# Collections
Collections are similar to what Excel users would called a Lookup, and you'll use them to extract variables from large datasets based on the inputs you’ve already been given.
For example, using a collection, you'd look down the first column of a dataset to find your key in a series of ABI codes, and you might then pick one particular rating from a larger table.
The output must resolve to a value (e.g. number, string, boolean, date). Collection queries must start with the word ***'collection'.***
## Using inputs from previous steps
The input value is wrapped in `{{ }}`**.**
For example, if the input is called `customer_age` and is equal to 30 in the quote, then:
'collection.filter(age<{{customer_age}}).map(age).first()' // 24
// resolves to
'collection.filter(age<30).map(age).first()' // 24
## Methods for use in Collections
#### `count()`
Counts the number of items in an response. It can be used after a filter has been applied.
*Example:*
name
age
is_male
bob
30
true
sue
24
false
jack
40
true
```javascript
'collection.count()' // 3
'collection.filter(age<30).count()' // 1
'collection.filter(is_male=true).filter(age>25).count()' // 2
```
#### `min(), max(), mean(), range(), sum()`
Finding the min number, max number, mean number, sum number and range between the two.
*Example:*
name
age
is_male
bob
30
true
sue
24
false
jack
40
true
```javascript
'collection.map(age).min()' // 24
'collection.map(age).max()' // 40
'collection.map(age).range()' // 16 i.e. 40 - 24
'collection.map(age).mean()' // 31.3333333333
'round(collection.map(age).mean(), 2)' // 31.33
'collection.map(age).sum()' // 94
```
#### `map()`
Returns an array of the selected column within a collection of items. You can not return an array but you can have it as a chained step to do work on a collection.
*Example*:
name
age
is_male
bob
30
true
sue
24
false
jack
40
true
```javascript
'collection.map(age)' // [30, 24, 40]
'collection.map(name)' // ["bob", "sue", "jack"]
// You must resolve the array to make an output
'collection.map(name).first()' // "bob"
```
#### `filter()`
Returns a filtered collection based on a columns criteria.
Filters will work on columns with strings, booleans and numbers. Filters can be chained.
You can also filter against value or empty/null data with the \~ flag
*Example*:
name
age
is_male
surname
bob
30
true
brown
sue
24
false
jack
40
true
brown
```javascript
'collection.filter(age>30).map(name)' // ["jack"]
'collection.filter(age<40).filter(is_male=true).map(name)' // ["bob"]
'collection.filter(age<40).filter(age>24).map(name)' // ["bob"]
'collection.filter(surname=brown~).count()' // 3 as the ~ returns empty cells too
// You must resolve the array to make an output
'collection.filter(age>30).map(name).first()' // "jack"
```
#### `unique`
Returns an array of the unique items from a column
*Example:*
name
age
is_male
bob
30
true
sue
24
false
jack
40
true
```javascript
'collection.unique(age).count()' // 3
'collection.unique(age).map(age).sum()' // 94
'collection.unique(is_male).count()' // 2
```
#### `first, last`
Returns an first or last element of an array as a value. This is a common way to resolve a result.
*Example:*
name
age
is_male
bob
30
true
sue
24
false
jack
40
true
```javascript
'collection.map(age).first()' // 30
'collection.map(age).last()' // 40
'collection.filter(age<30).map(age).first()' // 24
```
#### `exists`
Checks if property value exists i.e. not null, undefined, empty string or 0.
*Example:*
name
age
is_male
bob
30
true
sue
24
false
jack
40
true
```javascript
'collection.filter(age>50).exists()' // "false"
'collection.filter(age<50).exists()' // "true"
```
#### `date, age`
If property is a date, then you can use date() to find a hour, date, month or year from the date string, or use age() to find the age since now in hours, days, months or years.
You can format a date from US or UK date formats.
*Example*:
name
dob
is_male
dob_format
bob
1993-12-12
true
12/12/1993
sue
1999-01-01
false
01/01/1999
jack
1983-11-11
true
11/11/1983
```javascript
'collection.map(dob).first().date(YY)' // 1993
'collection.map(dob).first().date(MM)' // 12
'collection.map(dob).first().date(DD)' // 12
'collection.map(dob_format).first().date(UK)' // 1993-12-12
'collection.map(dob_format).first().date(US)' // 1993-12-12
'collection.map(dob).first().age(YY)' // 30
'collection.map(dob).first().age(MM)' // 360
'collection.map(dob).first().age(DD)' // 131800
```
A second argument can be provided to point at a date in the object instead of using now.
```javascript
'collection.map(dob).first().age(YY, "2020-01-01")' // 27
```
Date and age work with arrays.
```javascript
'collection.map(dob).age(YY)' // [ 30 , 24, 40 ]
```
#### `postcode`
If property is a string postcode, then you can find the area and sector.
Example:
name
dob
postcode
bob
1993-12-12
SW1A1AA
sue
1999-01-01
L201AR
jack
1983-11-11
M123RT
```javascript
'collection.map(postcode).first().postcode(area)'); // "SW"
'collection.map(postcode).first().postcode(sector)'); // "SW1A"
'collection.map(postcode).postcode(area)'); // ["SW", "L", "M"]
```
#### `regex`
If property is a string, then you can use regex() to parse a regular expression against the string and return the first match.
Example:
```javascript
'collection.map(postcode).first().regex(^[A-Za-z]{2}|^[A-Za-z]{1})'); // W
'collection.map(postcode).first().regex(^[A-Za-z]{2}|^[A-Za-z]{1})'); // PO
```
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://swallow-1.gitbook.io/swallow/your-first-model/essential-concepts/collections.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
