啊"
+}
+```
+
 常見問題
 ========
 1) 模型的數據是如何生成的？ https://github.com/fxsjy/jieba/issues/7
@@ -920,6 +983,67 @@ array(21) {
 }
 ```
 
+Function 5)：Use Traditional Chinese
+==============
+
+Example (Tutorial)
+
+```php
+ini_set('memory_limit', '1024M');
+
+require_once dirname(dirname(__FILE__))."/vendor/multi-array/MultiArray.php";
+require_once dirname(dirname(__FILE__))."/vendor/multi-array/Factory/MultiArrayFactory.php";
+require_once dirname(dirname(__FILE__))."/class/Jieba.php";
+require_once dirname(dirname(__FILE__))."/class/Finalseg.php";
+use Fukuball\Jieba\Jieba;
+use Fukuball\Jieba\Finalseg;
+Jieba::init(array('mode'=>'default','dict'=>'big'));
+Finalseg::init();
+
+$seg_list = Jieba::cut("怜香惜玉也得要看对象啊！");
+var_dump($seg_list);
+
+$seg_list = Jieba::cut("憐香惜玉也得要看對象啊！");
+var_dump($seg_list);
+```
+
+Output:
+
+```php
+array(7) {
+  [0]=>
+  string(12) "怜香惜玉"
+  [1]=>
+  string(3) "也"
+  [2]=>
+  string(3) "得"
+  [3]=>
+  string(3) "要"
+  [4]=>
+  string(3) "看"
+  [5]=>
+  string(6) "对象"
+  [6]=>
+  string(3) "啊"
+}
+array(7) {
+  [0]=>
+  string(12) "憐香惜玉"
+  [1]=>
+  string(3) "也"
+  [2]=>
+  string(3) "得"
+  [3]=>
+  string(3) "要"
+  [4]=>
+  string(3) "看"
+  [5]=>
+  string(6) "對象"
+  [6]=>
+  string(3) "啊"
+}
+```
+
 詞性說明
 ==============
 ```

diff --git a/composer.json b/composer.json
@@ -4,7 +4,7 @@
     "description": "結巴中文分詞（PHP 版本）：做最好的 PHP 中文分詞、中文斷詞組件",
     "keywords": ["Jieba", "PHP"],
     "license": "MIT",
-    "version": "0.24",
+    "version": "0.25",
     "authors": [
         {
             "name": "fukuball",

diff --git a/src/class/Jieba.php b/src/class/Jieba.php
@@ -59,6 +59,8 @@ public static function init($options = array())
 
         if ($options['dict']=='small') {
             $f_name = "dict.small.txt";
+        } else if ($options['dict']=='big') {
+            $f_name = "dict.big.txt";
         } else {
             $f_name = "dict.txt";
         }

diff --git a/src/cmd/demo.php b/src/cmd/demo.php
@@ -12,31 +12,48 @@
  * @version  GIT: <fukuball/jieba-php>
  * @link     https://github.com/fukuball/jieba-php
  */
-ini_set('memory_limit', '600M');
+ini_set('memory_limit', '1024M');
 
 require_once dirname(dirname(__FILE__))."/vendor/multi-array/MultiArray.php";
 require_once dirname(dirname(__FILE__))."/vendor/multi-array/Factory/MultiArrayFactory.php";
 require_once dirname(dirname(__FILE__))."/class/Jieba.php";
 require_once dirname(dirname(__FILE__))."/class/Finalseg.php";
 use Fukuball\Jieba\Jieba;
 use Fukuball\Jieba\Finalseg;
-Jieba::init(array('mode'=>'test','dict'=>'samll'));
+Jieba::init(array('mode'=>'test','dict'=>'big'));
 Finalseg::init();
 
 $seg_list = Jieba::cut("怜香惜玉也得要看对象啊！");
 var_dump($seg_list);
 
+$seg_list = Jieba::cut("憐香惜玉也得要看對象啊！");
+var_dump($seg_list);
+
 echo "Full Mode: \n";
 $seg_list = Jieba::cut("我来到北京清华大学", true);
 var_dump($seg_list);
 
+echo "Full Mode: \n";
+$seg_list = Jieba::cut("我來到北京清華大學", true);
+var_dump($seg_list);
+
 echo "Default Mode: \n";
 $seg_list = Jieba::cut("我来到北京清华大学", false);
 var_dump($seg_list);
 
