Trying to understand mikro-orm Identity Map

[alexojegu]

In the README.md I can read:

Thanks to Identity Map, you will always have only one instance of given entity in one context. This allows for some optimizations (skipping loading of already loaded entities), as well as comparison by identity (ent1 === ent2).

But when I do:

public async findId(id: string | number): Promise<Book | null> {
    await this.repository.findOne({ id: { $eq: id as number } });
    return this.repository.findOne({ id: { $eq: id as number } });
}

In the debug I can see:

[query] select `e0`.* from `book` as `e0` where `e0`.`id` = '7' limit 1 [took 3 ms]
[query] select `e0`.* from `book` as `e0` where `e0`.`id` = '7' limit 1 [took 2 ms]

So, how or where happens the "skipping loading of already loaded entities"?

[B4nan]

You need to load by PK, just by the value. There is no query processing, so anything more complex will always do the query.

public async findId(id: string | number): Promise<Book | null> {
    await this.repository.findOne(id as number);
    return this.repository.findOne(id as number);
}

[alexojegu]

And from what I see, it don't work for a array of PKs (just by the values):

public async findId(id: string | number): Promise<Book | null> {
    await this.repository.find([1, 2, 3]);
    await this.repository.find([1, 2, 3]);

    await this.repository.findOne(id as number);
    return this.repository.findOne(id as number);
}

Result in:

[query] select `e0`.* from `book` as `e0` where `e0`.`id` in (1, 2, 3) [took 2 ms]
[query] select `e0`.* from `book` as `e0` where `e0`.`id` in (1, 2, 3) [took 2 ms]
[query] select `e0`.* from `book` as `e0` where `e0`.`id` = '7' limit 1 [took 1 ms]