MongoDB retrieve only specific data - mongodb

I have collection like this:
{
"Post": {
"post": "These are following example of your station of the various stamps, and this can't be foured by the name.\n\n also I can't use this way to search string in the midle way, also what are you doing is the default factory",
"like": [
"rudi",
"tabootie",
"oknoorap",
"various",
"rusian_roulette"
],
"Comment": [
{
"comment_id": 1,
"name": "Anonymous",
"comment": "You are absolutely right dude, when you call me, you can host here",
"like": [
"rudi",
"stumble",
"upon",
"facebook"
]
"timestamp": {
"t": 9000,
"i": 1311245225
}
},
{
"comment_id": 2,
"name": "Anonymous",
"comment": "the guy is here",
"like": [
"rudi",
"stumble",
"upon",
"facebook"
]
"timestamp": {
"t": 10000,
"i": 1311245225
}
},
{
"comment_id": 2,
"name": "Oknoorap",
"comment": "the other guy is here",
"like": [
"rudi",
"stumble",
"upon",
"facebook"
]
"timestamp": {
"t": 11000,
"i": 1311245225
}
}
]
}
}
Could you help me? How to retrieve only for Post.Comment.comment_id = 2, negation for _id, post, etc

It is not possible to retrieve only parts of an array. You can limit fetching only per document/embedded document.

Java or Javascript?
Java:
//Query items
BasicDBObject query = new BasicDBObject();
query.put("Post.Comment.comment_id", 2);
//Show items
BasicDBObject showField = new BasicDBObject();
showField.put("post", 1); //show
showField.put("other", 0); //hide
//Execute query
DBCursor cursor = dbc.find(query, showField);
......

Although your document schema could probably be improved .. the Aggregation Framework in MongoDB 2.2 enables some new flexibility.
Note: you have two comments with comment_id = 2 in your example, which seems to be a mistake if you are using this to uniquely identify a comment. I'm assuming your third comment_id should actually be comment_id = 3.
Here is a commented example using aggregate():
db.posts.aggregate(
// Find specific post
{ $match : {
'_id' : 123,
}},
// Unwind the Post.Comment array into a stream of documents
{ $unwind : '$Post.Comment' },
// Match specific comment_id
{ $match : {
'Post.Comment.comment_id' : 2,
}},
// Limit results to the embedded Comment
{ $project: {
'Post.Comment': 1
}}
)
.. and the results:
{
"result" : [
{
"_id" : 123,
"Post" : {
"Comment" : {
"comment_id" : 2,
"name" : "Anonymous",
"comment" : "the guy is here",
"like" : [
"rudi",
"stumble",
"upon",
"facebook"
],
"timestamp" : {
"t" : 10000,
"i" : 1311245225
}
}
}
}
],
"ok" : 1
}

