On May 1, 10:47 pm, workoutlooktest....@gishpuppy.com wrote:
> Hello all,
> I have a table with 2 issues, issue_A and issue_B, a table with action
> items related to the issues,
>
> issue_A AI_A1
> issue_A AI_A2
> issue_A AI_A3
> issue_B AI_B1
> issue_B AI_B2
>
> and network elements that the issues affect:
>
> issue_A NE_A1
> issue_A NE_A2
> issue_B NE_B1
> issue_B NE_B2
> issue_B NE_B3
>
> and I would like a query whose result juxtaposes those 2 tables, as
> follows:
>
> issue_A AI_A1 NE_A1
> issue_A AI_A2 NE_A2
> issue_A AI_A3 NULL
> issue_B AI_B1 NE_B1
> issue_B AI_B2 NE_B2
> issue_B NULL NE_B3
>
> where the top and bottom NULLs could also be NE_A2, AI_B2
> respectively, I wouldn't mind.
> Any idea how to do that?
> Thanks, Robin
----------------------------------------
t=# select * from ai; i | ai
---------+-------issue_A | AI_A1issue_A | AI_A2issue_A | AI_A3issue_B | AI_B1issue_B | AI_B2
t=# select * from ne; i | ne
---------+-------issue_A | NE_A1issue_A | NE_A2issue_B | NE_B1issue_B | NE_B2issue_B | NE_B3
t=# SELECT COALESCE(ai.i, ne.i) AS i, ai.ai, ne.ne
t-# FROM ai FULL JOIN ne ON
t-# REPLACE(ai.ai, 'AI', '') = REPLACE(ne.ne, 'NE', ''); i | ai | ne
---------+-------+-------issue_A | AI_A1 | NE_A1issue_A | AI_A2 | NE_A2issue_A | AI_A3 |issue_B | AI_B1 | NE_B1issue_B
|AI_B2 | NE_B2issue_B | | NE_B3
----------------------------------------
HINT: You might want to normalize your tables further to avoid
unnecessary string handling.
