如何创建表，其中第一个表中的列是另一个表中的字段？

Question

TABLE_1

  t1_id  name
 -------------
  1      mark
  2     stieve

TABLE_2

t2_id   month  amt   t1_id
----------------------------
 1      jan    200     1
 2      feb    400     1
 3      jan    500     2

Expected output

VIEW_TABLE_3

 T1_ID   name    jan    feb
 -----------------------
   1      mark   200   400
   2     stieve  500    0

In the VIEW_TABLE_3 we need to make columns Jan and Feb which are fields in the Table_2 , with their respective amt.

Please help how can get this output?

Answer 1

I don't know which one RDBMS you are using, but In SQL-Server you can use PIVOT in following:

CREATE TABLE #t1
(
    t1_id INT,
    name NVARCHAR(60)
)
CREATE TABLE #t2
(
    t2_id INT,
    [month] NVARCHAR(20),
    amt INT,
    t1_id INT
)
INSERT INTO #t1 VALUES (1, 'Mark'), (2, 'Stieve')
INSERT INTO #t2 VALUES 
(1,      'jan',    200,         1),
(2,      'feb',    400,         1),
(3,      'jan',    500,         2)

SELECT t1_id, Name, COALESCE(jan, 0) jan, COALESCE(feb, 0) feb
FROM (
    SELECT #t1.t1_id, #t1.Name, #t2.[month], #t2.amt
    FROM #t1
    JOIN #t2 ON #t1.t1_id = #t2.t1_id
)x
PIVOT
(
MAX([amt])
FOR [month] IN ([jan], [feb])
)piv

DROP TABLE #t1
DROP TABLE #t2

OUTPUT:

t1_id   Name    jan     feb
  1     Mark    200     400
  2    Stieve   500      0

Answer 2

In PostgreSQL you can install the tablefunc extension:

CREATE EXTENSION tablefunc;

In that extension you will find the crosstab(text, text) function . This is an ugly function to work with (putting it mildly), but it will do what you want:

SELECT * FROM crosstab(
    'SELECT t1_id, name, month, coalesce(amount, 0) FROM table_1 JOIN table_2 USING (t1_id)',
    'VALUES (''jan''), (''feb''), (''mar''), (''apr''), (''may''), (''jun''), (''jul''), (''aug''), (''sep''), (''oct''), (''nov''), (''dec'')'
  ) AS (
    id int, name varchar,
    jan int, feb int, mar int, apr int, may int, jun int,
    jul int, aug int, sep int, oct int, nov int, dec int
);

The first parameter specifies the data to be cross-tabulated, in a specific format (check the documentation), the second parameter specifies the categories. I am assuming here that the months could be any of the year.

If you can have multiple records for the same person in a single month, then you can sum coalesce(amount, 0) in the first parameter, and add a GROUP BY 1, 2, 3 clause.

Finally, you can wrap this in a view:

CREATE VIEW view_table_3 AS
  SELECT * FROM crosstab(...