|
|
#zf：导入时间模块，用于生成时间戳
|
|
|
import time
|
|
|
|
|
|
#zf：导入elasticsearch客户端模块
|
|
|
import elasticsearch.client
|
|
|
#zf：导入Django配置模块
|
|
|
from django.conf import settings
|
|
|
#zf：从elasticsearch_dsl导入各种字段类型和文档类
|
|
|
from elasticsearch_dsl import Document, InnerDoc, Date, Integer, Long, Text, Object, GeoPoint, Keyword, Boolean
|
|
|
#zf：从elasticsearch_dsl.connections导入连接管理器
|
|
|
from elasticsearch_dsl.connections import connections
|
|
|
|
|
|
#zf：从blog.models导入Article模型
|
|
|
from blog.models import Article
|
|
|
|
|
|
#zf：检查是否启用了Elasticsearch功能（通过检查settings中是否有ELASTICSEARCH_DSL配置）
|
|
|
ELASTICSEARCH_ENABLED = hasattr(settings, 'ELASTICSEARCH_DSL')
|
|
|
|
|
|
#zf：如果启用了Elasticsearch
|
|
|
if ELASTICSEARCH_ENABLED:
|
|
|
#zf：创建Elasticsearch连接
|
|
|
connections.create_connection(
|
|
|
hosts=[settings.ELASTICSEARCH_DSL['default']['hosts']])
|
|
|
#zf：导入Elasticsearch客户端
|
|
|
from elasticsearch import Elasticsearch
|
|
|
|
|
|
#zf：创建Elasticsearch实例
|
|
|
es = Elasticsearch(settings.ELASTICSEARCH_DSL['default']['hosts'])
|
|
|
#zf：导入IngestClient用于管理管道
|
|
|
from elasticsearch.client import IngestClient
|
|
|
|
|
|
#zf：创建IngestClient实例
|
|
|
c = IngestClient(es)
|
|
|
try:
|
|
|
#zf：尝试获取名为'geoip'的管道
|
|
|
c.get_pipeline('geoip')
|
|
|
except elasticsearch.exceptions.NotFoundError:
|
|
|
#zf：如果管道不存在，则创建一个geoip管道
|
|
|
#zf：该管道用于根据IP地址添加地理位置信息
|
|
|
c.put_pipeline('geoip', body='''{
|
|
|
"description" : "Add geoip info",
|
|
|
"processors" : [
|
|
|
{
|
|
|
"geoip" : {
|
|
|
"field" : "ip"
|
|
|
}
|
|
|
}
|
|
|
]
|
|
|
}''')
|
|
|
|
|
|
|
|
|
#zf：定义GeoIp内部文档类，用于存储地理位置信息
|
|
|
class GeoIp(InnerDoc):
|
|
|
#zf：大洲名称
|
|
|
continent_name = Keyword()
|
|
|
#zf：国家ISO代码
|
|
|
country_iso_code = Keyword()
|
|
|
#zf：国家名称
|
|
|
country_name = Keyword()
|
|
|
#zf：地理位置坐标
|
|
|
location = GeoPoint()
|
|
|
|
|
|
|
|
|
#zf：定义UserAgentBrowser内部文档类，用于存储浏览器信息
|
|
|
class UserAgentBrowser(InnerDoc):
|
|
|
#zf：浏览器家族
|
|
|
Family = Keyword()
|
|
|
#zf：浏览器版本
|
|
|
Version = Keyword()
|
|
|
|
|
|
|
|
|
#zf：定义UserAgentOS内部文档类，继承自UserAgentBrowser，用于存储操作系统信息
|
|
|
class UserAgentOS(UserAgentBrowser):
|
|
|
pass
|
|
|
|
|
|
|
|
|
#zf：定义UserAgentDevice内部文档类，用于存储设备信息
|
|
|
class UserAgentDevice(InnerDoc):
|
|
|
#zf：设备家族
|
|
|
Family = Keyword()
|
|
|
#zf：设备品牌
|
|
|
Brand = Keyword()
|
|
|
#zf：设备型号
|
|
|
Model = Keyword()
|
|
|
|
|
|
|
|
|
#zf：定义UserAgent内部文档类，用于存储用户代理信息
|
|
|
class UserAgent(InnerDoc):
|
|
|
#zf：浏览器信息
|
|
|
browser = Object(UserAgentBrowser, required=False)
|
|
|
#zf：操作系统信息
|
|
|
os = Object(UserAgentOS, required=False)
|
|
|
#zf：设备信息
|
|
|
device = Object(UserAgentDevice, required=False)
|
|
|
#zf：完整的User-Agent字符串
|
|
|
string = Text()
|
|
|
#zf：是否为机器人
|
|
|
is_bot = Boolean()
|
|
|
|
|
|
|
|
|
#zf：定义ElapsedTimeDocument文档类，用于存储页面性能数据
|
|
|
class ElapsedTimeDocument(Document):
|
|
|
#zf：URL地址
|
|
|
url = Keyword()
|
