add: api nodejieba.textRankExtract

.gitignore

@@ -17,3 +17,4 @@ t.*
 node_modules
 npm-debug.log
 .idea
+package-lock.json

README.md

@@ -120,6 +120,13 @@ console.log(nodejieba.extract("升职加薪，当上CEO，走上人生巅峰。"
 //  { word: '升职', weight: 10.8561552143 },
 //  { word: '加薪', weight: 10.642581114 },
 //  { word: '巅峰', weight: 9.49395840471 } ]
+
+console.log(nodejieba.textRankExtract("升职加薪，当上CEO，走上人生巅峰。", topN));
+//[ { word: '当上', weight: 1 },
+//  { word: '不用', weight: 0.9898479330698993 },
+//  { word: '多久', weight: 0.9851260595435759 },
+//  { word: '加薪', weight: 0.9830464899847804 },
+//  { word: '升职', weight: 0.9802777682279076 } ]
 ```
 
 更详细的其他用法请看 [test/demo.js](test/demo.js)

README_EN.md

@@ -95,6 +95,13 @@ console.log(nodejieba.extract("升职加薪，当上CEO，走上人生巅峰。"
 //  { word: '升职', weight: 10.8561552143 },
 //  { word: '加薪', weight: 10.642581114 },
 //  { word: '巅峰', weight: 9.49395840471 } ]
+
+console.log(nodejieba.textRankExtract("升职加薪，当上CEO，走上人生巅峰。", topN));
+//[ { word: '当上', weight: 1 },
+//  { word: '不用', weight: 0.9898479330698993 },
+//  { word: '多久', weight: 0.9851260595435759 },
+//  { word: '加薪', weight: 0.9830464899847804 },
+//  { word: '升职', weight: 0.9802777682279076 } ]
 ```
 
 See details in [test/demo.js](test/demo.js)

index.js

@@ -44,6 +44,7 @@ wrapWithDictLoad("cutForSearch");
 wrapWithDictLoad("cutSmall");
 wrapWithDictLoad("tag");
 wrapWithDictLoad("extract");
+wrapWithDictLoad("textRankExtract");
 wrapWithDictLoad("insertWord");
 
 module.exports = exports;

lib/nodejieba.cpp

@@ -4,6 +4,7 @@
 
 #include "cppjieba/Jieba.hpp"
 #include "cppjieba/KeywordExtractor.hpp"
+#include "cppjieba/TextRankExtractor.hpp"
 
 NodeJieba::NodeJieba(Napi::Env env, Napi::Object exports) {
   DefineAddon(exports, {
@@ -15,6 +16,7 @@ NodeJieba::NodeJieba(Napi::Env env, Napi::Object exports) {
     InstanceMethod("cutSmall", &NodeJieba::cutSmall),
     InstanceMethod("tag", &NodeJieba::tag),
     InstanceMethod("extract", &NodeJieba::extract),
+    InstanceMethod("textRankExtract", &NodeJieba::textRankExtract),
     InstanceMethod("insertWord", &NodeJieba::insertWord)
   });
 }
@@ -37,6 +39,13 @@ Napi::Value NodeJieba::load(const Napi::CallbackInfo& info) {
                                   idfPath,
                                   stopWordsPath);
 
+  delete _text_rank_extractor_handle;
+  _text_rank_extractor_handle = new cppjieba
+		      ::TextRankExtractor(dictPath,
+                                  modelPath,
+                                  stopWordsPath,
+                                  userDictPath);
+
   return Napi::Boolean::New(info.Env(), true);
 }
 
