上一篇笔记仅是记录了一下简单的关联查询，根据笔记中的场景：将某一车辆关联的耗损记录全部放在了一个字段当中。不知道现在中有没有这种场景，我们的应用中没有类似的场景，可能我们更关注的是某车辆的总耗损金额和总营收金额、总里程数等，因此这篇笔记记录一下怎么得到这些数据。要想获取这些记录我们首先想到的就是先按照车辆ID分组之后再对指定字段求和，可能是先入为主的原因（有了上一篇笔记），我没有按照传统的思路走下去，而是想着看看能不能在上一篇笔记的基础之上得到想要的结果。
我们沿着这条路在分析一下，车辆相关的耗损和营收都可以放到一个字段中。也就是说我们不用分组了，车辆相关的耗损和营收记录都冗余到了车辆信息记录中，那么接下来就是对数组中的某一字段或者多个进行求和了……查了半天并没有找到类似$sum这样的聚合管道可以对数组中的字段求和，不过得益于mongodb管道的强大，用另外一个管道间接实现了这个功能。
db.getCollection('FormInstace').aggregate([
{
$match: {
"_id": { $in: ["1","2","3"] },
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{
$lookup:{
from:"FormInstace",
let: { tempMainRelationKey: "$_id" },
pipeline:[
{
$match: {
FormId: "507048044944692000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{ $match:
{ $expr:
{ $eq: [ "$1572493552001.id","$$tempMainRelationKey" ] }
}
},
],
as:'carWastage'
}
},
{
$lookup:{
from:"FormInstace",
let: { tempMainRelationKey: "$_id" },
pipeline:[
{
$match: {
FormId: "507048044944693000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{ $match:
{ $expr:
{ $eq: [ "$1572493553001.id","$$tempMainRelationKey" ] }
}
},
],
as:'carRevenue'
}
},
{
$addFields: {
carWastageStatistics: {
$reduce: {
input: "$carWastage",
initialValue: {wastageMoney:0},
in: {
wastageMoney:{ $add : ["$$value.wastageMoney", "$$this.1572493552005"] }
}
}
}
}
},
{
$addFields: {
carRevenueStatistics: {
$reduce: {
input: "$carRevenue",
initialValue: {revenueMoney:0,mileage:0},
in: {
revenueMoney:{$add : ["$$value.revenueMoney", "$$this.1572493553005"]},
mileage:{$add : ["$$value.mileage", "$$this.1572493553006"]}
}
}
}
}
}
]);
这次统计查询就查出了某些车辆的总耗损金额和总营收金额、总里程数；只不过将结果放到对象中了（方便统计一个表中的多个字段），但是现在看不到具体的结果，将代码在修改一下，看一下结果
db.getCollection('FormInstace').aggregate([
{
$match: {
"_id": { $in: ["1","2","3"] },
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{
$lookup:{
from:"FormInstace",
let: { tempMainRelationKey: "$_id" },
pipeline:[
{
$match: {
FormId: "507048044944692000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{ $match:
{ $expr:
{ $eq: [ "$1572493552001.id","$$tempMainRelationKey" ] }
}
},
],
as:'carWastage'
}
},
{
$lookup:{
from:"FormInstace",
let: { tempMainRelationKey: "$_id" },
pipeline:[
{
$match: {
FormId: "507048044944693000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{ $match:
{ $expr:
{ $eq: [ "$1572493553001.id","$$tempMainRelationKey" ] }
}
},
],
as:'carRevenue'
}
},
{
$addFields: {
carWastage_Money: {
$reduce: {
input: "$carWastage",
initialValue: 0,
in: { $add : ["$$value", "$$this.1572493552005"] }
}
}
}
},
{
$addFields: {
carRevenue_Money: {
$reduce: {
input: "$carRevenue",
initialValue: 0,
in: { $add : ["$$value", "$$this.1572493553005"] }
}
}
}
},
{
$addFields: {
carRevenue_Mileage: {
$reduce: {
input: "$carRevenue",
initialValue: 0,
in: { $add : ["$$value", "$$this.1572493553006"] }
}
}
}
}
]);
这样就一目了然了。统计结果对不对呢？我们用传统的分组之后再求和在查询一下，将结果做一下对比（顺便记录一下，分组求和）
db.getCollection('FormInstace').aggregate([
{
$match: {
"FormId":"507048044944692000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$match: {