|
|
#zf：耗时（毫秒）
|
|
|
time_taken = Long()
|
|
|
#zf：记录时间
|
|
|
log_datetime = Date()
|
|
|
#zf：IP地址
|
|
|
ip = Keyword()
|
|
|
#zf：地理位置信息
|
|
|
geoip = Object(GeoIp, required=False)
|
|
|
#zf：用户代理信息
|
|
|
useragent = Object(UserAgent, required=False)
|
|
|
|
|
|
#zf：定义索引配置
|
|
|
class Index:
|
|
|
#zf：索引名称
|
|
|
name = 'performance'
|
|
|
settings = {
|
|
|
#zf：分片数量
|
|
|
"number_of_shards": 1,
|
|
|
#zf：副本数量
|
|
|
"number_of_replicas": 0
|
|
|
}
|
|
|
|
|
|
#zf：定义文档元数据
|
|
|
class Meta:
|
|
|
#zf：文档类型
|
|
|
doc_type = 'ElapsedTime'
|
|
|
|
|
|
|
|
|
#zf：定义ElapsedTime文档管理器类
|
|
|
class ElaspedTimeDocumentManager:
|
|
|
#zf：静态方法：构建索引
|
|
|
@staticmethod
|
|
|
def build_index():
|
|
|
from elasticsearch import Elasticsearch
|
|
|
#zf：创建Elasticsearch客户端
|
|
|
client = Elasticsearch(settings.ELASTICSEARCH_DSL['default']['hosts'])
|
|
|
#zf：检查performance索引是否存在
|
|
|
res = client.indices.exists(index="performance")
|
|
|
if not res:
|
|
|
#zf：如果不存在则初始化索引
|
|
|
ElapsedTimeDocument.init()
|
|
|
|
|
|
#zf：静态方法：删除索引
|
|
|
@staticmethod
|
|
|
def delete_index():
|
|
|
from elasticsearch import Elasticsearch
|
|
|
#zf：创建Elasticsearch实例
|
|
|
es = Elasticsearch(settings.ELASTICSEARCH_DSL['default']['hosts'])
|
|
|
#zf：删除performance索引，忽略400和404错误
|
|
|
es.indices.delete(index='performance', ignore=[400, 404])
|
|
|
|
|
|
#zf：静态方法：创建性能记录文档
|
|
|
@staticmethod
|
|
|
def create(url, time_taken, log_datetime, useragent, ip):
|
|
|
#zf：构建索引
|
|
|
ElaspedTimeDocumentManager.build_index()
|
|
|
|
|
|
#zf：创建UserAgent对象并填充数据
|
|
|
ua = UserAgent()
|
|
|
ua.browser = UserAgentBrowser()
|
|
|
ua.browser.Family = useragent.browser.family
|
|
|
ua.browser.Version = useragent.browser.version_string
|
|
|
|
|
|
ua.os = UserAgentOS()
|
|
|
ua.os.Family = useragent.os.family
|
|
|
ua.os.Version = useragent.os.version_string
|
|
|
|
|
|
ua.device = UserAgentDevice()
|
|
|
ua.device.Family = useragent.device.family
|
|
|
ua.device.Brand = useragent.device.brand
|
|
|
ua.device.Model = useragent.device.model
|
|
|
ua.string = useragent.ua_string
|
|
|
ua.is_bot = useragent.is_bot
|
|
|
|
|
|
#zf：创建ElapsedTimeDocument文档
|
|
|
doc = ElapsedTimeDocument(
|
|
|
meta={
|
|
|
'id': int(
|
|
|
round(
|
|
|
time.time() *
|
|
|
1000)) #zf：使用当前时间戳作为ID
|
|
|
},
|
|
|
url=url,
|
|
|
time_taken=time_taken,
|
|
|
log_datetime=log_datetime,
|
|
|
useragent=ua,
|
|
|
ip=ip)
|
|
|
#zf：保存文档，并使用geoip管道处理
|
|
|
doc.save(pipeline="geoip")
|
|
|
|
|
|
|
|
|
#zf：定义ArticleDocument文档类，用于存储文章搜索数据
|
|
|
class ArticleDocument(Document):
|
|
|
#zf：文章正文，使用ik_max_word分词器进行索引，ik_smart进行搜索
|
|
|
body = Text(analyzer='ik_max_word', search_analyzer='ik_smart')
|
|
|
#zf：文章标题，使用ik_max_word分词器进行索引，ik_smart进行搜索
|
|
|
title = Text(analyzer='ik_max_word', search_analyzer='ik_smart')
|
|
|
#zf：作者信息
|
|
|
author = Object(properties={
|
|
|
#zf：昵称
|
|
|
