Skip to main content

Hello,

I have a following statement:

case when ($(DB_OPERATION) = 'INSERT')

then (
select max(uot.id)+1 as id, i.geometry_type as geometry_type, i.name as name,
i.name as label, i.name as label_mask
from unip_object_type uot
cross join Inserted i)

else select * from Inserted i

end

When I run this in my workspace, I get the following error:

InlineQuerier(InlineQueryFactory): Error preparing database query: near "case": syntax error Unable to execute statement

InlineQuerier(InlineQueryFactory): Database was unable to prepare query 'case when (UPDATE = 'INSERT') then (

select max(uot.id)+1 as id, i.geometry_type as geometry_type, i.name as name,

i.name as label, i.name as label_mask

from unip_object_type uot

cross join Inserted i)

else select * from Inserted i

end

'.

InlineQueryFactory instance was not properly defined. Consult logfile for details.

InlineQuerier(InlineQueryFactory): A fatal error has occurred. Check the logfile above for details

A fatal error has occurred. Check the logfile above for details

It has to be my mistake in the case statement. When I run the query just as

select max(uot.id)+1 as id, i.geometry_type as geometry_type, i.name as name,

i.name as label, i.name as label_mask

from unip_object_type uot

cross join Inserted i

It works. However I need to run this query only when $(DB_OPERATION) (user parameter) is set to INSERT, when it is UPDATE, I need to only do select.

 

 

 

Any help is really welcome.

Thanks,

Radek

One thing I have tried was changing the word INSERT to 'INSERT' or [insert]... none of them works


Hi @drakez, as far as I know, CASE clause cannot be used to select a SELECT statement that should be executed. I would use a Tester to branch the data flow into two streams depending on the "DB_OPERATION" parameter value, and then use two InlineQueliers for each SELECT statement.


Hi @drakez, as far as I know, CASE clause cannot be used to select a SELECT statement that should be executed. I would use a Tester to branch the data flow into two streams depending on the "DB_OPERATION" parameter value, and then use two InlineQueliers for each SELECT statement.

Hello and thanks @takashi!

My problem is - I know I can filter data based on DB_OPERATION value, however - I am trying to join two tables, where I have to use cross join (they have no common attributes). Any time I use cross join, even though one table basically does not exist, I get output data with "no_schema" - I need to have no output, not output with no_schema.

I have no idea if this clarifies what I am trying to do. When input is UPDATE - I need to merge two tables using some columns from original table, some columns from UPDATE table, when the input is INSERT, I am forced to use cross join (no common columns), so my inserted feature has id value based on max id from original table increased by one.


Reply