如何使用Lucene索引和搜索结构文本文件？

Question

This is the first time I use Lucene, and I have a text file like this :

id,name,address,hobby
1,namm1,address1,football
2,namm2,address2,football
3,namm3,address3,football
4,namm4,address4,football
5,namm5,address5,football
6,namm6,address6,basketball
7,namm7,address7,basketball
8,namm8,address1,football
9,namm9,address8,swimming
...

The file above is a text file which contains 1,000,000 lines. Now I want to find the record whose address is address1 and its hobby is football from the file, and then put the record into another file like this:

1,namm1,address1,football
8,namm8,address1,football
...

The first file is extremely large, so it would be very slow to find the record one after another. I want to and build an index (according to address and hobby) for the first file with Lucene. Then I can quickly find the record whose address is address1 and its hobby is football , and put it in a new file. I have never programmed with Lucene. Who can give me a similar example?

Answer 1

This is pretty simple. When you index a file with Lucene, you can define your own "Analyzer". In a nutshell, the analyzer extracts information from a source and puts it into "fields" of a lucene "document".

When you search something, you can define which fields Lucene should consider.

So the solution in your case is to write an analyzer which puts each column into a field. Use a MultiFieldQueryParser and in your query, specify the field names. For your example, the query would be

address:address1 hobby:football

Answer 2

我认为他根本不需要编写分析器，他可以使用许多内置分析器之一，使用Java代码来解析每一行并将每个值放入相应的字段中