Replacing null value with non-null value in a column
Hi All,
I would require your help dealing with empty values in a column
I have a file with "n" features. In one of the column (Example: Column A) certain values are null/empty. I would like to replace it with non-empty value. Also, Column A has only 2 distinct values - either null or date.
Example:
Feature Column A 1 Jan 2025 2 Jan 2025 3 Jan 2025 4 null 5 null 6 null 7 Jan 2025 8 Jan 2025
I would like to change Feature 4,5,6 to "Jan 2025" instead of null.
In Java world, I would store them in local variable and run column A in loop and replace with value if null. Not sure how to perform similar scenario in FME. Would really appreciate your help in achieving the above expected output.
Thanks in advance
Page 1 / 1
Have a look at the NullAttributeMapper. This will allow you to replace attributes with a defined value (in your case null) with another value.
Thinking about the above, that is fine if you’re expecting to replace all nulls with the same value. However if you data is more similar to the below, where it is ordered and you have some rows missing the date, that first approach won’t work.
This now allows you to get the value from the previous feature that has passed through the transformer. Add an Output Attribute called date with a conditional value
In the conditional value we want to check if the current value is null, if it is, take the previous features date. If the value is not null, take no action
Thank you so much for taking time to respond to my question.
Attribute creator works well if there is a non-null value in previous feature
However, if there are 2 or more null values next to each other, the result of new attribute is null as well
Example:
Feature date new_date 1 Jan 2025 Jan 2025 2 null Jan 2025 3 null null 4 Jan 2025 Jan 2025 5 null Jan 2025
In the above example - "date" is existing column and "new_date" is the column I created using Attribute creator "new_date" value for feature 3 is null because date value of feature 2 (previous feature) is null
1) Any thoughts on how I can handle this ?
2) Also, Just for your information, values in "date" column will be the same. it will not have different date. Is there any way to store non-null value in random using some variable and replace the null columns ?
Thanks,
you need to reference the same field you’re updating. In the above example:
Feature 1 has a date - do nothing
Feature 2 is null, look at date, date is Jan 2025, set new date to Jan 2025
Feature 3 is null, look at date, date is null, set new date to Jan 2025
….
If you want them to populate in new_date, the easiest way would be to copy date to new_date first using an AttributeCopier
2) Also, Just for your information, values in "date" column will be the same. it will not have different date. Is there any way to store non-null value in random using some variable and replace the null columns ?
Thanks,
I think in this case you want to use the NullAttributeMapper as @hkingsbury initially suggested
