Goal:
This sql and its result should be the same result from mongoDB's query code.
In order words, same result but for mongoDB.
Problem:
How to you make a lookup connection in relation to People and Role in Mongo DB's query code?
Info:
I'm new in mongo DB
SQL code
SELECT
a.*,
'.' AS '.',
b.*,
'.' AS '.',
c.*
FROM
[db1].[dbo].[People_Course_Grade] a
INNER JOIN [db1].[dbo].[People] b on a.PeopleId = b.PeopleId
INNER JOIN [db1].[dbo].[Role] c on b.RoleId = c.RoleId
Json data:
Role:
[{"RoleId":1,"Name":"Student"},{"RoleId":2,"Name":"Teacher"}]
People_Course_Grade:
[{"People_Course_GradeId":1,"PeopleId":1,"CourseId":1},
{"People_Course_GradeId":2,"PeopleId":2,"CourseId":1},
{"People_Course_GradeId":3,"PeopleId":3,"CourseId":2},
{"People_Course_GradeId":4,"PeopleId":1,"CourseId":2}]
Course:
[{"CourseId":1,"Name":"Java"},{"CourseId":2,"Name":"Java II"},
{"CourseId":3,"Name":"Statistik 1"}]
db.People_Course_Grade.aggregate([
{
$lookup:{
from: "People",
localField: "people_id",
foreignField: "_id",
as: "people"
}
},
{ $unwind:"$people" },
{
$project:{
course_id : 1,
people_id : 1,
// grade_id : 1,
Name : "$people.Name",
}
}
]);
You need to start with double $lookup since you have three collections. Then you can use $arrayElemAt to always get single element from lookup's result. To flatten your structure you can use $replaceRoot with $mergeObjects (promotes all the fields from people and course to root level.
db.People_Course_Grade.aggregate([
{
$lookup:{
from: "Role",
localField: "PeopleId",
foreignField: "RoleId",
as: "people"
}
},
{
$lookup:{
from: "Course",
localField: "CourseId",
foreignField: "CourseId",
as: "course"
}
},
{
$replaceRoot: {
newRoot: {
$mergeObjects: [
"$$ROOT",
{ $arrayElemAt: [ "$people", 0 ] },
{ $arrayElemAt: [ "$course", 0 ] },
]
}
}
},
{
$project: {
people: 0,
course: 0
}
}
])
Mongo Playground
$arrayElemAt can always be replaced with $unwind like you tried. You also have a naming conflict on name field so probably you need to run $project to rename one of those fields - otherwise you'll get only one of them in final result.
Related
I have 3 tables sale,company and saleItem with the following relations:
Sale.belongsTo(Company);
Company.hasMany(Sale);
Sale.hasMany(SaleItem, { as: "items" });
SaleItem.belongsTo(Sale);
I want to apply a filter on the company's name I saw that to do that we have to use $ at the start and end but it isnt working. Any ideas where I am going wrong?
