Formula syntax

Written by

Dataset fields in calculations
Constants
Operators
Formatting formulas
Comments
Logical operations
Strings
Converting types
Aggregation
Text markup

The syntax of calculated expressions in Yandex DataLens is similar to SQL.
Simple expressions can be described using ordinary arithmetic operations:

([Sales] - [Profit]) / 10

[Date] - #2019-01-24#

More complex expressions use multiple functions to implement various calculations, aggregations, and conversions of data from one type to another:

CONCAT(SUM([Category Sales]) / [Total Sales], ' %')

DATETRUNC([datetime], 'month')

Dataset fields in calculations

The syntax used for accessing dataset fields is similar to Transact-SQL, but in Yandex DataLens, the field name must be enclosed in square brackets ([]):

[Field name]

Constants

In addition to fields, operators, and functions, expressions can include constants of different data types:

Integer: 23, -4325653.
Fractional number: 0.0234, -1.0.
Date: #2020-01-01#.
Date and time: #2020-01-01 11:15:00#.
String: "String".
Boolean: TRUE, FALSE.
Geopoint: GEOPOINT("[55.7912,37.6872]").

Operators

Operators available in expressions:

Arithmetic: +, -, *, /.
```
([Sales per Order] * [OrderCount]) - [Profit]
```
```
([Profit] / [Cost Price]) * 100
```
```
[City] + " " + "city"
```
Addition (+), subtraction (-), and multiplication (*) operators behave differently depending on the argument type.
Exponentiation: ^.
```
[Mass] * [Speed] ^ 2
```
Remainder of a division: %.
```
[Sales] % 10 + [Cost Price] % 10
```
Boolean: AND, OR, NOT, IN, LIKE, IS TRUE, IS FALSE, and BETWEEN.
Comparisons: =, !=, <, <=, >, and >=.

Comparison operators let you create logical chains:
```
1 > x > -4 > y != 8
```

Full description of all operators.

Formatting formulas

Any formula can be written in one or several lines:

CONCAT(
    SUM([Category Sales]) / [Total Sales],
    ' %'
)

Comments

Comments are used to add explanations or to ignore parts of formulas:

One-line comment.
```
-- This is a one-line comment
```
Block comment.
```
/* This is a block
comment */
```

Logical operations

Logical functions are used for branching calculations in expressions:

CASE.

CASE [ProductID]
    WHEN 1 THEN "Bananas"
    WHEN 2 THEN "Apples"
    WHEN 3 THEN "Pears"
    ELSE "Other"
END

CASE( 
    [Color],
    "R", "Red",
    "G", "Green",
    "B", "Blue",
    "Not RGB" 
)

Full description of the CASE logical function.

IF.

IF([MassIndex] BETWEEN 18.5 AND 25, "Normally", "Not normal")

IF
    [Year] % 400 = 0 OR ([Year] % 4 = 0 AND [Year] % 100 != 0)
        THEN "Leap year"
    ELSE "Ordinary year"
END

IF
    [City] = "Moscow"
        THEN "This is the Capital"
    ELSEIF [City] = "St. Petersburg"
        THEN "This is the northern Capital"
    ELSE "Other city"
END

Full description of the IF logical function.

IFNULL, ISNULL, and ZN.

IFNULL([Cost Price], 10) * [OrderCount]

Full description of the IFNULL logical function.

IF(ISNULL([Product Name]) = TRUE, "Unnamed", [Product Name] + " " + [ProductID])

Full description of the ISNULL logical function.

ZN([Total Sales]) - ZN([Total Cost])

Full description of the ZN logical function.

Strings

The following string functions are used for processing text data:

CONCAT([Total Sales], "$")

IF(CONTAINS([Product Name], "RU"), [Product Cost] + " " + "RUB", [Product Cost] + " " + "USD")

REPLACE([OrderID], "2020", [Month])

IF(STARTSWITH([Region Name], "RU_"), SPLIT([Region Name], "_", 2), [Region Name])

Strings can be enclosed in single or double quotes. In this case, one type of quotation mark can be used inside the other:

FIND([Product Name], 'plus')

CONCAT('"', [Product Name], '"')

You can make different conversions of string data using special characters in formulas:

REPLACE([ShopAddress], "\n", " ")

"File path" + " = " + "\\" + [Folder] + '\\' + [Filename]

Note

Special characters such as \n, \t, and \r do not affect the display of the source data.

Converting types

Expression values can be converted from one type to another:

FLOAT([StringWithNumbers])

DATETIME(STR([Order Date]) + "-" + STR([Order Time]))

GEOPOINT([Latitude],[Longitude])

Full description of the type conversion functions.

Aggregation

To calculate the resulting values, use aggregate functions:

AVG([Check Total]) * COUNTD([CustomerID])

SUM_IF([Sales], [Category] = "Fruits")

IF
    MIN[Date] = #2020-01-01#
        THEN SUM[Sales] * 1.1
    ELSE SUM[Sales] * 1.2
END

Window functions let you aggregate values from a group of strings, without combining these strings into one. This distinguishes them from aggregate functions. You can also use window functions to calculate values for one string in the context of values from other strings.

SUM([Sales] WITHIN [PaymentType]) / SUM([Sales] TOTAL)

MSUM([Sales per Order], 1 WITHIN [ProductID] ORDER BY [Price])

RANK_DENSE(AVG([Price]), "desc" WITHIN [ShopID] BEFORE FILTER BY [PaymentType])

Window functions support grouping and sorting of records and the BEFORE FILTER BY option.

Text markup

To create text with markup, use markup functions:

MARKUP(BOLD("This"), " ", ITALIC("is"), " ", URL("https://example.com/", [LinkName])))

To use numeric constants when marking up URLs, convert them to the String type:

URL("https://example.com/?value=" + STR([Value]), [Value]))