Hi guys I want to calulate sum of rate for comments group by type like this
Post::with(['comments' => function ($q) {
$q->selectRaw('type, SUM(rate) as total_rate')
->groupBy('type');
}])
I'm waiting for a result like this:
0 => array:4 [
"id" => 5
"start_date" => "2022-01-01"
"end_date" => "2022-01-31"
"comments" => array:2 [
0 => array:3 [
"type" => "personal"
"total_rate" => 44244.0
]
1 => array:3 [
"type" => "business"
"total_rate" => 22358.0
]
]
but the result is
0 => array:4 [
"id" => 5
"start_date" => "2022-01-01"
"end_date" => "2022-01-31"
"comments" => []
]
Related
I access a lookup field in $project using $unwind but this breaks the accessibility of the other nested fields from the main collection. Is there any way to access the fields from both collections in $project. I thought of merging the arrays but still not sure if it's the right approach.
Users collection
{
"_id" : ObjectId("5a54f739fe0a00373e7ef1e8"),
"team" : {
"name" : "test",
},
"updated_at" : ISODate("2018-05-22T04:28:00Z"),
"created_at" : ISODate("2018-01-09T17:09:13Z"),
"users" : [
{
"updated_at" : ISODate("2018-11-22T11:55:22Z"),
"created_at" : ISODate("2018-01-09T17:09:13Z"),
"_id" : ObjectId("5a54f739fe0a00373e7ef1e9"),
"name" : test,
"status" : "active",
"title" : "Engineer",
},
{
"updated_at" : ISODate("2018-11-22T11:55:22Z"),
"created_at" : ISODate("2018-01-09T17:09:13Z"),
"_id" : ObjectId("5a54f739fe0a00373e7ef1e9"),
"name" : test1,
"status" : "passive",
"title" : "Tester",
}
]
}
Comments collection:
{
"_id" : ObjectId("6062178fc73fe806e45c9b69"),
"userId" : "5a54f739fe0a00373e7ef1e9",
'text' : 'this is a test',
"status" : "1",
"timestamp" : ISODate("2021-03-29T18:08:14.317Z")
}
Pipeline
$pipeline = [['$match' => [
'users' => [
'$elemMatch' => [
'field1' => $field1,
],
]
]
],
['$unwind' => '$users'],
['$match' => [
'users.field1' => $field1,
]
],
['$addFields' => ['userId' => ['$toString' => '$userId' ]]],
['$lookup' => [
'from' => 'comments',
'localField' => 'userId',
'foreignField' => 'userId',
'as' => 'userComments'
]
],
['$unwind' => '$userComments'],
['$project' => [
'comments' => [
'$switch' => [
'branches' => [
[ 'case' => [
'$eq' => ['$userComments.status','verified']
],
'then' => 1],
[ 'case' => [
'$lte' => ['$userComments.status', '']
],
'then' => 1],
],
'default' => 0
]
],
'status' => '$users.status',
'total' => [false],
]
],
['$group' => [
'_id' => $groupBy,
'text' => ['$sum' => '$comments'],
'total' => ['$sum' => '$total'],
'completed' => ['$sum' => '$status'],
]
],
];
result
{"_id" :"categories","text": 21,"total": 100,"completed":50}
I am trying to fetch data from mongoDB using lookup
collections companies
{
"id" : 1,
"company_id" : 2,
"user_group" : 1,
"company_name" : "xyz",
"created_on" : "00-00-0000"
}
collection users
{
"id" : 1,
"company_id" : 1,
"user_group" : 1,
"name" : "abcd",
"email" : "abcd#abcd.abcd"
}
{
"id" : 1,
"company_id" : 2,
"active": 1,
"user_group" : 1,
"name" : "efgh",
"email" : "efgh#efgh.efgh"
}
Query used to fetch data using php
$collection->aggregate([
['$match' => ['company_id' => 2]],
['$lookup' => [
'from' => 'users',
'localField' => 'user_group',
'foreignField' => 'user_group',
'as' => 'company_users',
]],
['$unwind' => ['path' => '$company_users', 'preserveNullAndEmptyArrays' => true]],
['$match' => ['$and' => [['company_users.company_id' => 2], ['company_users.active' => 1]]]],
['$project' => [
'_id' => false,
'company_id' => true,
'user_group' => true,
'company_users.name' => true,
'company_users.email' => true
]
]
]);
Query is working correctly but takes more time to retrieve data if document is greater than 1000
I'm using mongo and laravel.
