$locf
The $locf operator propagates the last observed non-null value forward within a partition in a windowed query.
$locf
The $locf operator propagates the last observed non-null value forward within a partition in a windowed query. The $locf operator is particularly useful in filling missing data points in time-series data or other datasets with gaps.
Parameters
| Parameter | Description |
|---|---|
input | The expression that resolves to the field whose value you want to propagate. |
sortBy | A document that specifies the sort order for the partition. |
Examples
Consider this sample document from the stores collection.
{
"_id": "0fcc0bf0-ed18-4ab8-b558-9848e18058f4",
"name": "First Up Consultants | Beverage Shop - Satterfieldmouth",
"location": {
"lat": -89.2384,
"lon": -46.4012
},
"staff": {
"totalStaff": {
"fullTime": 8,
"partTime": 20
}
},
"sales": {
"totalSales": 75670,
"salesByCategory": [
{
"categoryName": "Wine Accessories",
"totalSales": 34440
},
{
"categoryName": "Bitters",
"totalSales": 39496
},
{
"categoryName": "Rum",
"totalSales": 1734
}
]
},
"promotionEvents": [
{
"eventName": "Unbeatable Bargain Bash",
"promotionalDates": {
"startDate": {
"Year": 2024,
"Month": 6,
"Day": 23
},
"endDate": {
"Year": 2024,
"Month": 7,
"Day": 2
}
},
"discounts": [
{
"categoryName": "Whiskey",
"discountPercentage": 7
},
{
"categoryName": "Bitters",
"discountPercentage": 15
},
{
"categoryName": "Brandy",
"discountPercentage": 8
},
{
"categoryName": "Sports Drinks",
"discountPercentage": 22
},
{
"categoryName": "Vodka",
"discountPercentage": 19
}
]
},
{
"eventName": "Steal of a Deal Days",
"promotionalDates": {
"startDate": {
"Year": 2024,
"Month": 9,
"Day": 21
},
"endDate": {
"Year": 2024,
"Month": 9,
"Day": 29
}
},
"discounts": [
{
"categoryName": "Organic Wine",
"discountPercentage": 19
},
{
"categoryName": "White Wine",
"discountPercentage": 20
},
{
"categoryName": "Sparkling Wine",
"discountPercentage": 19
},
{
"categoryName": "Whiskey",
"discountPercentage": 17
},
{
"categoryName": "Vodka",
"discountPercentage": 23
}
]
}
]
}Example 1: Using $locf to fill missing time series data
To propagate the most recent non-null value for the lastUpdated field across stores within the "First Up Consultants" company, first run a query to partition the stores within the company. Then, use the $lecf operator to propagate the last non-null value for the field for all stores within the partition.
db.stores.aggregate([{
"$match": {
"company": {
"$in": ["First Up Consultants"]
}
}
}, {
"$setWindowFields": {
"partitionBy": "$name",
"sortBy": {
"sales.revenue": 1
},
"output": {
"lastUpdatedDate": {
"$locf": {
"lastUpdated": 1
}
}
}
}
}, {
"$project": {
"company": 1,
"name": 1,
"lastObservedDiscount": 1
}
}])The first three results returned by this query are:
[
{
"_id": "0f4c48fe-c43b-4083-a856-afe6dd902077",
"name": "First Up Consultants | Appliance Bargains - Feilmouth",
"company": "First Up Consultants"
},
{
"_id": "c4883253-7ccd-4054-a7e0-8aeb202307b5",
"name": "First Up Consultants | Appliance Bargains - New Kari",
"company": "First Up Consultants"
},
{
"_id": "a159ff5c-6ec5-4ca8-9672-e8903a54dd90",
"name": "First Up Consultants | Appliance Bargains - Schadenstad",
"company": "First Up Consultants"
}
]