"1572493552001.id": { $in:["1","2","3"]}
}
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{
$group:{
_id: "$1572493552001.id",
wastageMoney: { $sum: "$1572493552005"}
}
},
{ $sort : { _id: 1 } }
]);
db.getCollection('FormInstace').aggregate([
{
$match: {
"FormId":"507048044944693000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$match: {
"1572493553001.id": { $in:["1","2","3"]}
}
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{
$group:{
_id: "$1572493553001.id",
wastageMoney: { $sum: "$1572493553005"},
mileage: { $sum: "$1572493553006"}
}
},
{ $sort : { _id: 1 } }
]);
对比两次查询结果，两次不同的统计方式的结果相同。这样就放心了。好了就到这里了。
********************以下是修改补充部分********************
打脸了，好疼，┭┮﹏┭┮……曾记得以前使用$sum对数组中的某一属性求和时没有得到预期的结果，这篇笔记写完之后，打算在按照常规思路（$unwind、$group、$sum等）在写一种查询方式，但是但是但是测试的时候发现$sum好使了，写放一下代码
db.getCollection('FormInstace').aggregate([
{
$match: {
"_id": { $in: ["1","2","3"] },
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{
$lookup:{
from:"FormInstace",
let: { tempMainRelationKey: "$_id" },
pipeline:[
{
$match: {
FormId: "507048044944692000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{ $match:
{ $expr:
{ $eq: [ "$1572493552001.id","$$tempMainRelationKey" ] }
}
},
],
as:'carWastage'
}
},
{
$lookup:{
from:"FormInstace",
let: { tempMainRelationKey: "$_id" },
pipeline:[
{
$match: {
FormId: "507048044944693000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{ $match:
{ $expr:
{ $eq: [ "$1572493553001.id","$$tempMainRelationKey" ] }
}
},
],
as:'carRevenue'
}
},
{
$addFields: {
"carWastageStatistics.wastageMoney": { $sum: "$carWastage.1572493552005"},
"carRevenueStatistics.revenueMoney": { $sum: "$carRevenue.1572493553005"},
"carRevenueStatistics.mileage": { $sum: "$carRevenue.1572493553006"}
}
}
]);
db.getCollection('FormInstace').aggregate([
{
$match: {
"_id": { $in: ["1","2","3"] },
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{
$lookup:{
from:"FormInstace",
let: { tempMainRelationKey: "$_id" },
pipeline:[
{
$match: {
FormId: "507048044944692000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{ $match:
{ $expr:
{ $eq: [ "$1572493552001.id","$$tempMainRelationKey" ] }
}
},
],
as:'carWastage'
}
},
{
$lookup:{
from:"FormInstace",
let: { tempMainRelationKey: "$_id" },
pipeline:[
{
$match: {
FormId: "507048044944693000",
"FormItems.key": { $ne: null }
}
},
{
$addFields: {
FormValueObj: {
$arrayToObject: {
$map: {
input: "$FormItems",
as: "field",
in: [ "$$field.key", "$$field.value" ]
}
}
}
}
},
{
$replaceRoot: { newRoot: { $mergeObjects: [ "$FormValueObj", "$$ROOT" ] } }
},
{
$project: {
FormItems:0,
FormValueObj:0
}
},
{ $match:
{ $expr:
{ $eq: [ "$1572493553001.id","$$tempMainRelationKey" ] }
}
},
],
as:'carRevenue'
}
},
{
$addFields: {
"carWastage_Money": { $sum: "$carWastage.1572493552005"},
"carRevenue_Money": { $sum: "$carRevenue.1572493553005"},
"carRevenue_Mileage": { $sum: "$carRevenue.1572493553006"},
}
}
]);
下面看一下对比图，看一下改动情况
再看一下，执行结果截图
和上面的结果一模一样……对mongodb管道认识不够深入，应用中有需求就是查一下，太片面了，没有验证的东西千万别乱说，切记切记切记！！！特尴尬，￣□￣｜｜