Python插入Elasticsearch操作方法解析

Python  /  管理员 发布于 3年前   198

这篇文章主要介绍了Python插入Elasticsearch操作方法解析,文中通过示例代码介绍的非常详细，对大家的学习或者工作具有一定的参考学习价值,需要的朋友可以参考下

在用scrapy做爬虫的时候，需要将数据存入的es中。网上找了两种方法，照葫芦画瓢也能出来，暂记下来：

首先安装了es，版本是5.6.1的较早版本

用pip安装与es版本相对的es相关包

pip install elasticsearch-dsl==5.1.0

方法一：

以下是pipelines.py模块的完整代码

# -*- coding: utf-8 -*-# Define your item pipelines here## Don't forget to add your pipeline to the ITEM_PIPELINES setting# See: https://docs.scrapy.org/en/latest/topics/item-pipeline.htmlimport chardetclass SinafinancespiderPipeline(object):  def process_item(self, item, spider):    return item# 写入到es中,需要在settings中启用这个类 ExchangeratespiderESPipeline# 需要安装pip install elasticsearch-dsl==5.1.0 注意与es版本需要对应from elasticsearch_dsl import Date,Nested,Boolean,analyzer,Completion,Keyword,Text,Integer,DocTypefrom elasticsearch_dsl.connections import connectionsconnections.create_connection(hosts=['192.168.52.138'])from elasticsearch import Elasticsearches = Elasticsearch()class AticleType(DocType):  page_from = Keyword()  # domain报错  domain=Keyword()  cra_url=Keyword()  spider = Keyword()  cra_time = Keyword()  page_release_time = Keyword()  page_title = Text(analyzer="ik_max_word")  page_content = Text(analyzer="ik_max_word")class Meta:    index = "scrapy"    doc_type = "sinafinance"    # 以下settings和mappings都没起作用，暂且记下    settings = {      "number_of_shards": 3,    }    mappings = {      '_id':{'path':'cra_url'}    }class ExchangeratespiderESPipeline(DocType):  from elasticsearch5 import Elasticsearch  ES = ['192.168.52.138:9200']  es = Elasticsearch(ES,sniff_on_start=True)  def process_item(self, item, spider):    spider.logger.info("-----enter into insert ES")    article = AticleType()    article.page_from=item['page_from']    article.domain=item['domain']    article.cra_url =item['cra_url']    article.spider =item['spider']    article.cra_time =item['cra_time']    article.page_release_time =item['page_release_time']    article.page_title =item['page_title']    article.page_content =item['page_content']    article.save()    return item

以上方法能将数据写入es，但是如果重复爬取的话，会重复插入数据，因为 主键 ”_id” 是ES自己产生的，找不到自定义_id的入口。于是放弃。

方法二：实现自定义主键写入，覆盖插入

# -*- coding: utf-8 -*-# Define your item pipelines here## Don't forget to add your pipeline to the ITEM_PIPELINES setting# See: https://docs.scrapy.org/en/latest/topics/item-pipeline.htmlfrom elasticsearch5 import Elasticsearchclass SinafinancespiderPipeline(object):  def process_item(self, item, spider):    return item# 写入到es中,需要在settings中启用这个类 ExchangeratespiderESPipeline# 需要安装pip install elasticsearch-dsl==5.1.0 注意与es版本需要对应class SinafinancespiderESPipeline():  def __init__(self):    self.ES = ['192.168.52.138:9200']    # 创建es客户端    self.es = Elasticsearch(      self.ES,      # 启动前嗅探es集群服务器      sniff_on_start=True,      # es集群服务器结点连接异常时是否刷新es结点信息      sniff_on_connection_fail=True,      # 每60秒刷新节点信息      sniffer_timeout=60    )  def process_item(self, item, spider):    spider.logger.info("-----enter into insert ES")    doc = {      'page_from': item['page_from'],      'domain': item['domain'],      'spider': item['spider'],      'page_release_time': item['page_release_time'],      'page_title': item['page_title'],      'page_content': item['page_content'],      'cra_url': item['cra_url'],      'cra_time': item['cra_time']    }    self.es.index(index='scrapy', doc_type='sinafinance', body=doc, id=item['cra_url'])    return item

搜索数据的方法：

# 字典形式设置bodyquery = { 'query': {  'bool': {   'must': [    {'match': {'_all': 'python web'}}   ],   'filter': [    {'term': {'status': 2}}   ]  } }}ret = es.search(index='articles', doc_type='article', body=query)# 查询数据data = es.search(index='articles', doc_type='article', body=body)print(data)# 增加es.index(...)# 修改es.update(...)# 删除es.delete()

完成后

在settings.py模块中注册自定义的类

ITEM_PIPELINES = {  # 'sinafinancespider.pipelines.SinafinancespiderPipeline': 300,  'sinafinancespider.pipelines.SinafinancespiderESPipeline': 300,}

以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持。