I want to sort my aggregation only if a condition is met.
This is what I have so far:
{
$cond: {
if: { $gte: [sort, "like"] },
then: { $divide: { $sort : { total_likes : -1 } } },
else: { $divide: '' }
}
}
sort is a variable that comes from a query parameter.
I want to sort by total_likes, only if sort is "likes". If it's not, I want to leave it alone.
First of all, #schoenbl, if you want to match some condition in mongo aggregation, you should use $match aggregation. It will send the documents which fulfill the given condition.
if: { $gte: [sort, "like"] }
In MongoDB, you are not allowed to compare string using "gte" operator. For string comparison in MongoDB, you get two operators:
for case sensitive $cmp.
for case insensitive $strcasecmp.
then: { $divide: { $sort : { total_likes : -1 } } },
Next, you were using divide operator don't know what is your need but syntax is improper,
refer $divide, for better knowledge.
Also, you are doing sorting in $cond, which means you want to sort each element, and that is not possible because you can't sort without having a comparison as you are inside $cond operator and it is performing manipulation on a single document.
Now, according to your need, I have prepared the next stages which will give sorted document which contains "sort" equals to "like".
{$match:{"sort":"like"}},{$sort:{"total_likes":-1}}
Output:
{ "_id" : ObjectId("5d50569fbe39828b4a22fba2"), "name" : "kyle", "sort" : "like", "total_likes" : 5 }
{ "_id" : ObjectId("5d5056a6be39828b4a22fba3"), "name" : "jack", "sort" : "like", "total_likes" : 2 }
{ "_id" : ObjectId("5d5056abbe39828b4a22fba4"), "name" : "john", "sort" : "like", "total_likes" : 1 }
Related
Is there a possibility to calculate mathematical operation on already aggregated computed fields?
I have something like this:
([
{
"$unwind" : {
"path" : "$users"
}
},
{
"$match" : {
"users.r" : {
"$exists" : true
}
}
},
{
"$group" : {
"_id" : "$users.r",
"count" : {
"$sum" : 1
}
}
},
])
Which gives an output as:
{ "_id" : "A", "count" : 7 }
{ "_id" : "B", "count" : 49 }
Now I want to divide 7 by 49 or vice versa.
Is there a possibility to do that? I tried $project and $divide but had no luck.
Any help would be really appreciated.
Thank you,
From your question, it looks like you are assuming result count to be 2 only. In that case I can assume users.r can have only 2 values(apart from null).
The simplest thing I suggest is to do this arithmetic via javascript(if you're using it in mongo console) or in case of using it in progam, use the language you're using to access mongo) e.g.
var results = db.collection.aggregate([theAggregatePipelineQuery]).toArray();
print(results[0].count/results[1].count);
EDIT: I am sharing an alternative to above approach because OP commented about the constraint of not using javascript code and the need to be done only via query. Here it is
([
{ /**your existing aggregation stages that results in two rows as described in the question with a count field **/ },
{ $group: {"_id": 1, firstCount: {$first: "$count"}, lastCount: {$last: "$count"}
},
{ $project: { finalResult: { $divide: ['$firstCount','$lastCount']} } }
])
//The returned document has your answer under `finalResult` field
note: I'm using Mongodb 4 and I must use aggregation, because this is a step of a bigger aggregation
Problem
How to find in a collection documents that contains fields that starts with value from another field in same document ?
Let's start with this collection:
db.regextest.insert([
{"first":"Pizza", "second" : "Pizza"},
{"first":"Pizza", "second" : "not pizza"},
{"first":"Pizza", "second" : "not pizza"}
])
and an example query for exact match:
db.regextest.aggregate([
{
$match : { $expr: { $eq: [ "$first" ,"$second" ] } } }
])
I will get a single document
{
"_id" : ObjectId("5c49d44329ea754dc48b5ace"),
"first" : "Pizza", "second" : "Pizza"
}
And this is good.
But how to do the same, but with startsWith ? My plan was to use regex but seems that is not supported in aggregation so far.
With a find and a custom javascript function works fine:
db.regextest.find().forEach(
function(obj){
if (obj.first.startsWith(obj.second)){
print(obj);
}
}
)
And returns correctly:
{
"_id" : ObjectId("5c49d44329ea754dc48b5ace"),
"first" : "Pizza",
"second" : "Pizza"
}
How it's possible to get same result with aggregation framework ?
One idea is to use existing aggregation framework pipeline, out to a temp colletion and then run the find above, to get match I'm looking for. This seems to be a workaround, I hope someone have a better idea.
