Referencing Existing SAS Table in Teradata Execute

[SASRt]

Hello all,

I've tried searching the site but can't quite figure out what I am missing. I have successfully created a table in SAS that I would like to join with a table in SQL I am accessing through Teradata. I have tried to create a temporary table, but this doesn't seem to be working.

I have also tried to reference the location I have saved the SAS data file to, but this doesn't seem to be working within the Teradata execute().

The SAS code I am using to create a SAS data set:

proc sql;
connect to teradata as tera(server=? user=? password=? mode=teradata connection=global);
create table gold.apple as
select distinct code
from loc1.codes
where product = 'big apples'
group by codes;

I would like the below code to pull my previously saved SAS data set and join it with an existing SQL table in Teradata.

execute(
create multiset volatile table ig as (
select distinct s.1,
     g.code,
     s.2,
     s.3
from gold.apple as g left join silver.apple as s
group by code)
with data primary index (code) on commit preserve rows)
by tera;

When I try and run everything I get the following error: ERROR: Teradata execute: Database 'gold'  does not exist. I have tried the steps suggested here: https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-Access-to-Teradata-How-to-create-Temporary-t..., but it ends up saying I don't have the rights to create a table. I'm sure there's a way to create a linkage, I have a SAS data set and I can pull from Teradata without any issues. I'm hoping someone can point me in the right direction.

Thanks! 

[Kurt_Bremser · Accepted solution]

Teradata can only "see" its own tables.

You will have to create a libname that points to the Teradata database, and use your SAS code to create the dataset there:

libname gold /* all the necessary options to connect to the Teradata database "gold" */;

proc sql;
create table gold.apple as
select distinct code
from loc1.codes
where product = 'big apples'
group by codes;
quit;

Now you can do

proc sql;
connect to teradata as tera(server=? user=? password=? mode=teradata connection=global);

execute(
  create multiset volatile table ig as (
    select distinct s.1,
      g.code,
      s.2,
      s.3
    from gold.apple as g left join silver.apple as s
    group by code
  )
  with data primary index (code) on commit preserve rows
)
by tera;

quit;

as the dataset is now available as a Teradata table in the database.