Pylucene7.6创建倒排索引

 

import os,sys,glob
import lucene
import pandas as pd
import sys, os, lucene, threading, time
from java.io import File
from java.nio.file import Paths
from org.apache.lucene.analysis.miscellaneous import LimitTokenCountAnalyzer
from org.apache.lucene.analysis.standard import StandardAnalyzer
from org.apache.lucene.analysis.cjk import CJKAnalyzer
from org.apache.lucene.document import Document, Field, FieldType,StringField, TextField
from org.apache.lucene.index import FieldInfo, IndexWriter, IndexWriterConfig
from org.apache.lucene.store import SimpleFSDirectory
from org.apache.lucene.analysis.core import SimpleAnalyzer
from org.apache.lucene.util import Version

df=pd.read_json('/Users/...',orient='records',encoding='UTF-8')

lucene.initVM()
indir=Paths.get("/Users/shiyufei/Desktop/index") 
INDEXIDR= indir
indexdir= SimpleFSDirectory(indir) #实例化一个SimpleFSDirectory对象，将索引保存至本地文件之中，保存的路径为自定义的路径“INDEXIDR”

analyzer= CJKAnalyzer() #实例化一个CJKAnalyzer分析器，该分析器对中文支持较好,适用于中文内容的文本处理
config = IndexWriterConfig(analyzer)
writer = IndexWriter(indexdir, config) #实例化一个IndexWriter对象，索引写入类 

title=df['title']  
url=df['current_url']
text=df['news_text']
date=df['news_date']

for i in range(len(df)):
    document=Document()
    document.add(Field("title", "".join(df['title'][i]), TextField.TYPE_STORED)) 
    document.add(Field("url", "".join(df['current_url'][i]), StringField.TYPE_STORED))
    document.add(Field("text", "".join(df['news_text'][i]), TextField.TYPE_STORED))
    document.add(Field("date","".join(df['news_date'][i]), StringField.TYPE_STORED))
    writer.addDocument(document) #实例化IndexWriter对象，将文档document写入索引文件

writer.commit()
writer.close()

 通过luke查看构建的索引