导航
分群 OpenAPI-DSL 结构说明
最近更新时间：2024.05.29 16:21:47首次发布时间：2023.03.17 18:05:26
1 概述
本文档提供对火山引擎增长分析中规则分群使用的DSL结构详细说明。
2 DSL 结构说明
2.1 从前端获取分群DSL（推荐）
在分群配置页面中配置好规则，打开浏览器控制台；接着点击“预估人数”，查看overview接口的dsl_content字段。（注意：使用dsl_content的时候，最好把show_option节点整体删除）
2.2 整体结构
分群DSL 整体结构跟查询API使用的整体结构是一致的，整体结构请参考： “查询API”->“DSL结构”说明。
分群的区别是在content的queries字段，扩展了相应的含义。整体结构：
{
"version": 3, # DSL的版本，本文档中全部为3
"app_ids": [], # 目标app
"use_app_cloud_id": true, # app_ids参数中是否为app_cloud_id。请设置为true,false仅供内部使用
"periods": [], # 时间，分群这里固定为[]
"content": {}, # 具体的查询内容字段
"option": {}, # json格式数据，请求相关的选项参数，分群不需要该字段
"show_option": {} # json格式数据，在结果中原样返回，分群不需要该字段
}
具体规则定义在 cotent字段中：
{
"content": {
"query_type": "cohort_v3_idlist", # 固定使用这个类型
"queries":[ # 分群规则
[{query1},{query2}],
[{query3},{query4}]
]
"option":{}, # 分群参数
"profile_filters":[], # 固定为[]
"profile_groups":[] # 固定为[]
}
}
2.3 字段详细说明
2.3.1 option 字段
{
"cohort": {
"outer_logic": "and"
}
}
字段 | 取值与含义 | 是否必须 | 例子 |
|---|---|---|---|
outer_logic | 最外层的逻辑关系：and/or | 否 | 默认and |
2.3.2 queries 字段
[
[{query1},{query2}],
[{query3},{query4}]
...
]
2.3.2.1 query 逻辑关系
queries 是一个二维数组，定义了每一个规则的关系。
当 outer_logic 为and 时：内层数组的关系为 or，外层关系为 and。比如：
"queries":[ # 分群规则
[{query1},{query2}],
[{query3},{query4}]
]
"option":{
"cohort": {
"outer_logic": "and"
}
},
那么整体计算关系为：
[query1 or query2] and [query3 or query4]
当 outer_logic 为or 时：内层数组的关系为 and，外层关系为 or。比如：
"queries":[ # 分群规则
[{query1},{query2}],
[{query3},{query4}]
]
"option":{
"cohort": {
"outer_logic": "or"
}
},
那么整体计算关系为：
[query1 and query2] or [query3 and query4]
2.3.2.2 query 结构
整体结构，参考“查询API”->“DSL结构”->“content字段/queries”。
{
"event_name": "any_event",
"event_id":1,
"event_type": "origin",
"show_name": "总次数",
"groups_v2": [],
"filters": [],
"show_label": "A",
"event_indicator": "events",
"measure_info": {},
"condition":{},
"logic": true,
"sequence_first":true,
"next_query_lable":"B",
}
新增字段含义说明：
字段 | 取值与含义 | 是否必须 | 例子 |
|---|---|---|---|
condition | 分群的条件，结构同 DSL “condition” 结构 | 是 | |
logic | true表示用户做过、用户是；false 表示用户没做过、用户不是 | 是 | 注意：当logic为true的时候，并且用户做过，比如选择用户做过事件次数 大于 {n}，n的值一定是一个大于0的值。如果要选择 “用户做过事件A次数=0”，请使用“用户没做过事件A 的条件” |
sequence_first | true表示是“依次做过”，并且是依次做过的第一个事件 | 否 | “依次做过”下是必须的 |
next_query_label | “依次做过”的下一个事件的show_label值 | 否 |
condition 结构：
整体结构，参考“查询API”->“DSL结构”->“content字段/condition”
{
"property_operation": ">",
"property_values": [0],
"period": {}
}
新增字段含义说明：
字段 | 取值与含义 | 是否必须 | 例子 |
|---|---|---|---|
period | 分群条件的时间范围，参考 DSL period结构 | 否 | 选择事件、或者all_value用户属性的时候一定需要时间 |
3 样例
3.1 用户做过
{
"periods": [],
"version": 3,
"use_app_cloud_id": true,
"app_ids": [
164314
],
"content": {
"option": {
"cohort": {
"outer_logic": "and"
}
},
"query_type": "cohort_v3_idlist",
"profile_filters": [],
"profile_groups": [],
"queries": [
[
{
"show_label": "c06bed36-eede-4174-91bf-8f37eac3a8ee",
"event_name": "any_event",
"event_type": "origin",
"logic": true,
"event_indicator": "events",
"measure_info": {},
"condition": {
"property_operation": ">=",
"property_values": [
1
],
"period": {
"granularity": "day",
"type": "past_range",
"spans": [
{
"type": "past",
"past": {
"amount": 7,
"unit": "day"
}
},
{
"type": "past",
"past": {
"amount": 1,
"unit": "day"
}
}
],
"timezone": "Asia/Shanghai",
"week_start": 1
}
},
"filters": [
{
"expression": {
"logic": "and",
"conditions": []
}
}
]
}
]
]
}
}
3.2 用户没做过
{
"periods": [],
"version": 3,
"use_app_cloud_id": true,
"app_ids": [
164314
],
"content": {
"option": {
"cohort": {
"outer_logic": "and"
}
},
"query_type": "cohort_v3_idlist",
"profile_filters": [],
"profile_groups": [],
"queries": [
[
{
"show_label": "c06bed36-eede-4174-91bf-8f37eac3a8ee",
"event_name": "any_event",
"event_type": "origin",
"logic": false,
"event_indicator": "events",