When I try to execute the below code I get the error:
SqlError: (conn=201, no: 1054, SQLState: 42S22) Unknown column 'company.name' in 'where clause'
sql: SELECT `sale`.*, `company`.`id` AS `company.id`, `company`.`name` AS `company.name`, `items`.`id` AS `items.id`, `items`.`quantity` AS `items.quantity`, `items`.`price` AS `items.price`, `items`.`description` AS `items.description`, `items`.`margin` AS `items.margin`, `items`.`gst` AS `items.gst`, `items`.`createdAt` AS `items.createdAt`, `items`.`updatedAt` AS `items.updatedAt`, `items`.`saleId` AS `items.saleId`, `items`.`itemId` AS `items.itemId` FROM (SELECT `sale`.`id`, `sale`.`date`, `sale`.`type`, `sale`.`description`, `sale`.`poNumber`, `sale`.`poDate`, `sale`.`paymentType`, `sale`.`gst`, `sale`.`discount`, `sale`.`freight`, `sale`.`status`, `sale`.`saleStatus`, `sale`.`referenceNumber`, `sale`.`ftn`, `sale`.`quotationNumber`, `sale`.`showGST`, `sale`.`invoiceDate`, `sale`.`hasWithholdingTax`, `sale`.`serialNumber`, `sale`.`currency`, `sale`.`createdAt`, `sale`.`updatedAt`, `sale`.`companyId`, `sale`.`customerId` FROM `sale` AS `sale` WHERE `company`.`name` LIKE '%%' AND `sale`.`customerId` = 1 AND `sale`.`status` = 'ACTIVE' ORDER BY `id` DESC LIMIT 0, 15) AS `sale` LEFT OUTER JOIN `company` AS `company` ON `sale`.`companyId` = `company`.`id` LEFT OUTER JOIN `saleItem` AS `items` ON `sale`.`id` = `items`.`saleId` ORDER BY `id` DESC;
It works if I dont include the SaleItem table in query
Here is the code
await Sale.findAndCountAll({
include: [
{
model: Company,
attributes: ["name"],
as: "company",
},
{ model: SaleItem, as: "items" },
],
distinct: true,
where: {
"$company.name$": { [Op.like]: `%${search}%` },
customerId:1,
status: "ACTIVE",
},
})
If you see the generated SQL, company.name WHERE clause is incorrectly added to a subquery, so you can either turn off the subquery or you can add your where option within the include.
Option 1:
await Sale.findAndCountAll({
...,
subQuery: false
})
Option 2:
await Sale.findAndCountAll({
include: [
{
model: Company,
attributes: ["name"],
as: "company",
where: {
name: { [Op.like]: `%${search}%` }
}
},
{ model: SaleItem, as: "items" },
],
distinct: true,
where: {
customerId:1,
status: "ACTIVE",
},
})
I'm new in mongodb. Could you please tell me how to perform join operation in this. I've two collection:
Collection 1 ("user")
{
_id: "d04d53dc-fb88-433e-a1c5-dd41a68d7655",
userName: "XYZ User",
age: 12
}
Collection 2 ("square")
{
_id: "ef6f6ac2-a08a-4f68-a63c-0b4a70285427",
userId: "d04d53dc-fb88-433e-a1c5-dd41a68d7655",
side: 4,
area: 16
}
Now I want to retrieve the data from collection 2 is like this.
Expected output:
{
_id: "ef6f6ac2-a08a-4f68-a63c-0b4a70285427",
userId: "d04d53dc-fb88-433e-a1c5-dd41a68d7655",
userName: "XYZ User",
side: 4,
area: 16
}
Thanks in advance :)
Here's one way to do it.
db.square.aggregate([
{
"$lookup": {
"from": "user",
"localField": "userId",
"foreignField": "_id",
"as": "userDoc"
}
},
{
"$set": {
"userName": {
"$first": "$userDoc.userName"
}
}
},
{ "$unset": "userDoc" }
])
Try it on mongoplayground.net.
You can keep the first documentid (_id) in the second document as userId for refrence and after that, you can use the join feature supported by MongoDB 3.2 and later versions. You can use joins by using an aggregate query.
You can do it using the below example :
db.user.aggregate([
// Join with square table
{
$lookup:{
from: "square", // other table name
localField: "_id", // name of user table field
foreignField: "userId", // name of square table field
as: "square" // alias for userinfo table
}
},
{ $unwind:"$user_info" }, // $unwind used for getting data in object or for one record only
// define some conditions here
{
$match:{
$and:[{"userName" : "XYZ User"}]
}
},
// define which fields are you want to fetch
{
$project:{
_id: 1,
userId: "$square.userId",
userName: 1,
side: "$square.side",
area: "$square.area"
}
}
]);
The Result will be
{
_id: "ef6f6ac2-a08a-4f68-a63c-0b4a70285427",
userId: "d04d53dc-fb88-433e-a1c5-dd41a68d7655",
userName: "XYZ User",
side: 4,
area: 16
}
Cheers
I am new to MongoDB. I need to convert this SQL code to MongoDB
select TOP 5 r.regionName, COUNT(c.RegionID)
from region as r,
company as c
where c.RegionID = r._id
group by r.regionName
order by COUNT(c.RegionID) DESC;
Option 1. You can use the aggregation framework with $lookup, $group, $project , $sort and $limit stages, but this seems like a wrong approach since the true power to change relation database with mongoDB is the denormalization and avoidance of join ($lookup) like queries.
