CODEWITHSUNDEEP
azure-synapseazure-data-factory
DataSteve
DataSteve

Reputation: 61

mapping data flow column pattern type =='decimal' not changing decimal columns

I'm using a column pattern to catch nulls. My logic is very simple.

Matching condition

type=='decimal'

Column name expression

$$

Value expression

coalesce($$,toDecimal(0,38,18))

I can't get this to work it simply leaves NULL values in place. I can however use the expression

type!='string' && type!='date'

This does change the value of the columns with NULL values to 0. In my current case this is fine I have strings, dates and decimals but I can see how this is not ideal as I might have integers, doubles or other types and I would not want these to be converted to decimals and the list of && statements becomes quite long.

Can anyone supply details of how I should specify a decimal type or a better workaround if not? If there is a link to any good references would be helpful too, the official documentation is not especially specific.

To confirm I am working with decimal values: showing decimals

Upvotes: 2

Views: 1980

Answers (3)

gmonster1st
gmonster1st

Reputation: 95

I know this is an old post but I stumbled in the same issue and I fixed it by just assigning the precision and scale like this type=='decimal(38,20)' just note that the arguments go like (scale, precision). This worked for me.

Upvotes: 0

Howard Renollet
Howard Renollet

Reputation: 4739

You can also use typeMatch(type, 'number') if you want to match all number types. Per the documentation, other options are:

typeMatch(type, 'integral') /* matches short, integer, long */
typeMatch(type, 'fractional') /* matches double, float, decimal */
typeMatch(type, 'datetime') /* matches date, timestamp */

Upvotes: 1

Kiran-MSFT
Kiran-MSFT

Reputation: 234

Try this instead - Decimal has precision and scale

startsWith(type, 'decimal')

Upvotes: 3

Related Questions