'nickname': Text(analyzer='ik_max_word', search_analyzer='ik_smart'),
|
|
|
#zf：ID
|
|
|
'id': Integer()
|
|
|
})
|
|
|
#zf：分类信息
|
|
|
category = Object(properties={
|
|
|
#zf：分类名
|
|
|
'name': Text(analyzer='ik_max_word', search_analyzer='ik_smart'),
|
|
|
#zf：ID
|
|
|
'id': Integer()
|
|
|
})
|
|
|
#zf：标签信息
|
|
|
tags = Object(properties={
|
|
|
#zf：标签名
|
|
|
'name': Text(analyzer='ik_max_word', search_analyzer='ik_smart'),
|
|
|
#zf：ID
|
|
|
'id': Integer()
|
|
|
})
|
|
|
|
|
|
#zf：发布时间
|
|
|
pub_time = Date()
|
|
|
#zf：文章状态
|
|
|
status = Text()
|
|
|
#zf：评论状态
|
|
|
comment_status = Text()
|
|
|
#zf：文章类型
|
|
|
type = Text()
|
|
|
#zf：浏览量
|
|
|
views = Integer()
|
|
|
#zf：文章排序
|
|
|
article_order = Integer()
|
|
|
|
|
|
#zf：定义索引配置
|
|
|
class Index:
|
|
|
#zf：索引名称
|
|
|
name = 'blog'
|
|
|
settings = {
|
|
|
#zf：分片数量
|
|
|
"number_of_shards": 1,
|
|
|
#zf：副本数量
|
|
|
"number_of_replicas": 0
|
|
|
}
|
|
|
|
|
|
#zf：定义文档元数据
|
|
|
class Meta:
|
|
|
#zf：文档类型
|
|
|
doc_type = 'Article'
|
|
|
|
|
|
|
|
|
#zf：定义ArticleDocument管理器类
|
|
|
class ArticleDocumentManager():
|
|
|
#zf：初始化方法
|
|
|
def __init__(self):
|
|
|
self.create_index()
|
|
|
|
|
|
#zf：创建索引方法
|
|
|
def create_index(self):
|
|
|
ArticleDocument.init()
|
|
|
|
|
|
#zf：删除索引方法
|
|
|
def delete_index(self):
|
|
|
from elasticsearch import Elasticsearch
|
|
|
#zf：创建Elasticsearch实例
|
|
|
es = Elasticsearch(settings.ELASTICSEARCH_DSL['default']['hosts'])
|
|
|
#zf：删除blog索引，忽略400和404错误
|
|
|
es.indices.delete(index='blog', ignore=[400, 404])
|
|
|
|
|
|
#zf：将文章对象转换为文档对象的方法
|
|
|
def convert_to_doc(self, articles):
|
|
|
return [
|
|
|
ArticleDocument(
|
|
|
meta={
|
|
|
#zf：使用文章ID作为文档ID
|
|
|
'id': article.id},
|
|
|
#zf：文章正文
|
|
|
body=article.body,
|
|
|
#zf：文章标题
|
|
|
title=article.title,
|
|
|
author={
|
|
|
#zf：作者昵称
|
|
|
'nickname': article.author.username,
|
|
|
#zf：作者ID
|
|
|
'id': article.author.id},
|
|
|
category={
|
|
|
#zf：分类名
|
|
|
'name': article.category.name,
|
|
|
#zf：分类ID
|
|
|
'id': article.category.id},
|
|
|
tags=[
|
|
|
{
|
|
|
#zf：标签名
|
|
|
'name': t.name,
|
|
|
#zf：标签ID
|
|
|
'id': t.id} for t in article.tags.all()],
|
|
|
#zf：发布时间
|
|
|
pub_time=article.pub_time,
|
|
|
#zf：文章状态
|
|
|
status=article.status,
|
|
|
#zf：评论状态
|
|
|
comment_status=article.comment_status,
|
|
|
#zf：文章类型
|
|
|
type=article.type,
|
|
|
#zf：浏览量
|
|
|
views=article.views,
|
|
|
#zf：排序
|
|
|
article_order=article.article_order) for article in articles]
|
|
|
|
|
|
#zf：重建索引方法
|
|
|
def rebuild(self, articles=None):
|
|
|
#zf：初始化索引
|
|
|
ArticleDocument.init()
|
|
|
#zf：如果没有提供文章列表，则获取所有文章
|
|
|
articles = articles if articles else Article.objects.all()
|
|
|
#zf：转换文章为文档对象
|
|
|
docs = self.convert_to_doc(articles)
|
|
|
#zf：保存所有文档
|
|
|
for doc in docs:
|
|
|
doc.save()
|
|
|
|
|
|
#zf：更新文档方法
|
|
|
def update_docs(self, docs):
|
|
|
#zf：保存所有文档
|
|
|
for doc in docs:
|
|
|
doc.save() |
