performance request to a query with an include

[matthewchung74]

Is your feature request related to a problem? Please describe.

My feature is a performance request

I have a snippet of code

var query = new Parse.Query(MyComment);
query.include("user");
query.find().then(function (results) { ... }

where there are 2 MyComments each have a pointer to 2 a different user. I ran the request with my mongod set to verbose. In those logs, I see

2018-08-16T18:40:47.131-0700 I COMMAND  [conn1] command dev._User command: find { find: "_User", filter: { _id: "1BoyKbmjuW" }, ...
2018-08-16T18:40:47.156-0700 I COMMAND  [conn2] command dev._User command: find { find: "_User", filter: { _id: { $in: [ "1BoyKbmjuW", "KmbsMaQscS" ] }, ...

which makes me assume parse-server did an db query for the comments, then a separate query for each included user for a total of 3 queries. if i am understanding correctly, this would get pretty slow when there are 10+ comments.

Describe the solution you'd like

would it be possible to do all the includes in one request? i believe programmatically we could construct one query with a series of or statements such that request above would only result in 2 queries to the db.

Additional context
I am quite new to parse-server so please tell me if i am misunderstanding.

[doender]

@foobar8675 We are running into the same issue. Did you create a workaround yourself by splitting this into 2 queries?

[oallouch]

If you could try with 10 comments and 10 users, you would see better.

[matthewchung74]

I think I later watched the transaction via mongo logs and it did optimize the query correctly, meaning both users were part of the same in clause.

On Tue, Nov 13, 2018 at 2:40 AM Olivier Allouch ***@***.***> wrote: If you could try with 10 comments and 10 users, you would see better. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#4984 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABm4xCwSFGkskgU3mMEY0SqzqMOdRrQgks5uuqGLgaJpZM4WCBon> .

-- Regards, Matt