Option 2. You convert your multi-table relational database schema to document model and proceed with simple $group, $project, $sort and $limit stage aggregation query for the above task.
Since you have not provided any mongodb document examples it is hard to provide how your queries will look like ...
Despite of my comment I try to give a translation (not tested):
db.region.aggregate([
{
$lookup: // left outer join collections
{
from: "company",
localField: "_id",
foreignField: "RegionID",
as: "c"
}
},
{ $match: { c: { $ne: [] } } }, // remove non-matching documents (i.e. INNER JOIN)
{ $group: { _id: "$regionName", regions: { $addToSet: { "$c.RegionID" } } } }, // group and get distinct regions
{ $project: { regionName: "$_id", count: { $size: "$regions" } , _id: 0} } // some cosmetic and count
{ $sort: { regionName: 1 } }, // order result
{ $limit: 5 } // limit number or returned documents
])
I have two documents on mongodb, these are percentages and items. I'm good at SQL, I can write PLSql query as follows but i can not convert to mongodb query. Because my mongodb level of knowledge is at the beginning. Actually I know I have to use $gt for the and condition. But I don't know how I can say not exists or union keyword for mongodb. How can I write mongodb query? which keywords should i search for?
select p.*, "to_top" as list
from percentages p
where p.percentage > 5
and p.updatetime > sysdate - 1/24
and not exists (select 1
from items i
where i.id = p.p_id
and i.seller = p.seller)
order by p.percentage desc
union
select p2.*, "to_bottom" as list
from percentages p2
where p2.percentage > 5
and p2.updatetime > sysdate - 1/24
and exists (select 1
from items i2
where i2.id = p2.p_id
and i2.seller = p2.seller)
order by p2.percentage desc
There is no UNION for MongoDB. Luckely, each query is performed on the same collection and have very close condition, so we can implement "Mongo way" query.
Explanation
Normally, alsmost all complex SQL queries are done with the MongoDB aggregation framework.
We filter document by percentage / updatetime. Explanation why we need to use $expr
SQL JOIN / Subquery is done with the $lookup operator.
SQL SYSDATE in MongoDB way can be NOW or CLUSTER_TIME variable.
db.percentages.aggregate([
{
$match: {
percentage: { $gt: 5 },
$expr: {
$gt: [
"$updatetime",
{
$subtract: [
ISODate("2020-06-14T13:00:00Z"), //Change to $$NOW or $$CLUSTER_TIME
3600000
]
}
]
}
}
},
{
$lookup: {
from: "items",
let: {
p_id: "$p_id",
seller: "$seller"
},
pipeline: [
{
$match: {
$expr: {
$and: [
{
$eq: [ "$$p_id", "$id"]
},
{
$eq: [ "$$seller", "$seller"]
}
]
}
}
},
{
$limit: 1
}
],
as: "items"
}
},
{
$addFields: {
list: {
$cond: [
{
$eq: [{$size: "$items"}, 0]
},
"$to_top",
"$to_bottom"
]
},
items: "$$REMOVE"
}
},
{
$sort: { percentage: -1 }
}
])
MongoPlayground
Note: The MongoDB aggregation has the $facet operator that allows to perform different queries on the same collection.