@@ -189,3 +198,21 @@ Napi::Value NodeJieba::extract(const Napi::CallbackInfo& info) {
   WrapPairVector(info.Env(), words, outArray);
   return outArray;
 }
+
+Napi::Value NodeJieba::textRankExtract(const Napi::CallbackInfo& info) {
+  if (info.Length() != 2) {
+    return Napi::Boolean::New(info.Env(), false);
+  }
+  std::string sentence = info[0].As<Napi::String>();
+  size_t topN = info[1].As<Napi::Number>().Int32Value();
+
+  if( !_text_rank_extractor_handle ){
+    Napi::Error::New(info.Env(), "Before calling any other function you have to call load() first").ThrowAsJavaScriptException();
+  }
+
+  std::vector<std::pair<std::string, double> > words;
+  _text_rank_extractor_handle->Extract(sentence, words, topN);
+  Napi::Array outArray;
+  WrapPairVector(info.Env(), words, outArray);
+  return outArray;
+}

lib/nodejieba.h

@@ -3,7 +3,7 @@
 
 #include <napi.h>
 
-namespace cppjieba{ class Jieba; }
+namespace cppjieba{ class Jieba; class TextRankExtractor; }
 
 class NodeJieba : public Napi::Addon<NodeJieba> {
 public:
@@ -18,9 +18,11 @@ class NodeJieba : public Napi::Addon<NodeJieba> {
   Napi::Value cutSmall(const Napi::CallbackInfo& info);
   Napi::Value tag(const Napi::CallbackInfo& info);
   Napi::Value extract(const Napi::CallbackInfo& info);
+  Napi::Value textRankExtract(const Napi::CallbackInfo& info);
   Napi::Value insertWord(const Napi::CallbackInfo& info);
 
   cppjieba::Jieba* _jieba_handle{nullptr};
+  cppjieba::TextRankExtractor* _text_rank_extractor_handle{nullptr};
 };
 
 #endif // NODEJIBEA_SRC_NODEJIEBA_H

test/demo.js

@@ -29,6 +29,9 @@ var topN = 5;
 result = nodejieba.extract(sentence, topN);
 console.log(result);
 
+result = nodejieba.textRankExtract(sentence, topN);
+console.log(result);
+
 result = nodejieba.cut("男默女泪");
 console.log(result);
 nodejieba.insertWord("男默女泪");

test/test.js

@@ -298,6 +298,30 @@ describe("nodejieba", function() {
           }]);
   });
 
+  it('nodejieba.textRankExtract(sentence, 5)', function() {
+    nodejieba.textRankExtract(sentence, 5).should.eql([
+		  {
+              "weight": 1,
+		      "word": "当上"
+		  },
+		  {
+              "weight": 0.9898479330698993,
+		      "word": "不用"
+		  },
+		  {
+		      "weight": 0.9851260595435759,
+		      "word": "多久"
+		  },
+		  {
+		      "weight": 0.9830464899847804,
+		      "word": "加薪"
+		  },
+		  {
+		      "weight": 0.9802777682279076,
+		      "word": "升职"
+		  }]);
+  });
+
   it('nodejieba.cut("红掌拨清波")', function() {
     nodejieba.cut("红掌拨清波").should.eql([
       '红掌',

types/index.d.ts

@@ -24,6 +24,7 @@ declare module "nodejieba" {
   export function cutForSearch(sentence: string, strict?: boolean): string[];
   export function tag(sentence: string): TagResult[];
   export function extract(sentence: string, threshold: number): ExtractResult[];
+  export function textRankExtract(sentence: string, threshold: number): ExtractResult[];
   export function insertWord(sentence: string): boolean;
   export function cutSmall(sentence: string, small: number): boolean;
 }

typescript_demo.ts

@@ -25,11 +25,14 @@ const topN = 5;
 result = nodejieba.extract(sentence, topN);
 console.log(result);
 
+result = nodejieba.textRankExtract(sentence, topN);
+console.log(result);
+
 result = nodejieba.cut("男默女泪");
 console.log(result);
 nodejieba.insertWord("男默女泪");
 result = nodejieba.cut("男默女泪");
 console.log(result);
 
 result = nodejieba.cutSmall("南京市长江大桥", 3);
-console.log(result);
+console.log(result);