使用合并从 2 个表中返回第一个非空值

Question

I have 2 tables in my Postgres db. Each of them has a URL column that I want to query and return the first non-empty value. Table

test1 
id  name   url 

test2 
id   address url 

I have the id as an input. How can I write a query to coalesce the latest available url for any id?

test 1
ID   name         url 
1     A           null     
2     B           abc.com
3     C           xyz.com

test 2
ID   address         url 
1     X            return.com     
2     Y            null
3     Z            efg.com

After the coalesce I expect return.com to be returned for id = 1 By latest available URL I mean the first non-null entry.

Answer 1

From the comments, it's hard to determine what "latest" and "first" mean, as COALESCE does not know how to distinguish that (basically, you have to define the priority). If your example seeks return.com , then the proper syntax is:

SELECT COALESCE(b.url,a.url) FROM test_1 a JOIN test2 a ON a.id=b.id WHERE a.id=1

So in this example, you will always show b.url as long as it is not null. If you want it to dynamically prioritize a.url and b.url based on some time-based criteria, you'll need to write a plpgsql function or do some convoluted array nesting and string concatenation (or I'd want to see some other experts show off their chops here)

Disclosure: I work for EnterpriseDB (EDB)