问题
I´m using Lucene.NET 3.0.3 for indexing the content of word-, excel-, etc. documents and some custom fields for each document.
If I index a field named "title" as Field.Index.NOT_ANALYZED the Lucene-Index stored the field in correct form. The hole title is stored in a single token. That´s what I want.
e.g. title of document is "Lorem ipsum dolor"
field in Lucene-index: "Lorem ipsum dolor"
If I search using exact search in this field I get no results.
My searchterm looks like: title:"Lorem ipsum dolor"
For searching i´m use the same StandardAnalzer.
Why I can´t find the document?
回答1:
StandardAnalyzer is sensitive to whitespace, among other delimiters. That is, it tokenizes the search term into three tokens:
( Lorem, ipsum, dolor )
But you indexed field title using Field.Index.NOT_ANALYZED so none of the three tokens above can match the single token in this field:
( Lorem ipsum dolor )
Use KeywordAnalyzer, which tokenizes the entire field value as a single token. As always, you need to use the same analyzer for both indexing and searching.
来源：https://stackoverflow.com/questions/19836528/search-in-single-token-field-using-lucene-net