Related

How to save deletion in a deeply nested MongoDB document

I am new to MongoDB and I am using MongoDB shell to perform the operations.
I am working to remove the array named Process from all the Items, but it seems that I do not grasp the remove concept correctly.
The documents we use are deeply nested - we do not know how many items there are, or how deep the level of nesting.
What I tried so far is to use recursion to iterate through the items:
function removeAllProcessFields(docItems)
{
if(Array.isArray(docItems))
{
docItems.forEach(function(item)
{
print("idItem: "+item._id);
if(item.Process == null)
{
print("Process null");
}
else
{
$unset: { Process: ""}
}
removeAllProcessFields(item.Items);
})
}
}
var docs = db.getCollection('MyCollection').find({})
docs.forEach(function(doc)
{
print("idDoc: "+doc._id);
removeAllProcessFields(doc.Items);
})
But I have difficulties on using unset properly to save the operation.
An example document would be:
{
"_id": "622226d319517e83e8ed6151",
"Name": "test1",
"Description": "",
"Items": [{
"_id": "622226d319517e83e8ed614e",
"Name": "test-item",
"Description": "",
"Process": [{
"Name": "Step1"
}, {
"Name": "Step2"
}],
"Items": [{
"_id": "622226d319517e83e8ed614f",
"Name": "test-subItem1",
"Description": "",
"Process": [{
"Name": "StepSub1"
}, {
"Name": "StepSub2"
}, {
"Name": "StepSub3"
}],
"Items": []
},
{
"_id": "622226d319517e83e8ed6150",
"Name": "test-subItem2",
"Description": "",
"Process": [{
"Name": "StepSub4"
}, {
"Name": "StepSub5"
}, {
"Name": "StepSub6"
}],
"Items": []
}
]
}]
}
What I hope to achieve would be:
{
"_id": "622226d319517e83e8ed6151",
"Name": "test1",
"Description": "",
"Items": [{
"_id": "622226d319517e83e8ed614e",
"Name": "test-item",
"Description": "",
"Items": [{
"_id": "622226d319517e83e8ed614f",
"Name": "test-subItem1",
"Description": "",
"Items": []
},
{
"_id": "622226d319517e83e8ed6150",
"Name": "test-subItem2",
"Description": "",
"Items": []
}
]
}]
}
Something like this maybe using the $[] positional operator:
db.collection.update({},
{
$unset: {
"Items.$[].Items.$[].Process": 1,
"Items.$[].Process": 1
}
})
You just need to construct it in the recursion ...
playground
JavaScript recursive function example:
mongos> db.rec.find()
{ "_id" : ObjectId("622a6c46ae295edb276df8e2"), "Items" : [ { "a" : 1 }, { "Items" : [ { "Items" : [ { "Items" : [ ], "Process" : [ 1, 2, 3 ] } ], "Process" : [ 4, 5, 6 ] } ], "Process" : [ ] } ] }
mongos> db.rec.find().forEach(function(obj){ var id=obj._id,ar=[],z=""; function x(obj){ if(typeof obj.Items != "undefined" ){ obj.Items.forEach(function(k){ if( typeof k.Process !="undefined" ){ z=z+".Items.$[]";ar.push(z.substring(1)+".Process") }; if(typeof k.Items != "undefined"){x(k)}else{} }) }else{} };x(obj);ar.forEach(function(del){print( "db.collection.update({_id:ObjectId('"+id+"')},{$unset:{'"+del+"':1}})" );}) })
db.collection.update({_id:ObjectId('622a6c46ae295edb276df8e2')},{$unset:{'Items.$[].Process':1}})
db.collection.update({_id:ObjectId('622a6c46ae295edb276df8e2')},{$unset:{'Items.$[].Items.$[].Process':1}})
db.collection.update({_id:ObjectId('622a6c46ae295edb276df8e2')},{$unset:{'Items.$[].Items.$[].Items.$[].Process':1}})
mongos>
Explained:
Loop over all documents in collection with forEach
Define recursive function x that will loop over any number of nested Items and identify if there is Process field and push to array ar
Finally loop over array ar and construct the update $unset query , in the example only printed for safety , but you can improve generating single query per document and executing unset query ...
Assuming you are on v>=4.4 you can use the "merge onto self" feature of $merge plus defining a recursive function to sweep through the collection and surgically remove one or a list of fields at any level of the hierarchy. The same sort of needs arise when processing json-schema data which is also arbitrarily hierarchical.
The solution below has extra logic to "mark" documents that had any modifications so the others can be removed from the update set passed to $merge. It also can be further refined to reduce some variables; it was edited down from a more general solution that had to examine keys and values.
db.foo.aggregate([
{$replaceRoot: {newRoot: {$function: {
body: function(obj, target) {
var didSomething = false;
var process = function(holder, spot, value) {
// test FIRST since [] instanceof Object is true!
if(Array.isArray(value)) {
for(var jj = value.length - 1; jj >= 0; jj--) {
process(value, jj, value[jj]);
}
} else if(value instanceof Object) {
walkObj(value);
}
};
var walkObj = function(obj) {
Object.keys(obj).forEach(function(k) {
if(target.indexOf(k) > -1) {
delete obj[k];
didSomething = true;
} else {
process(obj, k, obj[k]);
}
});
}
// ENTRY POINT:
if(!Array.isArray(target)) {
target = [ target ]; // if not array, make it an array
}
walkObj(obj);
if(!didSomething) {
obj['__didNothing'] = true;
}
return obj;
},
// Invoke!
// You can delete multiple fields with an array, e.g.:
// ..., ['Process','Description']
args: [ "$$ROOT", 'Process' ],
lang: "js"
}}
}}
// Only let thru docs WITHOUT the marker:
,{$match: {'__didNothing':{$exists:false}} }
,{$merge: {
into: "foo",
on: [ "_id" ],
whenMatched: "merge",
whenNotMatched: "fail"
}}
]);

