Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi all,
I’m facing an issue with decimal values coming from VSAM via ARC Studio and replicated to BigQuery.
NUMSTRING_U123,12123So the decimal part is lost during replication.
Has anyone experienced this behavior?
Is there a recommended way to handle comma-based decimal values (e.g. replace with dot or change data type) before replication?
Thanks in advance!
Hello Domenico_Arcamone,
Thank you for reaching out to the Qlik community,
This is expected behaviour. Your value 123,12 is treated as a string / localized number. BigQuery expects a dot (.) as decimal separator and interprets a comma as thousands separator → so 123,12 becomes 123. Also, Qlik Replicate may map such fields as STRING (e.g. NUMSTRING_U), losing decimal semantics unless explicitly cast.
The best fix for this:-
Recommendation: Convert to proper numeric (DECIMAL) before replication to avoid data loss.
Hello Domenico_Arcamone,
Thank you for reaching out to the Qlik community,
This is expected behaviour. Your value 123,12 is treated as a string / localized number. BigQuery expects a dot (.) as decimal separator and interprets a comma as thousands separator → so 123,12 becomes 123. Also, Qlik Replicate may map such fields as STRING (e.g. NUMSTRING_U), losing decimal semantics unless explicitly cast.
The best fix for this:-
Recommendation: Convert to proper numeric (DECIMAL) before replication to avoid data loss.
Hi Rahul,
thanks for your reply, very clear.
Just to add: I also tried converting the field to a decimal type directly in ARC Studio, but in that case the value with comma (e.g. 123,12) is not correctly handled and the decimal part is still not preserved.
So I was wondering if there is any alternative way to handle this directly in ARC (e.g. field definition, formatting, or transformation), instead of applying the conversion in Replicate or BigQuery.
Thanks again for your support!