使用 SQL 查询获取父级的所有子级 - 分层数据

发布于 2025-01-13 08:20:36 字数 703 浏览 1 评论 0原文

我正在尝试学习如何查询分层数据以返回父级的所有子级（给出了父级电子邮件）。这是我的查询：

WITH emp AS (
  SELECT
    PersonnelNumber,
    Email,
    ReportsToPersonnelNbr,
    1 AS level
  FROM tbl
  WHERE Email = 'XYZ'
 
  UNION ALL
   
  SELECT
    e.PersonnelNumber,
    e.Email,
    e.ReportsToPersonnelNbr,
    emp.level + 1
  FROM tbl e INNER JOIN emp
  ON e.ReportsToPersonnelNbr = emp.PersonnelNumber
)
SELECT e.PersonnelNumber, e.Email, e.ReportsToPersonnelNbr, e.level
FROM emp e
ORDER BY e.level;

我得到了预期的结果，但是返回结果需要更长的时间。例如，2 分钟可返回约 300 行。我该如何解决这个问题？

更新：我运行了以下 sql 查询：

CREATE INDEX [IDX_ReportsToPersonnelNbr] ON tbl(ReportsToPersonnelNbr)

但是我仍然发现返回结果需要更长的时间。

原文

I'm trying to learn how to query hierarchical data to return all children of a parent (parent email is given). Here is my query:

WITH emp AS (
  SELECT
    PersonnelNumber,
    Email,
    ReportsToPersonnelNbr,
    1 AS level
  FROM tbl
  WHERE Email = 'XYZ'
 
  UNION ALL
   
  SELECT
    e.PersonnelNumber,
    e.Email,
    e.ReportsToPersonnelNbr,
    emp.level + 1
  FROM tbl e INNER JOIN emp
  ON e.ReportsToPersonnelNbr = emp.PersonnelNumber
)
SELECT e.PersonnelNumber, e.Email, e.ReportsToPersonnelNbr, e.level
FROM emp e
ORDER BY e.level;

I'm getting the results as expected however it is taking a longer time to return the results. For example, 2 minutes to return ~300 rows. How do I fix that?

UPDATE: I ran the following sql query:

CREATE INDEX [IDX_ReportsToPersonnelNbr] ON tbl(ReportsToPersonnelNbr)

however I still see it is taking a longer time to return the results.

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

倒数 2025-01-20 08:20:36

您需要更好地对表进行索引。考虑服务器将如何找到您想要的行，并记住您的索引需要覆盖所有必要的列。

为了使查询性能良好，您将需要以下两个索引

CREATE NONCLUSTERED INDEX [IDX_ReportsToPersonnelNbr] ON tbl
  (Email) INCLUDE (PersonnelNumber, ReportsToPersonnelNbr)
  WITH (DROP_EXISTING = ON);

CREATE NONCLUSTERED INDEX [IDX2_ReportsToPersonnelNbr] ON tbl
  (ReportsToPersonnelNbr) INCLUDE (PersonnelNumber, Email)
  WITH (DROP_EXISTING = ON);

You need to index your table better. Think about how the server is going to find the rows that you want, and bear in mind that your indexes need to cover all necessary columns.

For your query to perform well, you are going to want the following two indexes

CREATE NONCLUSTERED INDEX [IDX_ReportsToPersonnelNbr] ON tbl
  (Email) INCLUDE (PersonnelNumber, ReportsToPersonnelNbr)
  WITH (DROP_EXISTING = ON);

CREATE NONCLUSTERED INDEX [IDX2_ReportsToPersonnelNbr] ON tbl
  (ReportsToPersonnelNbr) INCLUDE (PersonnelNumber, Email)
  WITH (DROP_EXISTING = ON);

回复收藏 0 原文

~没有更多了~