Filter sub-document array using substring as criteria

My collection:
{
title: 'Computers',
maincategories:[
{
title: 'Monitors',
subcategories:[
{
title: '24 inch',
code: 'AFG'
}
]
}
]
}
I want query the code. The code is just the first part so I want to have all subcategories that contains the given search. So AFG101 would return this subcategories.
My query:
module.exports = (req, res) => {
var q = {
'maincategories.subcategories': {
$elemMatch: {
code: 'AFG101'
}
}
};
var query = mongoose.model('TypeCategory').find(q, {'maincategories.$': 1, 'title': 1});
query.exec((err, docs) => {
res.status(200).send(docs);
});
};
My problem:
How do I search for a part of a string? AFG101 should return all subcategories with property code containing any part of the string. So in this case, AFG would be a hit. Same as in this sql question: MySQL: What is a reverse version of LIKE?
How do I project the subcategories. Current query returns all subcategories. I only want to returns those hitting.
The best way to do this is in MongoDB 3.4 using the $indexOfCP string aggregation operator.
let code = "afg101";
db.collection.aggregate([
{ "$project": {
"title": 1,
"maincategories": {
"$map": {
"input": "$maincategories",
"as": "mc",
"in": {
"$filter": {
"input": "$$mc.subcategories",
"as": "subcat",
"cond": {
"$gt": [
{
"$indexOfCP": [
code,
{ "$toLower": "$$subcat.code" }
]
},
-1
]
}
}
}
}
}
}}
])
which returns:
{
"_id" : ObjectId("582cba57e6f570d40d77b3a8"),
"title" : "Computers",
"maincategories" : [
[
{
"title" : "24 inch",
"code" : "AFG"
}
]
]
}
You can read my other answers to similar question 1, 2 and 3.
From 3.2 backward, the only way to do this is with mapReduce.
db.collection.mapReduce(
function() {
var code = 'AFG101';
var maincategories = this.maincategories.map(function(sdoc) {
return {
"title": sdoc.title,
"subcategories": sdoc.subcategories.filter(function(scat) {
return code.indexOf(scat.code) != -1;
}
)};
});
emit(this._id, maincategories);
},
function(key, value) {},
{ "out": { "inline": 1 }
})
which yields something like this:
{
"results" : [
{
"_id" : ObjectId("582c9a1aa358615b6352c45a"),
"value" : [
{
"title" : "Monitors",
"subcategories" : [
{
"title" : "24 inch",
"code" : "AFG"
}
]
}
]
}
],
"timeMillis" : 15,
"counts" : {
"input" : 1,
"emit" : 1,
"reduce" : 0,
"output" : 1
},
"ok" : 1
}
Well, just like your question has two parts, I could think of two separate solution, however I don't see a way to join them together.
For first part $where can be used to do a reverse regex, but it's dirty, it's an overkill and it can't use any indexes, since $where runs on each documents.
db.TypeCategory.find({$where:function(){for(var i in this.maincategories)
{for(var j in this.maincategories[i].subcategories)
{if("AFG101".indexOf(this.maincategories[i].subcategories[j].code)>=0)
{return true}}}}},{"maincategories.subcategories.code":1})
Even if you use this option, it would need couple of boundary check and it cannot project two level of nested array. MongoDB doesn't support such projection (yet).
For that purpose we might go for aggregation
db.TypeCategory.aggregate([{$unwind:"$maincategories"},
{$unwind:"$maincategories.subcategories"},
{$match:{"maincategories.subcategories.code":"AFG"}},
{$group:{_id:"$_id","maincategories":{$push:"$maincategories"}}}
])
However I don't think there is a way to do reverse regex check in aggregation, but I might be wrong too. Also this aggregation is costly since there are two unwinds which can lead to overflow the memory limit for aggregation for a really large collection.
You can use $substr and do it
db.getCollection('cat').aggregate([
{"$unwind" : "$maincategories"},
{"$unwind" : "$maincategories.subcategories"},
{"$project" :
{"maincategories" : 1,
"title":1,"sub" : {"$substr" :["$maincategories.subcategories.code",0,3]}}},
{"$match" : {"sub" : "AFG"}},
{"$project" :
{"maincategories" : 1,
"title":1}
}
])

How to get back the new value after an update in a embedded array?

