如何获取所有出现的分层数据？

Question

在我的 PostgreSQL 数据库 (version: 11.4) 中，我有 table 调用 organizations。这个 table 有这样的结构：

| organization_id | organization_name | parent_organization_id | tree_organization_id | organization_rang |
|-----------------|-------------------|------------------------|----------------------|-------------------|
| 1               | Alphabet          |                        |                    | 1                 |
| 2               | Google            | 1                      |                  | 2                 |
| 3               | Calico            | 1                      |                  | 2                 |
| 4               | Youtube           | 2                      |                | 3                 |
| 5               | Nest              | 2                      |                | 3                 |
| 6               | Facebook          |                        |                    | 1                 |
| 7               | Whatsapp          | 5                      |                  | 2                 |
| 8               | Instagram         | 5                      |                  | 2                 |

如您所见，此 table 存储有关组织之间层次关系的信息。假设作为输入，我有特定的 organization_id 值。例如，它可以是 4 (Youtube)。我需要创建一个名为 dependencies 的新列，其中将存储所有具有组织 ID 的 tree_organization_id。这个新列的数据类型必须是 string array []text.

换句话说，我正在尝试获得类似的结果：

| organization_id | organization_name | parent_organization_id | tree_organization_id | organization_rang | dependencies              |
|-----------------|-------------------|------------------------|----------------------|-------------------|---------------------------|
| 1               | Alphabet          |                        |                    | 1                 | ['', '', '' ] |
| 2               | Google            | 1                      |                  | 2                 | ['', '']        |
| 4               | Youtube           | 2                      |                | 3                 | ['']                |

我使用了这样的 SQL 请求来找到组织的父级，但是如何正确计算 dependencies 列中的值？

with recursive hierarchy as (
    select
        organizations.organization_id,
        organizations.organization_name,
        organizations.parent_organization_id,
        organizations.tree_organization_id,
        organizations.organization_rang
    from
        organizations
    and
        organizations.organization_id in (4)
    union
    select
        a.organization_id,
        a.organization_name,
        a.organization_rang,
        a.parent_organization_id,
        a.tree_organization_id
    from
        organizations a
    inner join
        hierarchy b
    on
        a.organization_id = b.parent_organization_id
)
select
    hierarchy.organization_id,
    hierarchy.organization_name,
    hierarchy.parent_organization_id,
    hierarchy.tree_organization_id,
    hierarchy.organization_rang
from
    hierarchy
order by
    hierarchy.organization_rang;

Answer 1

如果我没理解错的话，你不需要递归 CTE。您可以只匹配 tree 列：

select o.*,
       (select array_agg(o2.tree order by o2.rang)
        from organizations o2
        where o2.tree like (o.tree || '%')
       ) as dependencies
from organizations o;

Here 是一个 db<>fiddle.

请注意，我将反斜杠更改为正斜杠。反斜杠是字符串中的转义字符，因此使用起来很挑剔。

如何获取所有出现的分层数据？

How do I get all occurrences of hierarchical data?

sql

postgresql

hierarchy