本月我陷入了這個問題。
爲了執行正確的查詢,您需要覆蓋一些乾草堆對象。我發現這篇文章非常有幫助Extending Haystack’s Elasticsearch backend。在開始時相當複雜,但一旦明白它是如何工作的......它的工作原理:-)
博客文章講授如何實現elasticsearch的嵌套查詢......以及...我實現了一個基本的multi_match query。
# -*- coding: utf-8 -*-
from __future__ import absolute_import
from django.conf import settings
from haystack.backends.elasticsearch_backend import (
ElasticsearchSearchBackend, ElasticsearchSearchEngine, ElasticsearchSearchQuery)
from haystack.query import SearchQuerySet
class ElasticsearchEngineBackendCustom(ElasticsearchSearchBackend):
DEFAULT_ANALYZER = "snowball"
def __init__(self, connection_alias, **connection_options):
super(ElasticsearchEngineBackendCustom, self).__init__(connection_alias, **connection_options)
user_settings = getattr(settings, 'ELASTICSEARCH_INDEX_SETTINGS', {})
if user_settings:
setattr(self, 'DEFAULT_SETTINGS', user_settings)
user_analyzer = getattr(settings, 'ELASTICSEARCH_DEFAULT_ANALYZER', '')
if user_analyzer:
setattr(self, 'DEFAULT_ANALYZER', user_analyzer)
def build_search_kwargs(self, query_string, sort_by=None, start_offset=0, end_offset=None,
fields='', highlight=False, facets=None,
date_facets=None, query_facets=None,
narrow_queries=None, spelling_query=None,
within=None, dwithin=None, distance_point=None,
models=None, limit_to_registered_models=None,
result_class=None, multi_match=None):
out = super(ElasticsearchEngineBackendCustom, self).build_search_kwargs(query_string, sort_by, start_offset,
end_offset,
fields, highlight, facets,
date_facets, query_facets,
narrow_queries, spelling_query,
within, dwithin, distance_point,
models, limit_to_registered_models,
result_class)
if multi_match:
out['query'] = {
'multi_match': {
'query': multi_match['query'],
'fields': multi_match['fields'],
'tie_breaker': multi_match['tie_breaker'],
'minimum_should_match': multi_match['minimum_should_match'],
}
}
return out
def build_schema(self, fields):
content_field_name, mapping = super(ElasticsearchEngineBackendCustom, self).build_schema(fields)
for field_name, field_class in fields.items():
field_mapping = mapping[field_class.index_fieldname]
if field_mapping['type'] == 'string' and field_class.indexed:
if not hasattr(field_class, 'facet_for') or field_class.field_type in ('ngram', 'edge_ngram'):
field_mapping['analyzer'] = getattr(field_class, 'analyzer', self.DEFAULT_ANALYZER)
mapping.update({field_class.index_fieldname: field_mapping})
return content_field_name, mapping
def multi_match_run(self, query, fields, minimum_should_match, tie_breaker):
from elasticsearch_dsl import Search
from elasticsearch_dsl.query import MultiMatch
raw = Search().using(self.conn).query(
MultiMatch(query=u'{}'.format(query), fields=fields, minimum_should_match=minimum_should_match, tie_breaker=tie_breaker)
).execute()
return self._process_results(raw)
class ElasticsearchSearchQueryCustom(ElasticsearchSearchQuery):
def multi_match(self, query, fields, minimum_should_match, tie_breaker):
results = self.backend.multi_match_run(query, fields, minimum_should_match, tie_breaker)
self._results = results.get('results', [])
self._hit_count = results.get('hits', 0)
def add_multi_match_query(self, query, fields, minimum_should_match, tie_breaker):
self.multi_match_query = {
'query': query,
'fields': fields,
'minimum_should_match': minimum_should_match,
'tie_breaker': tie_breaker
}
def build_params(self, spelling_query=None, **kwargs):
search_kwargs = super(ElasticsearchSearchQueryCustom, self).build_params(spelling_query, **kwargs)
if self.multi_match_query:
search_kwargs['multi_match'] = self.multi_match_query
return search_kwargs
class ElasticsearchSearchQuerySetCustom(SearchQuerySet):
def multi_match(self, query, fields, minimum_should_match="35%", tie_breaker=0.3):
clone = self._clone()
clone.query.add_multi_match_query(query, fields, minimum_should_match, tie_breaker)
clone.query.multi_match(query, fields, minimum_should_match, tie_breaker)
return clone
class ElasticsearchEngineCustom(ElasticsearchSearchEngine):
backend = ElasticsearchEngineBackendCustom
query = ElasticsearchSearchQueryCustom
正如你可以看到我用elasticsearc-dsl
執行查詢(MultiMatch)和這句話總結了博客文章:ElasticsearchSearchQuerySetCustom().multi_match(...)
呼叫的方式取決於ElasticsearchSearchQueryCustom
取決於ElasticsearchEngineBackendCustom
。
然後把你的設置elasticsearch配置,如:
ELASTICSEARCH_DEFAULT_ANALYZER = 'italian'
ELASTICSEARCH_INDEX_SETTINGS = {
"settings": {[...]}
}
您可以從Language Analyzers
抓住你的ELASTICSEARCH_INDEX_SETTINGS
語言(S)你需要還SearchForm
覆蓋:
# -*- coding: utf-8 -*-
from __future__ import absolute_import
from haystack.forms import SearchForm
from .backend import ElasticsearchSearchQuerySetCustom
class SearchFormCustom(SearchForm):
def search(self):
query = self.searchqueryset.query.clean(self.cleaned_data.get('q'))
if not self.is_valid() or not query:
return self.no_query_found()
sqs = ElasticsearchSearchQuerySetCustom().multi_match(query, ['title^8', 'text^0.5'])
return sqs
字段title
和text
必須在您的索引中並且脫字符字符i s用於在場上進行提升。
你需要重寫URL草堆模式,以便使用自定義窗體:
urlpatterns = patterns(
'search.views',
url('^$', search_view_factory(form_class=SearchFormCustom), name='haystack-search'),
)
就是這樣,HTH :-)
注重不使用result.object.something
但使用而不是你的索引上的字段,例如result.tilte
,因爲result.object.tilte
遇到數據庫!請參閱Haystack Best Practices
非常感謝您的詳細解答,但我轉而使用solr – Ihar