Edit: here the solution
db.regextest.aggregate([{
$project : {
"first" : 1,
"second" : 1,
fieldExists : {
$indexOfBytes : ['$first', '$second' , 0]
}
}
}, {
$match : {
fieldExists : {
$gt : -1
}
}
}
]);
The simplest way is to use $expr, first available in 3.6 like this:
{
$match: {
$expr: {
$eq: [
'$second',
{
$substr: ['$first', 0, { $strLenCP: '$second' }]
}
]
}
}
}
This compares the string in field second with the first N characters of first where N is the length of second string. If they are equal, then first starts with second.
4.2 adds support for $regex in aggregation expressions, but starts with is much simpler and doesn't need regular expressions.
I have a collection of persons whose schema looks like the collection of following documents.
Document: {
name:
age:
educations:[{
title:xyz,
passed_year:2005,
univercity:abc},
{
title:asd
passed_year:2007,
univercity:mno
}],
current_city:ghi
}
Now I wanna show all the persons who has not done xyz education from abc university in year 2005.
I think two possible queries for this need but not sure which one to use as both of them are giving me the output
Query 1:
db.persons.find({"education":{$ne:{$elemMatch:{"title":"xyz","passed_year":2005,"univercity":"abc"}}}})
Query 2:
db.persons.find({"education":{$not:{$elemMatch:{"title":"xyz","passed_year":2005,"univercity":"abc"}}}})
I'm quite confused about operator $ne and $not, which one should I use with $elemMatch as both of them are giving me the output.
Given this $elemMatch: {"title":"xyz","passed_year":2005,"univercity":"abc"} I think you want to exclude any documents which contain an sub document in the educations array which contains all of these pairs:
"title" : "xyz"
"passed_year" : 2005
"univercity" : "abc"
This query will achieve that:
db.persons.find({
"educations": {
$not: {
$elemMatch:{"title": "xyz", "passed_year": 2005, "univercity": "abc"}
}
}
})
In your question you wrote:
both of them are giving me the output
I suspect this is because your query is specifying education whereas the correct attribute name is educations. By specifying education you are adding a predicate which cannot be evaluated since it references a non existent document attribute so regardless of whether that predicate uses $ne or $not it will simply not be applied.
In answer to the question of which operator to use: $not or $ne: if you run the above query with .explain(true) you'll notice that the parsed query produced by Mongo is very different for each of these operators.
Using $ne
"parsedQuery" : {
"$not" : {
"educations" : {
"$eq" : {
"$elemMatch" : {
"title" : "xyz",
"passed_year" : 2005,
"univercity" : "abc"
}
}
}
}
}
Using $not:
"parsedQuery" : {
"$not" : {
"educations" : {
"$elemMatch" : {
"$and" : [
{
"passed_year" : {
"$eq" : 2005
}
},
{
"title" : {
"$eq" : "xyz"
}
},
{
"univercity" : {
"$eq" : "abc"
}
}
]
}
}
}
}
So, it looks like use of $ne causes Mongo to do something like this psuedo code ...
not educations equalTo "$elemMatch" : {"title" : "xyz", "passed_year" : 2005, "univercity" : "abc"}
... i.e. it treats the elemMatch clause as if it is the RHS of an equality operation whereas use of $not causes Mongo to actually evaluate the elemMatch clause.
For a given record id, how do I get the average of a sub document field if I have the following in MongoDB:
/* 0 */
{
"item" : "1",
"samples" : [
{
"key" : "test-key",
"value" : "1"
},
{
"key" : "test-key2",
"value" : "2"
}
]
}
/* 1 */
{
"item" : "1",
"samples" : [
{
"key" : "test-key",
"value" : "3"
},
{
"key" : "test-key2",
"value" : "4"
}
]
}
I want to get the average of the values where key = "test-key" for a given item id (in this case 1). So the average should be $avg (1 + 3) = 2
Thanks
You'll need to use the aggregation framework. The aggregation will end up looking something like this:
db.stack.aggregate([
{ $match: { "samples.key" : "test-key" } },
{ $unwind : "$samples" },
{ $match : { "samples.key" : "test-key" } },
{ $project : { "new_key" : "$samples.key", "new_value" : "$samples.value" } },
{ $group : { `_id` : "$new_key", answer : { $avg : "$new_value" } } }
])
The best way to think of the aggregation framework is like an assembly line. The query itself is an array of JSON documents, where each sub-document represents a different step in the assembly.
Step 1: $match
The first step is a basic filter, like a WHERE clause in SQL. We place this step first to filter out all documents that do not contain an array element containing test-key. Placing this at the beginning of the pipeline allows the aggregation to use indexes.