Given the following MongoDB collection:
{
"_id": ObjectId("56d6a7292c06e85687f44541"),
"name": "My ranking list",
"rankings": [
{
"_id": ObjectId("46d6a7292c06e85687f55542"),
"name": "Ranking 1",
"score": 1
},
{
"_id": ObjectId("46d6a7292c06e85687f55543"),
"name": "Ranking 2",
"score": 10
},
{
"_id": ObjectId("46d6a7292c06e85687f55544"),
"name": "Ranking 3",
"score": 15
},
]
}
Here is how I increase the score of a given ranking:
db.collection.update(
{ "_id": ObjectId("56d6a7292c06e85687f44541"), "rankings._id" : ObjectId("46d6a7292c06e85687f55543") },
{ $inc : { "rankings.$.score" : 1 } }
);
How do I get the new score value? In the previous query I increase the second ranking from 10 to 11... How do I get this new value back after the update?
If you are on MongoDB 3.0 or newer, you need to use the .findOneAndUpdate() and use projection option to specify the subset of fields to return. You also need to set returnNewDocument to true. Of course you need to use the $elemMatch projection operator here because you cannot use a positional projection and return the new document.
As someone pointed out:
You should be using .findOneAndUpdate() because .findAndModify() is highlighed as deprecated in every official language driver. The other thing is that the syntax and options are pretty consistent across drivers for .findOneAndUpdate(). With .findAndModify(), most drivers don't use the same single object with "query/update/fields" keys. So it's a bit less confusing when someone applies to another language to be consistent. Standardized API changes for .findOneAndUpdate() actually correspond to server release 3.x rather than 3.2.x. The full distinction being that the shell methods actually lagged behind the other drivers ( for once ! ) in implementing the method. So most drivers actually had a major release bump corresponding with the 3.x release with such changes.
db.collection.findOneAndUpdate(
{
"_id": ObjectId("56d6a7292c06e85687f44541"),
"rankings._id" : ObjectId("46d6a7292c06e85687f55543")
},
{ $inc : { "rankings.$.score" : 1 } },
{
"projection": {
"rankings": {
"$elemMatch": { "_id" : ObjectId("46d6a7292c06e85687f55543") }
}
},
"returnNewDocument": true
}
)
From MongoDB 3.0 onwards, you need to use findAndModify and the fields options also you need to set new to true in other to return the new value.
db.collection.findAndModify({
query: {
"_id": ObjectId("56d6a7292c06e85687f44541"),
"rankings._id" : ObjectId("46d6a7292c06e85687f55543")
},
update: { $inc : { "rankings.$.score" : 1 } },
new: true,
fields: {
"rankings": {
"$elemMatch": { "_id" : ObjectId("46d6a7292c06e85687f55543") }
}
}
})
Both queries yield:
{
"_id" : ObjectId("56d6a7292c06e85687f44541"),
"rankings" : [
{
"_id" : ObjectId("46d6a7292c06e85687f55543"),
"name" : "Ranking 2",
"score" : 11
}
]
}

How do I replace an entire array of subdocuments in MongoDB?

Here is an example document from my collection:
Books
[
id: 1,
links:
[
{text: "ABC", "url": "www.abc.com"},
{text: "XYZ", "url": "www.xyz.com"}
]
]
I want to replace the links array in one update operation. Here is an example of how the above document should be modified:
Books
[
id: 1,
links:
[
{text: "XYZ", "url": "www.xyz.com"},
{text: "efg", "url": "www.efg.com"}, <== NEW COPY OF THE ARRAY
{text: "ijk", "url": "www.ijk.com"}
]
]
As you can see, the links array has been replaced (old data removed, and new data added).
I am having very hard time with the Update.Set() because it says it MyLinks<> cannot be mapped to a BsonValue
I've tried many different ways of achieving this, and all of them fail, including .PushAllWrapped<WebLinkRoot>("links", myDoc.WebLinks).
Everything I've tried results in the new values being appended to the array, rather than the array being replaced.
As it seems MongoDB doesn't provide a simple method to replace an array of subdocument OR a method like .ClearArray(), what is the best way for me to ensure the array is cleared before adding new elements in a single query?
I am here because I saw 5k views on this post, I'm adding some stuff may be it help other who looking for answer of above
db.collectionName.insertOne({
'links': [
{
"text" : "XYZ",
"url" : "www.xyz.com"
}
]
});
now run this query which help to replace older data
db.collectionName.update(
{
_id: ObjectId("your object Id")
},
{
$set:
{
'links':[ {
"text" : "XYZ1",
"url" : "www.xyz.com1"
} ]
}
});
I think you have to do something like this:
var newArray = new BSONArray {
new BSONDocument { { "text", "XYZ" }, { "url", "www.xyz.com" } },
new BSONDocument { { "text", "efg" }, { "url", "www.efg.com" } },
new BSONDocument { { "text", "ijk" }, { "url", "www.ijk.com" } }
};
var update = Update.Set( "links", newArray );
collection.Update( query, update );
Or whatever method you can to cast as a valid BSONValue.
So equivalent to shell:
{ "links" : [ { "text" : "abc" } ] }
> db.collection.update(
{},
{ $set:
{ links: [
{ text: "xyz", url: "something" },
{ text: "zzz", url: "else" }
]}
})
>db.collection.find({},{ _id: 0, links:1 }).pretty()
{ "links" : [
{
"text" : "xyz",
"url" : "something"
},
{
"text" : "zzz",
"url" : "else"
}
]
}
So that works.
You clearly need something else other than embedded code. But hopefully that puts you on the right track.

