CODEWITHSUNDEEP
parsingpylucene
sunner
sunner

Reputation: 23

How to use PyLucene's MultiFieldQueryParser?

I'm using PyLucene and trying to use MultiFieldQueryParser with field boosting. I saw in the documentation that the constructor's fields are MultiFieldQueryParser(String[] fields, Analyzer analyzer, Map<String,Float> boosts) (boosts being optional) and wrote my constructor like this:

def mapFunc(a, b):
    return b

mapping = map(mapFunc, ("Post", "Title", "Comments", "Subreddit"), (1, 1.2, 0.1, 0.5))

def retrieve(storeDir, query, amt:int):
    searchDir = NIOFSDirectory(Paths.get(storeDir))
    searcher = IndexSearcher(DirectoryReader.open(searchDir))

    # parser = QueryParser("Post", StandardAnalyzer())
    # parser = MultiFieldQueryParser(["Post", "Title", "Comments", "Subreddit"], StandardAnalyzer(), mapping)
    parser = MultiFieldQueryParser(["Post", "Title", "Comments", "Subreddit"], StandardAnalyzer())
    parsed_query = parser.parse(query)

Didn't find much in terms of PyLucene documentation, and the few examples I found on Lucene were for Java and did not include examples for boosting.

I tried removing the boost mapping to see if it would work and got a different error saying 'parse' for 'QueryParserBase' object doesn't apply to a 'str' object (not sure what that's all about, the QueryParser worked just fine).

Using boost mapping gave me this error about invalid arguments. Not sure how to fix this, do I need to wrap the Python map in a Java container? If so, how would I do that?

Upvotes: 0

Views: 43

Answers (0)

Related Questions