I'm getting data in periods of time, usually 30 days. I want to group the data by day. I've tried $project $dayOfMonth and group by day but it grouping them in order of days in month and I want to be ordered in days in the period.
is there a way?
[
'$match' => [
"created_at" => [
'$gte' => new \MongoDB\BSON\UTCDateTime($thisMonth),
]
],
],
[
'$project' => [
'day' => [
'$dayOfMonth' => [
'date' => '$created_at',
]
],
]
],
[
'$group' => [
'_id' => '$day',
'count' => ['$sum' => 1],
],
],
sample:
{#940
flag::STD_PROP_LIST: false
flag::ARRAY_AS_PROPS: true
iteratorClass: "ArrayIterator"
storage: array:16 [
"_id" => ObjectId {#933
+"oid": "5b2ff00e35826377be16ff82"
}
"orderNumber" => "10000"
"userName" => "dGFraHRlLTkwMDQ2NDcyOJRbugaZlHWdMR+nCNzaUfY="
"updated_at" => UTCDateTime {#938
+"milliseconds": "1529868539000"
}
"created_at" => UTCDateTime {#939
+"milliseconds": "1529868302000"
}
]
}
You can use $dateFromString to converts a date object to a string according to a your desired format.
[ "$match" => [
"created_at" => [ '$gte' => new \MongoDB\BSON\UTCDateTime($thisMonth) ]
]],
[ "$group" => [
"_id" => [ "$dateToString" => [ "format" => "%Y-%m-%d", "date" => "$created_at" ] ],
"count" => [ "$sum" => 1 ]
]]
I have a collection named ad_log with created_at column in unix timestamp format. e.g. created_at: 1469007877.
How can I group my records by year and month of the year in laravel jenssegers mongodb library.
Supposing that your model is AdLog
Try this :
AdLog::raw()->agregate([
[ "$group" => [
"_id" => [
"year" => [ "$substr" => [ "$dateStr", 7, 4 ] ],
"month" => [ "$substr" => [ "$dateStr", 4, 2 ] ]
],
"count" => [ "$sum" => 1 ]
]]
]);
i get an error like
Array
(
[errmsg] => exception: the $unwind field path must be specified as a string
[code] => 15981
[ok] => 0
)
While i used following query for given embedded document (I want to sum of rate_number from my given record of table structure)
global $DB, $mongo;
$theObjId = new MongoId($post_id);
$collection = $mongo->getCollection('mongo_hw_posts');
$rt_sum = $collection->aggregate(
array('$unwind'=>$rate),
array('$group'=>
array(
'_id' => $theObjId
),
array(
'rate_number'=>array('$sum' =>$rate.'rate_number')
))
);
table structure
{
"_id": ObjectId("51ff3b38636e3b9803000001"),
"class_id": NumberInt(2986),
"created_by": NumberInt(1758),
"created_datetime": NumberInt(1375681336),
"deleted": NumberInt(0),
"learn": {
"0": {
"user_id": NumberInt(0),
"learn_date": NumberInt(0)
}
},
"parent_id": "0",
"post_text": "2%20C",
"post_type": "text_comment",
"rate": {
"0": {
"user_id": NumberInt(0),
"rate_date": NumberInt(0),
"rate_number": NumberInt(0)
},
"1": {
"user_id": NumberInt(1457),
"rate_date": NumberInt(1375764137),
"rate_number": NumberInt(3)
},
"2": {
"user_id": NumberInt(1619),
"rate_date": NumberInt(1375764694),
"rate_number": NumberInt(8)
}
},
"serialized_data": "",
"unique_key": "8bdddfe8137d14702b4517f7e8e88ee3",
"user_role": "student"
}
There are a few things wrong with your aggrgation command.