SCHEMA:
db.percentages.aggregate([
{$facet:{
q1:[...],
q2:[...],
}},
//We apply "UNION" the result documents for each pipeline into single array
{$project:{
data:{$concatArrays:["$q1","$q2"]}
}},
//Flatten array into single object
{$unwind:"$data"}
//Replace top-level document
{$replaceWith:"$data"}
])
MongoPlayground
why you don't import your mangoDB data into oracle and use sql(that is more easy and powerful than mango.)
I'm struggling in how to select from two tables using the Sequelize.
Actually I'm trying to do it:
SELECT * FROM users, clients WHERE user.id = clients.user_id
I have no idea how to user two tables as I described, the only thing I did that got some results were:
const clients = await Client.findAll({
attributes: ["user_id"],
});
const users = [];
for (const client of clients) {
let user = await User.findAll({
where: {
id: {
[Op.eq]: client.user_id
}
}
});
users.push(user);
}
Which return me something:
[
[
{
"id": 1,
"first_name": "Velda",
"middle_name": "Zboncak",
"last_name": "Kris",
"email": "vkris10#hotmail.com",
"created_at": "2020-02-07T20:09:29.484Z",
"updated_at": "2020-02-07T20:09:29.484Z"
}
]
];
Model and Assossiation
First of all, you need to create the correct associations in the model of your table. In this case for the User and the Client, it's supposed to be an Client.belongsTo(...)
Take a look at User model:
const { Model, DataTypes } = require("sequelize");
class User extends Model {
static init(sequelize) {
super.init({
first_name: DataTypes.STRING,
middle_name: DataTypes.STRING,
last_name: DataTypes.STRING,
email: DataTypes.STRING
}, { sequelize });
}
}
module.exports = User;
Take a look at Client model:
const { Model, DataTypes } = require("sequelize");
class Client extends Model {
static init(sequelize) {
super.init({
user_id: DataTypes.INTEGER // The foreign key
}, { sequelize });
}
static associate(models) {
Client.belongsTo(models.User, {
foreignKey: "id", // Column name of associated table
as: "user" // Alias for the table
});
}
}
module.exports = Client;
When associating tables you need to have in mind those values inside the associate method, being the foreignKey: "id" the column name inside the models.ModelName, which will be used to make the joins, and the as: "user" which are used as an alias for the table like SELECT t.column1 FROM table AS t;
Controller
Okay, now you have set your models, you need to set your controller, where the magic happens. As you said you want to fetch results using:
SELECT * FROM users, clients WHERE user.id = clients.user_id
But to achieve the same result you can follow the sql join method to fetch the results from db, so it will be something like this:
SELECT
"user"."first_name", "user"."middle_name", "user"."last_name", "user"."email"
FROM "clients" AS "client"
LEFT JOIN "users" AS "user"
ON "client"."id" = "user"."id";
Knowing that we can talk about including tables in sequelize, which is the same as associations
const Client = require("./path/to/models/Client");
module.exports = {
async fetchAll(req, res) {
const results = await Client.findAll({
limit: 25,
include: [
{
association: "user",
attributes: ["first_name", "middle_name", "last_name", "email"]
}
]
});
return res.json(results);
},
};
Now lets talk about what is going on in the code:
The Model.findAll({}) will fetch all the result inside the specified table, in this case clients table.
The limit: 25 will limit your results in only 25 rows, you are free to remove or edit as you need.
The include: [], it will do the joins through the tables you specify, as you need only the users table, we are going to use only one object, so the assossiation: "user" will make this connection between tables, you must use the same alias you set inside the model. And at least the attributes: ["columns"] is where you set all the fields you want to fetch.
And that's it, you make you request, and the result of this will be exactly the same join as I mentioned. And the results will be:
[
{
"id": 1,
"user_id": 1,
"user": {
"first_name": "John",
"middle_name": "Ironsight",
"last_name": "Doe",
"email": "johndoe#example.com"
}
}, {...}
]
Can use where in include. Find the document at here
let user_id = client.user_id;
users = await User.findAll({
include: [
{
model: Client,
as: 'client',
where: {
user_id: user_id
}
}
]
});