NUMVALUE Function
Converts a string formatted as a number into an Integer or Decimal value by parsing out the specified decimal and group separators. A string or a function returning formatted numbers of String type or a column containing formatted numbers of string type can be inputs.
You can use this function to convert String values that have locale-specific formatting to locale-independent values of Integer or Decimal type by removing the formatting separators.
If the source value does not include a valid input for this function, a missing value is returned.
This function supports negative input values.
Supports input for multiple locale types and returns the output of the appropriate numeric type.
You can use decimal separators and grouping separators when working with other currency formats. More information is below.
Nota
If the decimal separator and group separator arguments are not specified, then a null value is returned.
Wrangle vs. SQL: This function is part of Wrangle, a proprietary data transformation language. Wrangle is not SQL. For more information, see Wrangle Language.
Basic Usage
Numvalue(<code>Discount</code>, "<code>,</code>" ,"<code>."</code>)
Output: Returns the values from the Discount
column by converting the formatted number to a numeric value using decimal and group separators. For example, if the Discount
column has values in percentage, then it is converted into the corresponding numeric value. Example: 10% to 0.10.
Nota
If multiple percent signs are used in a column, then this function returns the decimal values accordingly. For example, an input of 9%%
returns 0.0009
.
Nota
This function removes any currency indicators. For example, £ 100
as an input value returns 100
.
Syntax and Arguments
<code class="listtype-code listindent1 list-code1 lang-bash">Numvalue(formattedNumber, ["grouping_separator"], ["decimal_separator"])</code>
Argument | Required? | Data Type | Description |
---|---|---|---|
formattedNumber | Y | string | A string, a column of strings, or a function returning a string |
grouping_separator | Y | string | A grouping representing grouping separator. By default, comma (,) is used as the grouping separator. |
decimal_separator | Y | string | A string used to separate the integer and fractional part of the result. If not specified, then a null value is returned. |
For more information on syntax standards, see Language Documentation Syntax Notes.
formatted_number
Literal string value, a function that returns a string value, or the name of a column containing string values to be converted into a numeric value.
Multiple columns and wildcards are not supported.
Usage Notes:
Required? | Data Type | Example Value |
---|---|---|
Yes | string, a column of strings, or a function returning a formatted number string or column of strings | Discount |
grouping_separator
The string used to group a set of digits in the input values. Separators must be enclosed with double quotes ("
) or single quotes ('
). For example, a comma (,
) is used as a grouping separator in the U.S.A. (“10,000”), whereas space is used in France (“10 000”).
Nota
When you provide invalid separators or wrong separators, you may get an error in the Formula column.
Usage Notes:
Required? | Data Type | Example Value |
---|---|---|
Yes | String literal | ',' |
decimal_separator
The string used to indicate the decimal point in the input values. Separators must be enclosed with double quotes ("
) or single quotes ('
).
A decimal separator is used to separate the fractional part of a number written in decimal form. For example, a period (.
) is used as a decimal separator in the U.S.A. ("1234.12"), whereas comma (,
) is used in France ("1234,12").
Usage Notes:
Required? | Data Type | Example Value |
---|---|---|
Yes | String literal | '.' |
Grouping Separator | Decimal Separator | Example Locale |
---|---|---|
Comma ( | Period ( | U.S locale |
Period ( | Comma ( | Spanish locale |
Space | Comma ( | French locale |
Examples
Suggerimento
For additional examples, see Common Tasks.
Example - formatting price and percentages
This example shows how to convert the formatted number of string type into a numeric value. The following table shows different types of products and their total sales in the UK region. From this example, you must convert the total sales value for the U.S region.
First, you must convert the global currency formats into a generic numeric value, then proceed with the conversion calculation.
Source:
Products | Total_Sales_UK |
---|---|
Baby Foods | £ 100.00 |
Medicines | £ 150.00 |
Groceries | £ 200.00 |
Kitchen Supplies | £ 25.00 |
Cosmetics | £ 250.00 |
Snacks | £ 50.00 |
Transformation:
The first transformation is to convert the Total_Sales_UK
column into a numeric value. In this case, when a Numvalue
function is applied, the value with the currency symbols returns only the numeric value. You can see the currency symbol is removed when using the Numvalue
function.
Transformation Name | |
---|---|
Parameter: columns | Total_Sales_UK |
Parameter: Formula | NUMVALUE (Total_Sales_UK, ",",".") |
The second step consists of converting the U.K Pounds (£) to U.S Dollars ($) for better computations. In this step, you multiply the Total_Sales_UK
column with the ratio conversion rate. Let's say the current conversion rate is 1.36, then multiply the U.K Pounds with 1.36 to convert to U.S Dollars. As a part of this transformation, the Total_Sales_US
column is created.
Transformation Name | |
---|---|
Parameter: Formula type | Single row formula |
Parameter: Formula | Total_Sales_UK * 1.36 |
Parameter: New column name | Total_Sales_US |
After you get the U.S conversion, you can format the values using the Numformat
function. The Numformat
function formats a numeric set of values according to the specified number formatting.
Transformation Name | |
---|---|
Parameter: columns | Total_Sales_US |
Parameter: Formula | NUMFORMAT(Total_Sales_US, '$###,###.00', ',', '.') |
Results
The output data should look like the following:
Products | Total_Sales_UK | Total_Sales_US |
---|---|---|
Baby Foods | £ 100.00 | $136.00 |
Medicines | £ 150.00 | $204.00 |
Groceries | £ 200.00 | $272.00 |
Kitchen Supplies | £ 25.00 | $34.00 |
Cosmetics | £ 250.00 | $340.00 |
Snacks | £ 50.00 | $68.00 |