$unwind
Instead of:
array( '$unwind' => $rate ),
You need to use:
array( '$unwind' => '$rate'),
$rate is not just a PHP variable, but a field-value expression in MongoDB.
But you can't use the $unwind like this either, because of:
"errmsg" : "exception: Value at end of $unwind field path '$rate' must be an Array, but is a Object",
That's because rate is:
"rate": {
"0": {
"user_id": NumberInt(0),
"rate_date": NumberInt(0),
"rate_number": NumberInt(0)
},
"1": {
"user_id": NumberInt(1457),
"rate_date": NumberInt(1375764137),
"rate_number": NumberInt(3)
},
"2": {
"user_id": NumberInt(1619),
"rate_date": NumberInt(1375764694),
"rate_number": NumberInt(8)
}
}
But it needs to look like:
"rate": [
{
"user_id": NumberInt(0),
"rate_date": NumberInt(0),
"rate_number": NumberInt(0)
},
{
"user_id": NumberInt(1457),
"rate_date": NumberInt(1375764137),
"rate_number": NumberInt(3)
},
{
"user_id": NumberInt(1619),
"rate_date": NumberInt(1375764694),
"rate_number": NumberInt(8)
}
]
Otherwise, $unwind will not work. You need to change your documents for this.
$group
Your $group is also wrong, instead of:
array('$group'=>
array(
'_id' => $theObjId
),
array(
'rate_number'=>array('$sum' =>$rate.'rate_number')
)
)
You need to make it syntax wise:
array('$group'=>
array(
'_id' => $theObjId
'rate_number'=>array('$sum' => '$rate.rate_number')
)
)
I don't understand why you have:
'_id' => $theObjId
Are you trying to only summarize the rates for one post? If that's the case, you will need to add a $match and change the $theObjId to null, something like this:
$rt_sum = $collection->aggregate(
array( '$match' => array( '_id' => $theObjId ) ),
array( '$unwind' => '$rate' ),
array( '$group'=>
array(
'_id' => null,
'rate_number' => array('$sum' => '$rate.rate_number')
)
)
);
The full example is here:
<?php
$m = new MongoClient;
$c = $m->test->so;
$c->drop();
$post_id = "51ff3b38636e3b9803000001";
$theObjId = new MongoId($post_id);
$c->insert( array(
"_id" => new MongoId("51ff3b38636e3b9803000001"),
"class_id" => 2986,
"created_by" => 1758,
"created_datetime" => 1375681336,
"deleted" => 0,
"learn" => array(
array(
"user_id" => 0,
"learn_date" => 0
)
),
"parent_id" => "0",
"post_text" => "2%20C",
"post_type" => "text_comment",
"rate" => array(
array(
"user_id" => 0,
"rate_date" => 0,
"rate_number" => 0
),
array(
"user_id" => 1457,
"rate_date" => 1375764137,
"rate_number" => 3
),
array(
"user_id" => 1619,
"rate_date" => 1375764694,
"rate_number" => 8
)
),
"serialized_data" => "",
"unique_key" => "8bdddfe8137d14702b4517f7e8e88ee3",
"user_role" => "student"
) );
$rt_sum = $c->aggregate(
array( '$match' => array( '_id' => $theObjId ) ),
array( '$unwind' => '$rate' ),
array( '$group'=>
array(
'_id' => null,
'rate_number' => array('$sum' => '$rate.rate_number')
)
)
);
var_dump ($rt_sum);
And the output is:
array(2) {
'result' =>
array(1) {
[0] =>
array(2) {
'_id' => NULL
'rate_number' => int(11)
}
}
'ok' =>
double(1)
}