# MongoDB

## Prerequisite
- NoSQL => not only SQL
- Non-Relational => You still can store relational data (just in different way)
- Document DataBase
- Stored in `bson (binary json)` but represented as `json`
- [Some Concepts about MongoDB](https://dotblogs.com.tw/sungnoone/2012/01/11/65274)
- [Cheat Sheet](https://gist.github.com/codeSTACKr/53fd03c7f75d40d07797b8e4e47d78ec)

## Mongoose

> The **`.save()`** is an instance method of the model, while the **`.create()`** is called directly from the Model as a method call, being static in nature, and takes the object as a first parameter.

const mongoose = require('mongoose');

const notificationSchema = mongoose.Schema({
    "datetime" : {
        type: Date,
        default: Date.now
    },
    "ownerId":{
        type:String
    },
    "customerId" : {
        type:String
    },
    "title" : {
        type:String
    },
    "message" : {
        type:String
    }
});

const Notification = mongoose.model('Notification', notificationsSchema);


function saveNotification1(data) {
    var notification = new Notification(data);
    notification.save(function (err) {
        if (err) return handleError(err);
        // saved!
    })
}

function saveNotification2(data) {
    Notification.create(data, function (err, small) {
    if (err) return handleError(err);
    // saved!
    })
}

- Example2

router.post("/", (req, res) => {
    const user = new User({
        isAdmin: req.body.isAdmin,
        chineseName: req.body.chineseName,
    });

    user.save()
        .then((data) => res.json(data))
        .catch((err) => {
            res.json(`err: ${err}`);
        });
});

- Example2 (try, catch)

    try {
        const data = await user.save();
        res.json(data);
    } catch (err) {
        res.json(`err: ${err}`);
    }

- `$set`
- https://docs.mongodb.com/manual/reference/operator/update/set/

## Mongoose (CRUD)

- `model.create(docs)`
- returns all the data

model.create({
    name: "Allen",
    age: 15,
    school: "NTHU",
    FAQ: [{"Ques":"Q1", "Ans":"A1"}],
    
})

<hr />

- `model.find(filter, projection, { limit, skip, sort}).select()`
- limit: `Number`
- skip: `Number`
- sort: `object`
- Ex: { name: 1 }
- `.select()`: 配合 `projection`
- returns all the data

model.find(
    { name: "Allen" },
    'name age',
    {
        limit,
        skip,
        sort
    }
).select().lean();

<hr />

- `model.findOne(filter)`
- returns all the data

model.findOne(
    { _id: "621bb7afa12afd5f13d00e4d" }
).lean();

<hr />

- `model.updateOne(filter, update)`
- filter: `object`
- update: `object`
- MongoDB will update only the first document that matches filter regardless of the value of the multi option.
- return `object` containing:
- `matchedCount`
- number of documents matched
- `modifiedCount`
- number of documents modified
- `acknowledge`
- boolean indicating everything went smoothly
- `upsertId`
- null or an id containing a document that had to be upserted
- `upsertedCount`
- number indicating how many documents had to be upserted. Will either be 0 or 1

model.updateOne(
    { _id: "621bb7afa12afd5f13d00e4d" },
    { name: "modified", age: 45 }
).lean();

<hr />

- `model.deleteOne(filter)`
- filter: `object`
- returns an `object` containing:
- `deletedCount`: 1

model.deleteOne(
    { _id: "621bb7afa12afd5f13d00e4d" }
)

<hr />

- `model.deleteMany()`
- filter: `object`
- returns an `object` containing:
- `deletedCount: x`: where x is the number of documents deleted.

model.deleteMany(
    { name: "Allen", age: { $gte: 18 } }
)

<hr />

- [operators](https://docs.mongodb.com/manual/reference/operator/query/)
- `$or` `$gte`

## Cooperate with Docker

- https://hub.docker.com/_/mongo