ElasticSearch学习笔记

Dockerfile 构建（安装ik）

FROM elasticsearch:7.13.2
ARG ANALYSIS_IK_URL=https://github.com/medcl/elasticsearch-analysis-ik/releases/download/v7.13.2/elasticsearch-analysis-ik-7.13.2.zip
ENV ANALYSIS_IK_URL=${ANALYSIS_IK_URL}
RUN yes|/usr/share/elasticsearch/bin/elasticsearch-plugin install ${ANALYSIS_IK_URL}
RUN echo 'xpack.security.enabled: true' >> /usr/share/elasticsearch/config/elasticsearch.yml
RUN echo 'xpack.security.transport.ssl.enabled: true' >> /usr/share/elasticsearch/config/elasticsearch.yml

Elasticsearch 设置密码

# 自由设置密码
/usr/share/elasticsearch/bin/elasticsearch-setup-passwords interactive
# 自动生成密码
/usr/share/elasticsearch/bin/elasticsearch-setup-passwords auto

Elasticsearch DSL 之 Search

简单示例

from elasticsearch import Elasticsearch
from elasticsearch_dsl import Search, Q, A
if __name__ == '__main__':
    es = Elasticsearch()
    es_index = "document"
    s = Search().index(es_index).using(es)
    s = s.query("has_child", type='author', query=Q("match_all"))
    # counts = s.count()
    # d = s.to_dict()
    es_response = s.execute()

获取查询总数

# 获取查询总数
counts = s.count()

匹配

 # 匹配
 s = s.query("match", title='Interface')

多字段匹配

# 多字段匹配
s = s.query("multi_match", fields=['title', 'source'], query='Interface')

精确过滤

# 精确过滤
s = s.filter("term", is_show=1)

排序

# 排序
s = s.sort('-view_times')

高亮

# 高亮
s = s.highlight('title')

指定返回字段

# 指定返回字段
s = s.source(includes=['title', 'source'], excludes=["child_.*"])

分页

# 分页
s = s[0:10]

求和

# 求和
s.aggs.metric('sum', 'sum', field='view_times')

求平均数

# 求平均数
s.aggs.metric('avg', 'avg', field='view_times')

按范围分组统计

# 按范围分组统计
s.aggs.bucket('range', 'range', field='view_times', ranges=[
    {"to": 100},
    {"from": 100, "to": 5000},
    {"from": 5000}
])

分组统计

# 分组统计
s.aggs.bucket(
    "group", A('terms', field='keywords', size=10000, order={'_count': 'desc'})
).metric(
    # 分组后再求和
    'sum', 'sum', field='view_times'
)
# 分组后再求平均数
s.aggs['group'].metric('avg', 'avg', field='view_times')

Nested Query(嵌套查询)

s = s.query(
        "nested",
        path="fee",
        query={
            "bool": {
                "filter": [
                    {
                        "term": {
                            "fee.title": "测试"
                        }
                    }
                ]
            }
        }
    )

Nested Aggregation（嵌套聚合）

s.aggs.bucket(
        name="apply_fee_bucket",
        agg_type="nested",
        path="apply_fee",
        aggs={
            "min_amount": {"min": {"field": "apply_fee.amount"}},
            "max_amount": {"max": {"field": "apply_fee.amount"}},
            "group_by": {"terms": {"field": "apply_fee.true_title.raw"}}
        }
    )

Elasticsearch DSL 之 UpdateByQuery

直接赋值

from elasticsearch import Elasticsearch
from elasticsearch_dsl import UpdateByQuery
es_response = UpdateByQuery().using(es).index( "document").filter(
    "term", _routing=es_doc_id
).filter(
    "term", child_index=child_index
).script( # 修改值
    inline="ctx._source.child_name = params.child_name",
    params={
        "child_name": "张三"
    },
    lang="painless"
).execute()

在原数据基础上更新

from elasticsearch import Elasticsearch
from elasticsearch_dsl import UpdateByQuery
es_response = UpdateByQuery().using(es).index( "document").filter(
    "term", _routing=es_doc_id
).filter(
    "term", child_index=child_index
).script( # += 操作
    inline="ctx._source.view_times += params.num",
    params={"num": 3},
    upsert={"view_times": 0},
    lang="painless"
).script( # ++ 操作
    inline="ctx._source.download_times++",
    upsert={"download_times": 0},
    lang="painless"
).execute()