select only subdocuments or arrays

{
"_id":{
"oid":"4f33bf69873dbc73a7d21dc3"
},
"country":"IND",
"states":[{
"name":"orissa",
"direction":"east",
"population":41947358,
"districts":[{
"name":"puri",
"headquarter":"puri",
"population":1498604
},
{
"name":"khordha",
"headquarter":"bhubaneswar",
"population":1874405
}
]
},
{
"name":"andhra pradesh",
"direction":"south",
"population":84665533,
"districts":[{
"name":"rangareddi",
"headquarter":"hyderabad",
"population":3506670
},
{
"name":"vishakhapatnam",
"headquarter":"vishakhapatnam",
"population":3789823
}
]
}
]
}
In above collection(i.e countries) i have only one document , and i want to fetch the details about a particular state (lets say "country.states.name" : "orissa" ) ,But i want my result as here under instead of entire document .Is there a way in Mogo...
{
"name": "orissa",
"direction": "east",
"population": 41947358,
"districts": [
{
"name": "puri",
"headquarter": "puri",
"population": 1498604
},
{
"name": "khordha",
"headquarter": "bhubaneswar",
"population": 1874405
}
]
}
Thanks
Tried this:
db.countries.aggregate(
{
"$project": {
"state": "$states",
"_id": 0
}
},
{
"$unwind": "$state"
},
{
"$group": {
"_id": "$state.name",
"state": {
"$first": "$state"
}
}
},
{
"$match": {
"_id": "orissa"
}
}
);
And got:
{
"result" : [
{
"_id" : "orissa",
"state" : {
"name" : "orissa",
"direction" : "east",
"population" : 41947358,
"districts" : [
{
"name" : "puri",
"headquarter" : "puri",
"population" : 1498604
},
{
"name" : "khordha",
"headquarter" : "bhubaneswar",
"population" : 1874405
}
]
}
}
],
"ok" : 1
You can't do it right now, but you will be able to with $unwind in the aggregation framework. You can try it now with the experimental 2.1 branch, the stable version will come out in 2.2, probably in a few months.
Any query in mongodb always return root document.
There is only one way for you to load one sub document with parent via $slice if you know ordinal number of state in nested array:
// skip ordinalNumberOfState -1, limit 1
db.countries.find({_id: 1}, {states:{$slice: [ordinalNumber -1 , 1]}})
$slice work in default order (as documents was inserted in nested array).
Also if you don't need fields from a country you can include only _id and states in result:
db.countries.find({_id: 1}, {states:{$slice: [ordinalNumber -1 , 1]}, _id: 1})
Then result document will looks like this one:
{
"_id":{
"oid":"4f33bf69873dbc73a7d21dc3"
},
"states":[{
"name":"orissa",
"direction":"east",
"population":41947358,
"districts":[{
"name":"puri",
"headquarter":"puri",
"population":1498604
},
{
"name":"khordha",
"headquarter":"bhubaneswar",
"population":1874405
}
]
}]
}
db.countries.find({ "states": { "$elemMatch": { "name": orissa }}},{"country" : 1, "states.$": 1 })
If you don't want to use aggregate, you can do it pretty easily at the application layer using underscore (included by default):
var country = Groops.findOne({"property":value);
var state _.where(country, {"state":statename});
This will give you the entire state record that matches statename. Very convenient.