MongoDB. Remove subelement from array using parent elements for query [duplicate] - mongodb

Question

This question already has answers here:
How to remove an element from a doubly-nested array in a MongoDB document
(5 answers)
Closed 2 years ago.
Here is my collection object structure
"_id" : ObjectId("xxxxxxxxxxxxxxxxxxxxxxxx"),
"Name": "HelloWorld",
"OtherFields": "OtherValues",
"Projects" : [
{
"Project" : {
"key" : 111
},
"Category" : [
{
"No" : "123"
},
{
"No" : "987"
}
]
},
{
"Project" : {
"key" : 222
},
"Category" : [
{
"No" : "123"
},
{
"No" : "987"
}
]
}
]
I want to delete element with Category No「123」in Project with key 111
Closer thing that I did was
db.project_collection.updateOne({"_id": ObjectId("collection_obj_id")}, {"$pull": { "Projects": {"$and": [ {"Project.key": 111)}, {"Category.No": "123"} ] }} })
Live example
This command removing whole element inside "Projects" object
But I want to delete only "Projects" → "Category" → [0] subelement which has No 123.
When I trying to change "Projects" after "$pull" to something like "Projects.Category" or "Projects.$.Category"
The positional operator did not find the match needed from the query
error occures.
So is it possible to delete subelement without creating own backend logic?

Answer 1

Knowing project.key and category.no you can use this query:
db.collection.update({
"_id": 1,
"Projects.Project.key": 111
},
{
"$pull": {
"Projects.$.Category": {
"No": "123"
}
}
})
With the updated JSON, the $ operator is nedded.
This query basically point to a document with key: 111 and _id: 1 and using $ do the pull.
So it will pull the 'pointed' element which match no: "123".
Updated example here
I have added another document with _id: 2 to check that only one with _id:1 is updated.