"measure_info": {},
"condition": {
"property_operation": ">=",
"property_values": [
1
],
"period": {
"granularity": "day",
"type": "past_range",
"spans": [
{
"type": "past",
"past": {
"amount": 7,
"unit": "day"
}
},
{
"type": "past",
"past": {
"amount": 1,
"unit": "day"
}
}
],
"timezone": "Asia/Shanghai",
"week_start": 1
}
},
"filters": [
{
"expression": {
"logic": "and",
"conditions": []
}
}
]
}
]
]
}
}
3.3 用户是
{
"periods": [],
"version": 3,
"use_app_cloud_id": true,
"app_ids": [
164314
],
"content": {
"option": {
"cohort": {
"outer_logic": "and"
}
},
"query_type": "cohort_v3_idlist",
"profile_filters": [],
"profile_groups": [],
"queries": [
[
{
"show_label": "c06bed36-eede-4174-91bf-8f37eac3a8ee",
"event_indicator": "events",
"logic": true,
"condition": {
"period": {
"granularity": "day",
"type": "past_range",
"spans": [
{
"type": "past",
"past": {
"amount": 7,
"unit": "day"
}
},
{
"type": "past",
"past": {
"amount": 1,
"unit": "day"
}
}
],
"timezone": "Asia/Shanghai",
"week_start": 1
},
"property_operation": ">",
"property_values": [
0
]
},
"filters": [
{
"expression": {
"logic": "and",
"conditions": [
{
"property_type": "profile",
"property_name": "custom_Age",
"property_compose_type": "origin",
"property_operation": "=",
"property_values": [
35
]
}
]
}
}
]
}
]
]
}
}
3.4 用户不是
{
"periods": [],
"version": 3,
"use_app_cloud_id": true,
"app_ids": [
164314
],
"content": {
"option": {
"cohort": {
"outer_logic": "and"
}
},
"query_type": "cohort_v3_idlist",
"profile_filters": [],
"profile_groups": [],
"queries": [
[
{
"show_label": "c06bed36-eede-4174-91bf-8f37eac3a8ee",
"event_indicator": "events",
"logic": false,
"condition": {
"period": {
"granularity": "day",
"type": "past_range",
"spans": [
{
"type": "past",
"past": {
"amount": 7,
"unit": "day"
}
},
{
"type": "past",
"past": {
"amount": 1,
"unit": "day"
}
}
],
"timezone": "Asia/Shanghai",
"week_start": 1
},
"property_operation": ">",
"property_values": [
0
]
},
"filters": [
{
"expression": {
"logic": "and",
"conditions": [
{
"property_type": "profile",
"property_name": "custom_Age",
"property_compose_type": "origin",
"property_operation": "=",
"property_values": [
35
]
}
]
}
}
]
}
]
]
}
}
3.5 行为序列
{
"periods": [],
"version": 3,
"use_app_cloud_id": true,
"app_ids": [
164314
],
"content": {
"option": {
"cohort": {
"outer_logic": "and"
}
},
"query_type": "cohort_v3_idlist",
"profile_filters": [],
"profile_groups": [],
"queries": [
[
{
"show_label": "1c21dddc-bcc6-4389-90b2-09d6f66099b4",
"event_name": "predefine_pageview",
"event_id": 237198,
"event_type": "origin",
"logic": true,
"event_indicator": "events",
"sequence_first": true,
"next_query_label": "e0edb616-3530-4316-9604-9141099db401",
"condition": {
"period": {
"granularity": "day",
"type": "past_range",
"spans": [
{
"type": "past",
"past": {
"amount": 7,
"unit": "day"
}
},
{
"type": "past",
"past": {
"amount": 1,
"unit": "day"
}
}
],
"timezone": "Asia/Shanghai",
"week_start": 1
},
"property_operation": ">",
"property_values": [
0
]
},
"filters": [
{
"expression": {
"logic": "and",
"conditions": []
}
}
]
},
{
"show_label": "e0edb616-3530-4316-9604-9141099db401",
"event_name": "order_pay_success",
"event_id": 298968,
"event_type": "origin",
"logic": true,
"event_indicator": "events",
"sequence_first": false,
"next_query_label": null,
"condition": {
"period": {
"granularity": "day",
"type": "past_range",
"spans": [
{
"type": "past",
"past": {
"amount": 7,
"unit": "day"
}
},
{
"type": "past",
"past": {
"amount": 1,
"unit": "day"
}
}
],
"timezone": "Asia/Shanghai",
"week_start": 1
},
"property_operation": ">",
"property_values": [
0
]
},
"filters": [
{
"expression": {
"logic": "and",
"conditions": []
}
}
]
}
]
]
}
}