postgresSQL 在update语句中使用别名 和 关联表（子查询）

update apt_taxiway m 
set code = n.code, name = n.name
from (
select name as code, name, uuid, version, airport_code from am_taxiway_area where name like '%-%' and version = '2020_14'
union all
select a.name as code, a.name, a.uuid, a.version, a.airport_code 
from am_taxiway_area a right join (
select name as code from am_taxiway_area where name not like '%-%' and version = '2020_14' group by name having count(name) = 1 
) b on a.name = b.code
where a.version = '2020_14'
union all
select a.name as code, a.name || '_' || a.order1 as name, a.uuid, a.version, a.airport_code 
from am_taxiway_area a right join (
select name as code from am_taxiway_area where name not like '%-%' and version = '2020_14' group by name having count(name) > 1 
) b on a.name = b.code
where a.version = '2020_14') n
where m.uuid = n.uuid and m.version = '2020_14';

同mysql和oracle不同，pgsql的update语法中关联表不使用join ，而是使用from，关联条件不用on，而是写在where中，和条件写在一起。

使用别名更新时，被更新表的字段不能用别名，如m.code = n.code是不行的。