Use es to realize epidemic map or take out order function (including code and data)

0、 introduction

The current situation , We use the relevant functions of the epidemic map almost every day , The core function is geographic location retrieval , Use ES Can be very easy to achieve , Let me briefly talk about how to design and implement several core functions .

This article does not cover the preceding paragraph UI Design and implementation of and client call code , If everyone needs , You can leave me a message , If the demand is relatively large, you can consider making a complete project open source , Like it and support it

PS： To study this article, we need to know something about Elasticsearch Location search Have a certain understanding of the basic functions of , Recommended reading ：todo

The data used in this case , You can download it at the end of the article .

1、 Function module

The following figure shows the basic block of the epidemic map , The purpose of this paper is to apply geographical location retrieval to project landing , Therefore, business modules unrelated to core functions will not be listed . The main functions are shown in the figure

The main functions include ：

• Administrative region information entry
• Institutional information
• Health code status reporting
• Search for nearby nucleic acid testing institutions （ The hospital ）
• Check the number of confirmed cases in a certain area （ Within the Administrative Region ）
• Query the distribution of confirmed persons in a certain area
• Query the administrative region of a hospital
• Count the number of confirmed cases in cities in various provinces

2、 Index structure design

2.1 Regional index

The linkage between provinces and cities is very common in many scenes , Its table structure is also very simple , Through one pid perhaps pcode You can save the cascading relationship .

But when searching geographical location , There is only Spatial relations There is no such thing as logical relationship , Different geographical locations do not save their relationships through associated fields , So when designing the index , Whether the province or city adopts structured logical storage , It depends on the business . Only from the geographical location retrieval analysis , No matter the province 、 City 、 Area or street , They all belong to Geometry , So there is no need for structured storage , Every _doc Save one geo_shape That is a reasonable way .

I still retain the logical relationship between provinces and cities , But this is not useful for the functions to be realized in this paper . It is recommended that each province be included in the index 、 City 、 region 、 Create a separate document for the street , Use polygon Storage .

ES Support only basic geometry , Irregular geometry is not supported , How to store such graphics in provinces and regions ？

In fact, irregular geometry can be regarded as polygons with many sides , There are enough sides , It can accurately describe an irregular polygon . This is actually the same as when you play games , Some scenes that need to display curves , In fact, it is described by countless polygons , When the number of sides is enough , You can't see that he is a polygon .

Indexes mapping

PUT area
{
    
  "mappings": {
    
    "properties": {
    
      "location": {
    
        "type": "geo_shape"
      },
      "city": {
    
      "properties": {
    
          "location": {
    
            "type": "geo_shape"
          },
          "district": {
    
            "properties": {
    
              "location": {
    
                "type": "geo_shape"
              }
            }
          }
        }
      }
    }
  }
}

2.2 Institutional index

The organization can be ： The hospital 、 School 、 bridge 、 company 、 Any unit such as tunnel , The specific content depends on what you want to search when you do the function , Generally, the mechanism is a point coordinate , namely ：geo_shape：point. To simplify the code , I only store one hospital information here , It can also be understood as the network of nucleic acid testing institutions .

Indexes mapping as follows ：

PUT hospital
{
    
  "mappings": {
    
    "properties": {
    
      "properties" : {
    
        "address" : {
    
          "type" : "text",
          "fields" : {
    
            "keyword" : {
    
              "type" : "keyword",
              "ignore_above" : 256
            }
          }
        },
        "district" : {
    
          "type" : "text",
          "fields" : {
    
            "keyword" : {
    
              "type" : "keyword",
              "ignore_above" : 256
            }
          }
        },
        "location" : {
    
          "type" : "geo_shape"
        },
        "lv" : {
    
          "type" : "text",
          "fields" : {
    
            "keyword" : {
    
              "type" : "keyword",
              "ignore_above" : 256
            }
          }
        },
        "name" : {
    
