fix thread problem (#176)

Author: IceMimosa

CHANGES.md

@@ -0,0 +1,15 @@
+# Change Log
+
+* 1.1.3
+  * 新增自定义地址设置
+* 1.1.4
+  * 修复一些匹配错误的bug
+* 1.1.6
+  * 升级地址库和包版本, 修复一些匹配错误的地址
+* 1.2.0
+  * geocoding项目转移到组织：bitlap/geocoding
+* 1.3.0
+  * 新增自定义地址文件库配置参数
+  * 添加自定义地址新增replace参数
+* 1.3.1 
+  * 修复normalizing方法多线程调用报错的问题

README.md

@@ -18,7 +18,7 @@
         <!--  从1.2.0开始迁移到bitlap，原groupId: io.patamon.geocoding     -->
         <groupId>org.bitlap</groupId> 
         <artifactId>geocoding</artifactId>
-        <version>1.3.0</version>
+        <version>1.3.1</version>
     </dependency>
 </dependencies>
 ```
@@ -176,17 +176,12 @@ Address(
 
 ## Release Log
 
-* 1.1.3
-  * 新增自定义地址设置
-* 1.1.4
-  * 修复一些匹配错误的bug
-* 1.1.6
-  * 升级地址库和包版本, 修复一些匹配错误的地址
-* 1.2.0
-  - geocoding项目转移到组织：bitlap/geocoding
-* 1.3.0
-  * 新增自定义地址文件库配置参数
-  * 添加自定义地址新增replace参数
+[Change Log](./CHANGES.md)
+
+
+## LICENSE
+
+MIT
 
 [1]:https://lsp.wuliu.taobao.com/locationservice/addr/output_address_town.do
 [2]:http://www.stats.gov.cn/tjsj/tjbz/tjyqhdmhcxhfdm/2015/index.html

pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>org.bitlap</groupId>
     <artifactId>geocoding</artifactId>
-    <version>1.3.0</version>
+    <version>1.3.1</version>
 
     <name>geocoding</name>
     <description>地理编码技术，提供地址标准化和相似度计算。</description>
@@ -293,7 +293,7 @@
                     <plugin>
                         <groupId>org.sonatype.plugins</groupId>
                         <artifactId>nexus-staging-maven-plugin</artifactId>
-                        <version>1.6.12</version>
+                        <version>1.6.13</version>
                         <extensions>true</extensions>
                         <configuration>
                             <serverId>central</serverId>

src/main/java/org/bitlap/geocoding/core/Context.kt

@@ -3,7 +3,6 @@ package org.bitlap.geocoding.core
 import org.bitlap.geocoding.core.impl.DefaultAddressInterpreter
 import org.bitlap.geocoding.core.impl.DefaultAddressPersister
 import org.bitlap.geocoding.core.impl.DefaultRegionCache
-import org.bitlap.geocoding.core.impl.RegionInterpreterVisitor
 import org.bitlap.geocoding.core.impl.SimilarityComputer
 
 /**
@@ -16,8 +15,7 @@ open class Context(
     val dataClassPath: String,
     val strict: Boolean,
     val persister: AddressPersister = DefaultAddressPersister(DefaultRegionCache(dataClassPath)),
-    val visitor: TermIndexVisitor = RegionInterpreterVisitor(persister, strict),
-    val interpreter: AddressInterpreter = DefaultAddressInterpreter(persister, visitor),
+    val interpreter: AddressInterpreter = DefaultAddressInterpreter(persister, strict),
     val computer: Computer = SimilarityComputer(),
 ) {
 

src/main/java/org/bitlap/geocoding/core/impl/DefaultAddressInterpreter.kt

@@ -21,7 +21,7 @@ import java.util.regex.Pattern
  * Created by IceMimosa
  * Date: 2017/1/17
  */
-open class DefaultAddressInterpreter(val persister: AddressPersister, val visitor: TermIndexVisitor) : AddressInterpreter {
+open class DefaultAddressInterpreter(val persister: AddressPersister, val strict: Boolean) : AddressInterpreter {
 
     private var indexBuilder: TermIndexBuilder? = null
     private val ignoringRegionNames = mutableListOf(
@@ -207,7 +207,7 @@ open class DefaultAddressInterpreter(val persister: AddressPersister, val visito
      * 将`脏`地址进行标准化处理, 解析成 [AddressEntity]
      */
     override fun interpret(address: String?): AddressEntity? {
-        return interpret(address, visitor)
+        return interpret(address, RegionInterpreterVisitor(persister, strict))
     }
 
     private fun interpret(address: String?, visitor: TermIndexVisitor): AddressEntity? {

src/main/java/org/bitlap/geocoding/core/impl/SimilarityComputer.kt

@@ -25,11 +25,11 @@ import org.bitlap.geocoding.utils.isNumericChars
 /**
  * Desc: 相似度算法相关逻辑
  *
- * 1. >>>>> 关于 TF-IDF
- *  TC: 词数 Term Count, 某个词在文档中出现的次数
- *  TF: 词频 Term Frequency, 某个词在文档中出现的频率. TF = 该词在文档中出现的次数 / 该文档的总词数
- *  IDF: 逆文档词频 Inverse Document Frequency. IDF = log( 语料库文档总数 / ( 包含该词的文档数 + 1 ) ). 分母加1是为了防止分母出现0的情况
- *  TF-IDF: 词条的特征值, TF-IDF = TF * IDF
+ * * 关于 TF-IDF
+ *   * TC: 词数 Term Count, 某个词在文档中出现的次数
+ *   * TF: 词频 Term Frequency, 某个词在文档中出现的频率. TF = 该词在文档中出现的次数 / 该文档的总词数
+ *   * IDF: 逆文档词频 Inverse Document Frequency. IDF = log( 语料库文档总数 / ( 包含该词的文档数 + 1 ) ). 分母加1是为了防止分母出现0的情况
+ *   * TF-IDF: 词条的特征值, TF-IDF = TF * IDF
  *
  * Mail: chk19940609@gmail.com
  * Created by IceMimosa

src/test/java/org/bitlap/geocoding/TestSimilarity.kt

@@ -1,6 +1,8 @@
 package org.bitlap.geocoding
 
 import org.junit.Test
+import java.util.concurrent.Callable
+import java.util.concurrent.Executors
 
 /**
  * Desc: 测试相似度
@@ -49,4 +51,22 @@ open class TestSimilarity {
 
         println("相似度结果分析 >>>>>>>>> " + Geocoding.similarityWithResult(addr1, addr2))
     }
-}
+
+    @Test
+    fun test_similarity_threads() {
+        val pool = Executors.newFixedThreadPool(10)
+
+        val addr1 = "中国湖南郴州宜章县梅田镇【梅田镇】(梅田镇附近)"
+        val addr2 = "湖南省郴州市宜章县梅田镇上寮村2组"
+
+        (1 .. 1000).map {
+            pool.submit(Callable {
+                Geocoding.similarity(addr1, addr2)
+            })
+        }.forEach {
+            val r = it.get()
+            assert(0.8164965809277261 == r)
+        }
+        pool.shutdown()
+    }
+}