Alibaba architects spent a year sorting out the "Lucene advanced document", and you are also a big factory employee!

Lucene It's a set for

Full text search and search

Open source library , from Apache The software foundation supports and provides .Lucene Provides a simple but powerful application interface , Be able to do full-text indexing and search , stay Java In the development environment Lucene Is a mature free open source tool ; In itself ,

Lucene Now and in recent years , The most popular free Java Information retrieval library

.

in addition ,Lucene No crawler function , If you need to get content, you need to build your own crawler Application . Lucene Only index and search .

You need to get this Alibaba architect spent a year sorting 《Lucene Advanced documentation 》 Friends of can click in to find out , Click on

——【 Portal 】——

that will do .

The content structure is divided into 13 Parts of .

Directory as follows ：

One 、 Theoretical basis of search technology

Lucene The goal is to provide software developers with an easy-to-use toolkit , In order to facilitate the realization of full-text retrieval in the target system , Or build a complete full-text search engine based on this .

• Why learn Lucene

• Data query method

• Application scenario of full-text retrieval technology

Two 、Lucene Introduce

The computer indexing program scans every word in the article , Make an index of each word , Indicate the number and location of the word in the article , When the user inquires , The search program will search according to the index established in advance , And the search results will be fed back to the user's search method .

• What is full-text retrieval

• What is? Lucene

• Lucene Official website

3、 ... and 、Lucene The process of full-text retrieval

• Index and search flow chart

• Indexing process

• Search process

Four 、Lucene introduction

• Lucene Get ready

• development environment

• establish Java engineering

• Indexing process

• Use Luke Look at the index

• Search process

5、 ... and 、Field Domain type

Field It's a field in the document , Include Field Name and Field It's worth two parts , A document can contain more than one Field,Document It's just Field One of the carriers of ,Field The value is the content to be indexed , It's also the content to search for .

• Field attribute

• Field Common types

• Field modify

6、 ... and 、 Index maintenance

Managers change book information through e-commerce system , At this time, the relational database is updated , If you use lucene Search for book information , Need to be in the database table book Update information as it changes lucene The index library .

• demand

• Add index

• Modify the index

• Delete index

7、 ... and 、 Word segmentation is

In the face of Document Before indexing the contents of , You need to use a word splitter to segment words  , The purpose of word segmentation is to search . The main process of word segmentation is to segment words first and then filter .

• Participle understanding

• Analyzer Use time

• Lucene Native word breaker

• Third party Chinese word segmentation

8、 ... and 、Lucene Advanced search

QueryParser Support the default search domain , The first - Parameters are default search fields . If in execution parse Method time , If the query syntax contains a domain name, search from the specified domain name , If there are only keywords to query , Search results from the default search domain .

• Text search

• Numerical range search

• Combined search

Nine 、 Search for cases

• Introduce dependencies

• Add pages and resources to the project

• Create packages and startup classes

• The configuration file

• Business code

Ten 、Lucene The underlying storage structure ( senior )

• Understand in detail lucene Storage structure

• Index library physical files

• Index library file extension comparison table

• Construction of dictionary

11、 ... and 、Lucene Optimize ( senior )

• confifig.setMaxBufffferedDocs(100000);  Control to write a new segment It's stored in memory before document Number of , Setting a larger number can speed up indexing .（
  The higher the value, the faster the indexing speed ,  But it will consume more memory
  ）

• indexWriter.forceMerge( Number of documents );  Set up N Documents are merged into one segment （
  The higher the value, the faster the indexing speed ,  The slower the search speed ;  The smaller the value, the slower the indexing speed ,  The faster the search
  ）

• Solve a large number of disks IO

• Choose the right word breaker

• Choose an appropriate location to store the index library

• Search for api The choice of

Twelve 、Lucene Correlation ranking ( senior )

Lucene Score the relevance of query keywords and index documents , Those with high scores are in front .

13、 ... and 、Lucene Precautions for use ( senior )

• Keywords are case sensitive  OR AND TO Key words are case sensitive ,lucene Only in capitals , Lowercase words are treated as ordinary words .

• Reading and writing are mutually exclusive   There can only be one write to the index at a time , Search while writing

• File lock   Forced exit in the process of writing index will be in tmp The catalog leaves a lock file , Make future write operations impossible , You can delete it manually

• Time format  lucene Only one time format is supported yyMMddHHmmss, So you pass a yy-MM-dd HH:mm:ss Time for lucene It's not going to be dealt with as time

• Set up boost  Sometimes when searching, the weight of a field needs to be larger , For example, you may think that articles with keywords in the title are more valuable than articles with keywords in the text , You can put the title of boost Set it bigger , Then the search results will give priority to the articles with keywords in the title

You need to get this Alibaba architect spent a year sorting 《Lucene Advanced learning documentation 》 Friends of can click in to find out , Click on

——【 Portal 】——

that will do .