Let's say I have a collection with this single document:
{
"_id" : ObjectId("…"),
"cartId" : "61",
"items" : [
{
"prodType" : "hardware",
"prod" : "screwdriver",
"checked": false
},
{
"prodType" : "hardware",
"prod" : "hammer",
"checked": false
},
{
"prodType" : "decor",
"prod" : "vase",
"checked": false
}
]
}
And I want to do findAndModify to find any hardware product and modify its checked field. Then it will look like this:
db.col.findAndModify({
query: {
items: {
$elemMatch: {
prodType: "hardware"
}
}
},
update: {
$set: {
"items.$.checked": true
}
}
})
Okay, but this isn't the whole story. findAndModify will return the whole matched document, and I want to project specifically the array item that was matched (and also modified), so I'll add a fields section to my query:
db.col.findAndModify({
query: {
items: {
$elemMatch: {
prodType: "hardware"
}
}
},
update: {
$set: {
"items.$.checked": true
}
},
fields: {
items: {
$elemMatch: {
prodType: "hardware"
}
}
}
})
And now to the question: does MongoDB guarantee that the returned array item from my query is the exact same one that was matched (and modified) in the update section even though we have two items matching the criteria?
YES. It will return only the first sub-document that matched your criteria and was modified in the update section as shown here
According to the official docs, then yes - the projected array element is the exact one that was modified using the same one modified by the positional operator.
$ (update) states:
the positional $ operator acts as a placeholder for the first element that matches the query document
and $elemMatch (projection) states:
The $elemMatch operator limits the contents of an <array> field from the query results to contain only the first element matching the $elemMatch condition
They both apply to the first array element so it directly implies that the modified array element is the one that is projected
Related
I would like to update each elements (object) in an array of a company.
Here my actual data :
{
_id: ObjectId("60d31024860ce0400b586111")
contracts:
[
{
name: 1.pdf
url: "https://someurl"
createdAt: 2021-06-23T10:42:44.594+00:00
}
{
name: 2.pdf
url: "https://someurl"
}
{
name: 3.pdf
url: "https://someurl"
}
]
}
I would like to add a defined date on each object (in contracts) that has no "updatedAt" key.
Here what I tried :
db.companies.update({ _id: ObjectId("60d31024860ce0400b586111"),"contracts.createdAt": { $exists: false } },{ $set: { "contracts.$.createdAt": "test" } })
but I got this error :
"The positional operator did not find the match needed from the query."
I have also tried this and it works, but I don't wanna query by file name. I just wanna add "createdAt" on each elements found that has no "createdAt"
db.companies.update({ "contracts.name": "2.pdf" },{ $set: { "contracts.$.createdAt": "atest" } })
I think you need to use the filtered position operator:
$ - updates the first matched array element
$[] - updates all the matched elements with a specific condition
The specific condition is mentioned in the arrayFilters key.
db.students.update(
{ },
{ $set: { "contracts.$[element].createdAt" : "atest"} },
{ multi: true,
arrayFilters: [ { "element.createdAt": { $exists: false } } ]
}
)
multi - true is to apply the operation on all the matching documents.
Also notice, how the first query parameter is empty, which means the query runs for all documents. I used it based on the second query you wrote but you can also add in an ObjectID query there.
In the mongo documentation unsetting a field is done with $unset. I can't quite grasp how it works, but it seems like it should be simple.
The following operation uses the $unset operator to remove the tags field:
db.books.update( { _id: 1 }, { $unset: { tags: 1 } } )
My confusion arises when setting what to unset. What is the value 1 for in the $unset clause?
As per the $unset documentation :-
The $unset operator deletes a particular field.
Syntax : { $unset: { <field1>: "", ... } }
The specified value in the $unset expression (i.e. "") does not impact the operation.
If the field does not exist, then $unset does nothing (i.e. no operation).
So you can use
db.books.update( { _id: 1 }, { $unset: { tags: 1 } } )
OR
db.books.update( { _id: 1 }, { $unset: { tags: 0 } } )
OR
db.books.update( { _id: 1 }, { $unset: { tags: "" } } )
All the above queries will delete tags field.
Hope your doubt is clear now.
{$unset : { tags : 1 } } will clear the field tags from the document. The value 1 is just to tell that, clear this field tags from the document.
If you want to clear multiple fields, you need to write {$unset : { tags : 1, randomField : 1} } and like that.
You can refer official documentation of $unset for further info.
According to the documentation:
The $unset operator deletes a particular field. Consider the
following syntax:
{ $unset: { field1: "", ... } }
The specified value in the $unset
expression (i.e. "") does not impact the operation.
If the field does not exist, then $unset does nothing (i.e. no
operation).