          "type" : "text",
          "fields" : {
    
            "keyword" : {
    
              "type" : "keyword",
              "ignore_above" : 256
            }
          }
        }
      }
    }
  }
}

2.3 Personnel information index （ Report information ）

Personnel information reporting actually includes health codes 、 Trip code （ Trajectory ）、 Nucleic acid test report 、 Health status and other information , Information reporting basically takes place in the access control code scanning registration 、 The time when the nucleic acid test results occur , The travel trajectory reporting time can be calculated and reported by changing the base station connected to the mobile phone . In order to simplify the function , Only one status field is reserved for our personnel's health information ： And health or infection （ Diagnosis ）.

Indexes mapping：

PUT case_person
{
    
  "mappings": {
    
    "properties": {
    
      "date": {
    
        "type": "date"
      },
      "location": {
    
        "type": "geo_shape"
      },
      "name": {
    
        "type": "text",
        "fields": {
    
          "keyword": {
    
            "type": "keyword",
            "ignore_above": 256
          }
        }
      },
      "status": {
    
        "type": "text",
        "fields": {
    
          "keyword": {
    
            "type": "keyword",
            "ignore_above": 256
          }
        }
      }
    }
  }
}

3、 Realization

3.1 Statistics of new items on the current day

Here is just a simple statistic , No longer distinguish between native 、 foreign 、 Asymptomatic and other business fields , Statistics of new items on the current day , It means satisfaction

• The time of diagnosis is the same day
• The status is confirmed

All personnel information of two conditions , The code is as follows ：

GET case_person/_search
{
    
  "size": 0, 
  "query": {
    
    "bool": {
    
      "must": [
        {
    "range": {
    
          "date": {
    
            "gte": "now/d"
          }
        }},
        {
    
          "term": {
    
            "status.keyword": {
    
              "value": " Diagnosis "
            }
          }
        }
      ]
    }
  },
  "aggs": {
    
    "today_add": {
    
      "value_count": {
    
        "field": "status.keyword"
      }
    }
  }
}

3.2 Check the surrounding confirmed cases

Inquire about the confirmed cases in the surrounding areas or the confirmed cases in the Administrative Region , It can be applied to businesses such as judging whether the region belongs to medium and high risk areas , It's also very simple to implement . Just provide two conditions

• Current longitude and latitude coordinates
• Search radius

If it's in APP in , The current coordinate is directly called SDK The location interface provided in , If it is for testing , The coordinates can be obtained by selecting points on the map through the open interface provided by Baidu map . The operation steps are: ：

Scrolling pages , Find the development documentation section of the page , Click the coordinate picker

here , Real time coordinates will be displayed where the mouse passes in the map ：

You can choose your favorite coordinates in the map , Simulate your current position . For example, the longitude and latitude of my current position is ：

• “lon”:116.238334,
• “lat”:39.900112

Take Yonghui supermarket as the center of the circle , The radius is three kilometers , Search for Amusement park... Octagonal 、 Babaoshan 、 Yuquan Road 、 Wukesong has four subway stations （ The subway station here can be regarded as a confirmed case ）, That is, there are four confirmed cases within three kilometers .

Such a search , It can also be used to search nearby nucleic acid detection institutions ： The code is as follows ：

3.2 Search the specified units in the specified area

Search for units in the specified area , You can search various units , You can also search various administrative regions .
For example, search Beijing - All nucleic acid testing institutions in Haidian District ：

The code is as follows ：

GET province_bak/_search
{
    
  "query": {
    
    "term": {
    
      "name.keyword": {
    
        "value": " Haidian District "
      }
    }
  }
}
GET hospital/_search
{
    
   "_source": {
    "include":["name","district"]},
  "query": {
    
    "geo_shape": {
    
      "location": {
    
        "indexed_shape": {
    
          "index": "province_bak",
          "id": "110108",
          "path": "location"
        },
        "relation": "within"
      }
    }
  }
}

Index data download ：https://t.zsxq.com/037YJUJia