How to group data by time period? - php

Question

I have data on invoices and I want to group them by time period wall. for example 0-10 and 11-20:
I need to solve them with $aggregate in mongoDB.
{
'_id': '1',
'value': 10,
'due_date': '20221001'
},
{
'_id': '2',
'value': 10,
'due_date': '20221012'
},
{
'_id': '2',
'value': 10,
'due_date': '20221030'
},
I need to group by period 0-10 days, 11-20 days and sum the values: For the example above the result would be:
[{
"_id": '0-10 days',
"total": 10,
},
{
"_id": '11-20 days',
"total": 10,
},
{
"_id": '>20 days',
"total": 10,
}]
I try with:
['$facet' => [
['due_date_one' => [
['$match' => [
'due_date' => [
'$gt' => new UTCDateTime((new Carbon())-> subDays(100) -> getTimestamp()),
'$lte' => date('Y-m-d', strtotime('now'))
]
]
]],
]]

Answer 1

You could use $bucket or add a few $addFields and $group stages:
Example mongo playground - https://mongoplayground.net/p/B0zl_GGH4sG
Example Documents:
[
{
"_id": "1a",
"value": 10,
"due_date": "20221001"
},
{
"_id": "1b",
"value": 5,
"due_date": "20221102"
},
{
"_id": "1c",
"value": 7,
"due_date": "20221102"
},
{
"_id": "2a",
"value": 10,
"due_date": "20221012"
},
{
"_id": "2b",
"value": 7,
"due_date": "20221113"
},
{
"_id": "2c",
"value": 8,
"due_date": "20221113"
},
{
"_id": "3a",
"value": 10,
"due_date": "20221030"
},
{
"_id": "3b",
"value": 9,
"due_date": "20221131"
},
{
"_id": "3c",
"value": 11,
"due_date": "20221131"
}
]
Aggregation query:
db.collection.aggregate([
{
$addFields: {
day: {
$toInt: { $substr: [ "$due_date", 6, 2 ] }
}
}
},
{
$addFields: {
bucketDate: {
$switch: {
branches: [
{ case: { $gt: [ "$day", 20 ] }, then: ">20 days" },
{ case: { $gt: [ "$day", 10 ] }, then: "11-20 days" }
],
"default": "0-10 days"
}
}
}
},
{
$addFields: {
bucketDateWithMonth: {
$concat: [
{ $substr: [ "$due_date", 0, 6 ] },
" ",
"$bucketDate"
]
}
}
},
{
$group: {
//_id: "$bucketDate", //No grouped month
_id: "$bucketDateWithMonth", //With grouped month
count: { $sum: 1 },
value: { $sum: "$value" }
}
}
])
Output: (grouped month)
[
{
"_id": "202210 0-10 days",
"count": 1,
"value": 10
},
{
"_id": "202211 0-10 days",
"count": 2,
"value": 12
},
{
"_id": "202210 11-20 days",
"count": 1,
"value": 10
},
{
"_id": "202211 11-20 days",
"count": 2,
"value": 15
},
{
"_id": "202210 \u003e20 days",
"count": 1,
"value": 10
},
{
"_id": "202211 \u003e20 days",
"count": 2,
"value": 20
}
]
Output: (No grouped month)
[
{
"_id": "\u003e20 days",
"count": 3,
"value": 30
},
{
"_id": "0-10 days",
"count": 3,
"value": 22
},
{
"_id": "11-20 days",
"count": 3,
"value": 25
}
]