When used with $ to match an array element, $unset replaces the
matching element with null rather than removing the matching element
from the array.
The $unset operator deletes a particular field.
I would like to add a document if it does not exist and else add an element to one of it's sub-documents.
db.test.update(
{
name : 'Peter'
},
$setOnInsert : {
name : 'Peter',
visits: { 'en' : ['today'], 'us' : [] }
},
$push : {
visits.en : 'today'
},
{ upsert : true }
)
If Peter exists, add an element to its visists.en or visists.us arrays. Else, create a document for Peter. This document should have the format for visits which should contain the current element ('today').
My issue is that I have "have conflicting mods in update".
I.e. (afaik), I cannot write to two things in one query. Yet how can I solve this dilemma?
You could implement it without $setOnInsert operator.
db.test.update(
{
name : 'Peter'
},
{
$push : {
"visits.en" : 'today'
}
},
{ upsert : true }
)
If Peter exists, element 'today' will be added to its visits.en array. Else, will be created a document for Peter, with visits object, that will be contain array en with 'today' element.
And I think, that error occured because of you using same property (visits) in two operations ($setOnInsert and $push).
You can still use $setOnInsert but when $setOnInsert and $push doesn't updates in the same fields as mentioned before.
N.b: We use $addToSet if you don't want a duplicated values in your Array
db.test.update(
{
name : 'Peter'
},
{
$setOnInsert: {name : 'Peter'},
$addToSet: {"visits.en": 'today'} // or $push
},
{upsert: true})
I have a document with an array field, similar to this:
{
"_id" : "....",
"Statuses" : [
{ "Type" : 1, "Timestamp" : ISODate(...) },
{ "Type" : 2, "Timestamp" : ISODate(...) },
//Etc. etc.
]
}
How can I update a specific Status item's Timestamp, by specifying its Type value?
From mongodb shell you can do this by
db.your_collection.update(
{ _id: ObjectId("your_objectid"), "Statuses.Type": 1 },
{ $set: { "Statuses.$.Timestamp": "new timestamp" } }
)
so the c# equivalent
var query = Query.And(
Query.EQ("_id", "your_doc_id"),
Query.EQ("Statuses.Type", 1)
);
var result = your_collection.Update(
query,
Update.Set("Statuses.$.Timestamp", "new timestamp", UpdateFlags.Multi,SafeMode.True)
);
This will update the specific document, you can remove _id filter if you wanted to update the whole collection
Starting with MongoDB 3.6, the $[<identifier>] positional operator may be used. Unlike the $ positional operator — which updates at most one array element per document — the $[<identifier>] operator will update every matching array element. This is useful for scenarios where a given document may have multiple matching array elements that need to be updated.
db.yourCollection.update(
{ _id: "...." },
{ $set: {"Statuses.$[element].Timestamp": ISODate("2021-06-23T03:47:18.548Z")} },
{ arrayFilters: [{"element.Type": 1}] }
);
The arrayFilters option matches the array elements to update, and the $[element] is used within the $set update operator to indicate that only array elements that matched the arrayFilter should be updated.
I have a document with an array field, similar to this:
{
"_id" : "....",
"Statuses" : [
{ "Type" : 1, "Timestamp" : ISODate(...) },
{ "Type" : 2, "Timestamp" : ISODate(...) },
//Etc. etc.
]
}
How can I update a specific Status item's Timestamp, by specifying its Type value?
From mongodb shell you can do this by
db.your_collection.update(
{ _id: ObjectId("your_objectid"), "Statuses.Type": 1 },
{ $set: { "Statuses.$.Timestamp": "new timestamp" } }
)
so the c# equivalent
var query = Query.And(
Query.EQ("_id", "your_doc_id"),
Query.EQ("Statuses.Type", 1)
);
var result = your_collection.Update(
query,
Update.Set("Statuses.$.Timestamp", "new timestamp", UpdateFlags.Multi,SafeMode.True)
);
This will update the specific document, you can remove _id filter if you wanted to update the whole collection
Starting with MongoDB 3.6, the $[<identifier>] positional operator may be used. Unlike the $ positional operator — which updates at most one array element per document — the $[<identifier>] operator will update every matching array element. This is useful for scenarios where a given document may have multiple matching array elements that need to be updated.
db.yourCollection.update(
{ _id: "...." },
{ $set: {"Statuses.$[element].Timestamp": ISODate("2021-06-23T03:47:18.548Z")} },
{ arrayFilters: [{"element.Type": 1}] }
);
The arrayFilters option matches the array elements to update, and the $[element] is used within the $set update operator to indicate that only array elements that matched the arrayFilter should be updated.