如何通过多值分层构面字段获取 Lucene 中构面最低粒度的构面计数？

编程问答 2022-05-27

问题描述

我有包含标题、作者和类型的文档。我已将这些文件编入索引如下。我不确定是否应该为此使用多值和分层方面。

# Imports
import lucene

from org.apache.lucene.index import IndexWriter,IndexWriterConfig,DirectoryReader
from org.apache.lucene.store import SimpleFSDirectory
from java.nio.file import Paths
from org.apache.lucene.analysis.en import EnglishAnalyzer
from org.apache.lucene.facet import Facets,FacetField,FacetResult,FacetsConfig,FacetsCollector
from org.apache.lucene.facet.taxonomy import FastTaxonomyFacetCounts
from org.apache.lucene.facet.taxonomy.directory import DirectoryTaxonomyWriter,DirectoryTaxonomyReader
from org.apache.lucene.facet import DrillSideways,DrillDownQuery
from org.apache.luce3ne.document import Document,Field,TextField
from org.apache.lucene.search import BooleanClause,IndexSearcher
from org.apache.lucene.queryparser.classic import QueryParser


# Indexing
index_path = "../index_try"
taxo_path = "../taxo_try"

lucene.initVM()

config = IndexWriterConfig(EnglishAnalyzer())
config.setopenMode(IndexWriterConfig.OpenMode.CREATE)

facets_config = FacetsConfig()
facets_config.setHierarchical("author_type",True)
facets_config.setMultiValued("author_type",True)

titles = ["the quick fox","the quick brown","the quick brown fox"]
author_types = [[["a","book"],["b","book"]],[["b","cd"],["c","cd"]],[["c","dvd"],["d","dvd"]]]

taxo_store = SimpleFSDirectory(Paths.get(taxo_path))
taxo_writer = DirectoryTaxonomyWriter(taxo_store,IndexWriterConfig.OpenMode.CREATE)
index_store = SimpleFSDirectory(Paths.get(index_path))
index_writer = IndexWriter(index_store,config)

for idx,title in enumerate(titles):
    doc = Document()
    doc.add(Field("title",title,TextField.TYPE_STORED))
    for author_type in author_types[idx]:
        doc.add(FacetField("author_type",author_type))
    index_writer.addDocument(facets_config.build(taxo_writer,doc))
    
taxo_writer.close()
taxo_store.close()

index_writer.close()
index_store.close()

当我使用查询进行搜索时，我想按标题类型查看每个作者的标题计数。

#Searching
taxo_store = SimpleFSDirectory(Paths.get(taxo_path))
taxo_reader = DirectoryTaxonomyReader(taxo_store)

index_store = SimpleFSDirectory(Paths.get(index_path))
index_reader = DirectoryReader.open(store)

searcher = IndexSearcher(reader)

query = QueryParser("title",EnglishAnalyzer())
parsed_query = query.parse("""(title: quick)""")
drill_down_query =  DrillDownQuery(facets_config,parsed_query)

facets_collector = FacetsCollector()
FacetsCollector.search(searcher,drill_down_query,100,facets_collector)

facets = FastTaxonomyFacetCounts(taxo_reader,facets_config,facets_collector)
results = facets.getTopChildren(1000,"author_type")

results 包含以下内容。

<FacetResult: dim=author_type path=[] value=-1 childCount=4
  b (2)
  c (2)
  a (1)
  d (1)
>

我希望看到 b (2) 实际上由一本书和一张 cd 组成，而 c (2) 是一张 cd 和一张 dvd。

我如何实现这一目标？任何帮助都会很棒，谢谢！

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）

faceted-searchlucenelucenepylucene