Zero foundation self-study SQL course | window function

Hello everyone , I'm Ning Yi .

Today is our first 24 course ： Window function .

Window function , Also called OLAP(Online Anallytical Processing, On line analytical processing ）, The database data can be analyzed and processed in real time .

Window function is a syntax function commonly used by data analysts , The interview is almost a must for a large factory .

This section belongs to the high-level content , It's hard to understand , We must knock on the computer by ourselves , Look at the effect .

Only in this way can we deepen our understanding . The previous course talked about how to install a database , You can check it on my homepage ~

Basic grammar

< Window function > OVER (
[PARTITION BY < Column name for grouping >] --  Optional 
[ORDER BY < Column name for sorting >] --  Optional 
)

above  < Window function >  The location of , You can put the following two functions ：

（1） Aggregate functions ： Such as sum.avg,count,max,min etc. ;

（2） Special window functions ： such as rank, dense_rank, row_number, I'll talk about it in detail .

We usually put window functions in select In the sentence .

1、 Aggregate functions

Window functions are also used for grouping and sorting , And aggregate functions +GROUP BY The effect is similar to . But the records generated by window functions will not be aggregated , Each row of data generates a record .

example ： stay Scores In the score sheet , find Sid by 7-10 Of the students , And calculate the total score of each student to show .

SELECT
  Sid,
  SUM(score) AS " Total score "
FROM Scores
WHERE Sid BETWEEN 7 AND 10
GROUP BY Sid;

But the problem is , While we know the total score of students , I also want to know the specific score of each subject , This cannot be done with the above statement .

Why? ？

because SELECT If the following field is an existing column in the table , be GROUP BY Clause must also have this column .

So if you want to know the specific scores of each subject , stay SELECT We will add Cid、score Two , stay GROUP BY If these two columns are also added to the clause , It would be contrary to our topic .

This knowledge point we are in the 15 Make peace 16 I've emphasized everything , You can click the home page to learn more , Let's not talk about it now .

We want our colleagues to know the total score of students 、 Course number 、 fraction , You need to use the window function .

SELECT
  Sid,Cid,score,
  SUM(score) OVER (PARTITION BY Sid) AS " Total score "
FROM Scores
WHERE Sid BETWEEN 7 AND 10;

Supplementary knowledge ：PARTITION BY The meaning of

PARTITION BY Partition means partition , Follow GROUP BY The meaning of grouping is similar .

If you don't write PARTITION BY, It means that the entire data set belongs to a partition .

For example, above us SQL sentence , If you omit PARTITION BY

SELECT
  Sid,Cid,score,
  SUM(score) OVER() AS " Total score "
FROM Scores
WHERE Sid BETWEEN 7 AND 10;

It turns out that , Will calculate the total score of all students .

2、 Special window functions

Common special window functions are ：

（1） Get data ranking ：

ROW_NUMBER()： Regardless of the parallel ranking , For example, the top three scores are 88,88,77, The ranking is 1,2,3.

RANK()： If there is a row with parallel ranking , Will occupy the next place . For example, the top three scores are 88,88,77, The ranking is 1,1,3.

DEBSE_RANK()： If the row is tied , Don't occupy the next place . For example, the top three scores are 88,88,77, The ranking is 1,1,2.

example ： stay Scores In the score sheet , find Sid by 7-10 Of the students , And calculate the ranking from high to low .

SELECT *,
  ROW_NUMBER() OVER(
    ORDER BY score DESC
  ) AS " ranking "
FROM Scores
WHERE Sid BETWEEN 7 AND 10;

Used above ROW_NUMBER() Window function , You can see the increase “ ranking ” This column . If you will ROW_NUMBER() Replace with RANK(), The return result is as follows ：

（2） Get first or last ：

FIRST_VALUE(< Name >)： Get first place .

LAST_VALUE(< Name >)： Get the last place .

example ： stay Scores In the score sheet , find Sid by 7-10 Of the students , Get the highest score of each student .

SELECT *,
  FIRST_VALUE(score) OVER(
    PARTITION BY Sid
    ORDER BY score DESC
  ) AS " Top grade "
FROM Scores
WHERE Sid BETWEEN 7 AND 10;

（3） Offset function ：

LEAD(< Name >,< The number n>)： Access the downward offset from the current line n Row data .

LAG(< Name >,< The number n>)： Offset from current row access n Row data .

NTH_VALUE(< Name >,< The number n>)： From the... In the result set N Row get data .

（4） Distribution function ：

CUME_DIST()： Less than... In the group 、 Equal to current rank The number of rows for the value / The total number of rows in the group .

PERCENT_RANK()： Returns the percentage sort of each row of a column , Each line follows the formula (rank-1) / (rows-1) Calculate .

NTILE(< The number n>)： Divide the result set into n Group , And show which group a piece of data is assigned to .

Homework ： stay Scores Verify the offset function and distribution function in , See what results can be returned .

such as ： Get the... Below 2 Line offset fraction .

SELECT *,
  LEAD(score,2) OVER(
    ORDER BY score DESC
  ) AS " Get the... Below 2 That's ok score value "
FROM Scores
WHERE Sid BETWEEN 7 AND 10;

We SQL The introductory to advanced courses are over . I will continue to go out later SQL Advanced courses , Include views 、 Indexes 、 Concurrent 、 Deadlock 、 trigger 、 event 、 Business 、 Stored procedure, etc .

Recording and broadcasting classes and live broadcasting classes will also keep up ~