select id_a
from table1
except
select t1.id_a
from table1 t1
inner join table2 t2
on t1.id_b=t2.id_b
where t2.c1 = F9.00
and t2.ds = G

此代码将在底部获取不需要的ID，在顶部获取所有ID，然后将它们排除在外

像这样：

select t1.id_a
from t1
  join t2 on t1.id_b = t2.id_b
where (t2.cl, t2.ds) <> ('F9.00', 'G')
group by t1.id_a
having count(*) = (select count(*) 
                   from t1 as t3
                   where t3.id_a = t1.id_a);

这会得到没有 'F9.00', 'G' 存在，并且只存在其结果行数与该行的总行数相同的那些 id_a .
在指定平台后编辑：
上面是ansisql-我不知道kylin是否支持ansisql。如果 where (t2.cl, t2.ds) <> ('F9.00', 'G') 不起作用，你需要使用 where t2.cl <> 'F9.00' and t2.ds <> 'G'

你可以用 CONCAT 在c1和ds列上：

SELECT DISTINCT id_a
FROM T1 INNER JOIN T2
ON T1.id_b = T2.id_b
WHERE id_a NOT IN
(SELECT id_a
 FROM T1 INNER JOIN T2
 ON T1.id_b = T2.id_b
 WHERE CONCAT(c1, ds) = 'F9.00G')