Step 2: $unwind
The second step, $unwind, is used for separating each of the elements in the "samples" array so we can perform operations across all of them. If you run the query with just that step, you'll see what I mean.
Long story short :
{ name : "bob",
children : [ {"name" : mary}, { "name" : "sue" } ]
}
becomes two documents :
{ name : "bob", children : [ { "name" : mary } ] }
{ name : "bob", children : [ { "name" : sue } ] }
Step 3: $match
The third step, $match, is an exact duplicate of the first $match stage, but has a different purpose. Since it follows $unwind, this stage filters out previous array elements, now documents, that don't match the filter criteria. In this case, we keep only documents where samples.key = "test-key"
Step 4: $project (Optional)
The fourth step, $project, restructures the document. In this case, I pulled the items out of the array so I could reference them directly. Using the example above..
{ name : "bob", children : [ { "name" : mary } ] }
becomes
{ new_name : "bob", new_child_name : mary }
Note that this step is entirely optional; later stages could be completed even without this $project after a few minor changes. In most cases $project is entirely cosmetic; aggregations have numerous optimizations under the hood such that manually including or excluding fields in a $project should not be necessary.
Step 5: $group
Finally, $group is where the magic happens. The _id value what you will be "grouping by" in the SQL world. The second field is saying to average over the value that I defined in the $project step. You can easily substitute $sum to perform a sum, but a count operation is typically done the following way: my_count : { $sum : 1 }.
The most important thing to note here is that the majority of the work being done is to format the data to a point where performing the operation is simple.
Final Note
Lastly, I wanted to note that this would not work on the example data provided since samples.value is defined as text, which can't be used in arithmetic operations. If you're interested, changing the type of a field is described here: MongoDB How to change the type of a field
I'm trying to retrieve an object from an array in mongodb. Below is my document:
{
"_id" : ObjectId("53e9b43968425b29ecc87ffd"),
"firstname" : "john",
"lastname" : "smith",
"trips" : [
{
"submitted" : 1407824585356,
"tripCategory" : "staff",
"tripID" : "1"
},
{
"tripID" : "2",
"tripCategory" : "volunteer"
},
{
"tripID" : "3",
"tripCategory" : "individual"
}
]
}
My ultimate goal is to update only when trips.submitted is absent so I thought I could query and determine what the mongo find behavior would look like
if I used the $and query operator. So I try this:
db.users.find({
$and: [
{ "trips.tripID": "1" },
{ "trips": { $elemMatch: { submitted: { $exists: true } } } }
]
},
{ "trips.$" : 1 } //projection limits to the FIRST matching element
)
and I get this back:
{
"_id" : ObjectId("53e9b43968425b29ecc87ffd"),
"trips" : [
{
"submitted" : 1407824585356,
"tripCategory" : "staff",
"tripID" : "1"
}
]
}
Great. This is what I want. However, when I run this query:
db.users.find({
$and: [
{ "trips.tripID": "2" },
{ "trips": { $elemMatch: { submitted: { $exists: true } } } }
]
},
{ "trips.$" : 1 } //projection limits to the FIRST matching element
)
I get the same result as the first! So I know there's something odd about my query that isn't correct. But I dont know what. The only thing I've changed between the queries is "trips.tripID" : "2", which in my head, should have prompted mongo to return no results. What is wrong with my query?
If you know the array is in a specific order you can refer to a specific index in the array like this:-
db.trips.find({"trips.0.submitted" : {$exists:true}})
Or you could simply element match on both values:
db.trips.find({"trips" : {$elemMatch : {"tripID" : "1",
"submitted" : {$exists:true}
}}})
Your query, by contrast, is looking for a document where both are true, not an element within the trips field that holds for both.
The output for your query is correct. Your query asks mongo to return a document which has the given tripId and the field submitted within its trips array. The document you have provided in your question satisfies both conditions for both tripIds. You are getting the first element in the array trips because of your projection.
I have assumed you will be filtering records by the person's name and then retrieving the elements inside trips based on the field-exists criteria. The output you are expecting can be obtained using the following:
db.users.aggregate(
[
{$match:
{
"firstname" : "john",
"lastname" : "smith"
}
},
{$unwind: "$trips"},
{$match:
{
"trips.tripID": "1" ,
"trips.submitted": { $exists: true }
}
}
]
)
The aggregation pipeline works as follows. The first $match operator filters one document (in this case the document for john smith) The $unwind operator in mongodb aggregation unwinds the specified array (trips in this case), in effect denormalizing the sub-records associated with the parent records. The second $match operator filters the denormalized/unwound documents further to obtain the one required as per your query.