+echo "Default Mode: \n";
+$seg_list = Jieba::cut("我來到北京清華大學", false);
+var_dump($seg_list);
+
 $seg_list = Jieba::cut("他来到了网易杭研大厦");
 var_dump($seg_list);
 
+$seg_list = Jieba::cut("他來到了網易杭研大廈");
+var_dump($seg_list);
+
 $seg_list = Jieba::cutForSearch("小明硕士毕业于中国科学院计算所，后在日本京都大学深造");
 var_dump($seg_list);
+
+$seg_list = Jieba::cutForSearch("小明碩士畢業于中國科學院計算所，後在日本京都大學深造");
+var_dump($seg_list);
 ?>
diff --git a/src/cmd/demo_extract_tags.php b/src/cmd/demo_extract_tags.php
@@ -12,7 +12,7 @@
  * @version  GIT: <fukuball/jieba-php>
  * @link     https://github.com/fukuball/jieba-php
  */
-ini_set('memory_limit', '600M');
+ini_set('memory_limit', '1024M');
 
 require_once dirname(dirname(__FILE__))."/vendor/multi-array/MultiArray.php";
 require_once dirname(dirname(__FILE__))."/vendor/multi-array/Factory/MultiArrayFactory.php";
@@ -22,7 +22,7 @@
 use Fukuball\Jieba\Jieba;
 use Fukuball\Jieba\Finalseg;
 use Fukuball\Jieba\JiebaAnalyse;
-Jieba::init(array('mode'=>'test','dict'=>'samll'));
+Jieba::init(array('mode'=>'test','dict'=>'big'));
 Finalseg::init();
 JiebaAnalyse::init();
 

diff --git a/src/cmd/demo_posseg.php b/src/cmd/demo_posseg.php
@@ -12,7 +12,7 @@
  * @version  GIT: <fukuball/jieba-php>
  * @link     https://github.com/fukuball/jieba-php
  */
-ini_set('memory_limit', '600M');
+ini_set('memory_limit', '1024M');
 
 require_once dirname(dirname(__FILE__))."/vendor/multi-array/MultiArray.php";
 require_once dirname(dirname(__FILE__))."/vendor/multi-array/Factory/MultiArrayFactory.php";
@@ -22,13 +22,19 @@
 use Fukuball\Jieba\Jieba;
 use Fukuball\Jieba\Finalseg;
 use Fukuball\Jieba\Posseg;
-Jieba::init();
+Jieba::init(array('mode'=>'test','dict'=>'big'));
 Finalseg::init();
 Posseg::init();
 
 $seg_list = Posseg::cut("这是一个伸手不见五指的黑夜。我叫孙悟空，我爱北京，我爱Python和C++。");
 var_dump($seg_list);
 
-//$seg_list = Posseg::posTagReadable($seg_list);
-//var_dump($seg_list);
+$seg_list = Posseg::posTagReadable($seg_list);
+var_dump($seg_list);
+
+$seg_list = Posseg::cut("這是一個伸手不見五指的黑夜。我叫孫悟空，我愛北京，我愛Python和C++");
+var_dump($seg_list);
+
+$seg_list = Posseg::posTagReadable($seg_list);
+var_dump($seg_list);
 ?>
diff --git a/src/cmd/gen_dict_json.php b/src/cmd/gen_dict_json.php
@@ -18,7 +18,7 @@
 require_once dirname(dirname(__FILE__))."/vendor/multi-array/Factory/MultiArrayFactory.php";
 use Fukuball\Tebru\MultiArray;
 
-$content = fopen(dirname(dirname(__FILE__))."/dict/dict.txt", "r");
+$content = fopen(dirname(dirname(__FILE__))."/dict/dict.big.txt", "r");
 
 $trie = new MultiArray(array());
 
@@ -39,6 +39,6 @@
 
 }
 
-file_put_contents(dirname(dirname(__FILE__))."/dict/dict.txt.json", json_encode($trie->storage));
-file_put_contents(dirname(dirname(__FILE__))."/dict/dict.txt.cache.json", json_encode($trie->cache));
+file_put_contents(dirname(dirname(__FILE__))."/dict/dict.big.txt.json", json_encode($trie->storage));
+file_put_contents(dirname(dirname(__FILE__))."/dict/dict.big.txt.cache.json", json_encode($trie->cache));
 ?>