Merge pull request #2 from duguying/develop

Support Chinese Word Segmentation to deal with polyphone, release v0.4.0

.travis.yml

@@ -2,6 +2,7 @@ language: c
 compiler: 
     - "gcc"
 before_script:
+    - "sudo apt-get update -qq"
     - "sudo apt-get install php5-dev"
 script: 
     - "phpize"

CMakeLists.txt

@@ -22,26 +22,33 @@ add_definitions(
 set(PHPSDK "C:/php/SDK")
 
 link_directories(
-	"${PHPSDK}/lib"
+	# "${PHPSDK}/lib"
 	# "D:/wamp/bin/php/php5.3.10/dev"
 )
 include_directories(
-	"${PHPSDK}/include"
-	"${PHPSDK}/include/main"
-	"${PHPSDK}/include/TSRM"
-	"${PHPSDK}/include/win32"
-	"${PHPSDK}/include/Zend"
+	# "${PHPSDK}/include"
+	# "${PHPSDK}/include/main"
+	# "${PHPSDK}/include/TSRM"
+	# "${PHPSDK}/include/win32"
+	# "${PHPSDK}/include/Zend"
 )
 
 set(
 	PY_LIST
-	./pinyin.c
-	./py_hashtable.c
-	./py_pinyin.c
+	./src/pinyin.c
+	./src/hashtable.c
+	./src/py_pinyin.c
 )
-add_library(php_pinyin SHARED ${PY_LIST})
 
-target_link_libraries(php_pinyin php5ts)
+set(
+    PY_TEST
+    ./src/hashtable.c
+    ./src/py_pinyin.c
+    ./src/test_pinyin.c
+)
 
+# add_library(php_pinyin SHARED ${PY_LIST})
+# target_link_libraries(php_pinyin php5ts)
 
+add_executable(pinyin_test ${PY_TEST})
 

README.md

@@ -19,31 +19,22 @@ make
 make test
 ```
 
-**win32(cmake)**
+## configuration
+add following into `php.ini`
 
-使用cmake编译，首先，您必须确保您本地存在PHP SDK(这个SDK可以由您编译php得来)，以及目标php版本的静态库(如`php5ts.lib`)，您可以按以下步骤进行：
-
-- 打开VC命令行(以目标版本为VC9为例)
-
-- 若您的SDK版本与目标版本不一致，搜索文件`${PHPSDK}/include/main/config.w32.h`中的字符串`PHP_COMPILER_ID`，将该宏的定义改为对应目标版本标识，如此例为`VC9`。
-
-- 设置`CMakeList.txt`中`add_definitions`的对应参数以符合自己的环境。
-
-- 设置`CMakeList.txt`中`PHPSDK`路径。
-
-- 在VC9命令行中编译：
-
-	```shell
-	mkdir build & cd build
-	cmake -G"NMake Makefiles" ..
-	nmake
-	```
-
-  如此便可得到php_pinyin.dll
+```
+pinyin.chars=/path/to/chars.csv
+pinyin.words=/path/to/words.csv
+```
 
-**about**
+## usage
 
-the file `pinyin.inc`, it is a resource file. of course, you needn't change the file name or create a `pinyin.inc`, the file `pinyin.inc` will be built by a php script in the `/data/` directory. 
+```php
+<?php
+echo pinyin("汉"),"\n";
+echo pinyin("わたしわ阿飞, and my English name is Rex Lee. 网名是独孤影！ ^_^。下面是一段多音分词歧义测试，这个人无伤无臭味。"),"\n";
+?>
+```
 
 # License #
 

clean

@@ -0,0 +1,28 @@
+rm -rf autom4te.cache
+rm -rf build
+rm -rf include
+rm -rf modules
+
+rm acinclude.m4
+rm aclocal.m4
+rm config.sub
+rm configure
+rm configure.in
+rm config.status
+rm config.guess
+rm config.h
+rm config.h.in
+rm config.log
+rm config.nice
+rm Makefile*
+rm ltmain.sh
+rm libtool
+rm mkinstalldirs
+rm missing
+rm install-sh
+rm run-tests.php
+
+rm -rf *.la
+rm -rf src/*.lo
+rm -rf *.a
+rm -rf *.so

config.m4

@@ -3,7 +3,7 @@ PHP_ARG_WITH(pinyin, for pinyin support,
 
 PHP_ADD_INCLUDE(pinyin)
 
-PHP_NEW_EXTENSION(pinyin, pinyin.c \
-						py_hashtable.c \
-						py_pinyin.c \
+PHP_NEW_EXTENSION(pinyin, src/pinyin.c \
+						src/hashtable.c \
+						src/py_pinyin.c \
 				,$ext_shared)

config.w32

@@ -1,6 +1,6 @@
 ARG_ENABLE("pinyin", "for pinyin support", "yes");
 
-EXTENSION("pinyin", "pinyin.c \
-					py_hashtable.c \
-					py_pinyin.c");
+EXTENSION("pinyin", "src/pinyin.c \
+					src/hashtable.c \
+					src/py_pinyin.c");
 

data/README.md

@@ -1,11 +1,5 @@
-# get the latest data #
+## about data ##
 
-you can get the latest data file(pinyin.inc) from the internet via run the script `getData.sh`.
+here is the data for pinyin-php extension
 
-```
-./getData.sh
-```
 
-## about pinyin-php resource build ##
-
-the raw data is in the MySQL database, I read the data from the database and build the data into three variables.They are `char* cnchar`,`char* pinyin` and `char* py`. Chinese characters is in the variable `cnchar`. Chinese PinYin characters is in the variable `pinyin`. The PinYin alphabet is in the variable `py`.