@mysqlクエリ最大連続日数sql

SELECT uid, min(date) AS mindate, max(date) AS maxdate, max(countday) countday, group_concat(date) dates FROM( SELECT uid, date ,@countday := ( CASE WHEN ( DATEDIFF( date ,@last_date) = 1) THEN (@countday + 1) ELSE 1 END) AS countday, @last_date := date FROM(SELECT uid , date FROM test_demo ORDER BY uid,date) AS t1, (select @countday:=0,@last_date:=’’) as t2 ) AS t3 GROUP BY uid
SELECT DATEDIFF(‘2015-09-02’,‘2015-09-01’)
分解sql SELECT uid,date FROM test_demo ORDER BY uid,dateソート検出id,日付
@countday @last_dateこれは変数宣言です
:=割り当て
@countday:=(CASE WHEN(DATEDIFF(date,@last_date)=1)THEN(@countday+1)ELSE 1 END DATEDIFF(d 1,d 2)2つの時間の時間差を計算2つの時間差が1の場合、連続日数、@countday+1変数に1を加算します.
@last_date:=date dataを変数@last_に割り当てるdate
したがって、比較するたびに、現在の行dataと前の行dataが比較されます.
(select@countday:=0,@last_date:=’)as付与変数
試験データCREATE TABLE test_demo(id int(11)NOT NULL AUTO_INCREMENT, uid int(11) NOT NULL DEFAULT ‘0’, date date NOT NULL, PRIMARY KEY ( id ), UNIQUE KEY idx_ud ( uid , date ) ) ENGINE=InnoDB AUTO_INCREMENT=23 DEFAULT CHARSET=latin1
INSERT INTO test_demo ( id , uid , date ) VALUES
(1, 1, ‘20111-011-01’),
(2, 1, ‘20111-011-02’),
(17, 1, ‘20111-011-03’),
(18, 1, ‘20111-011-04’),
(3, 2, ‘20111-011-01’),
(40, 2, ‘20111-011-02’),
(118, 2, ‘20111-011-03’),
(20, 2, ‘20111-011-04’),
(11, 3, ‘20111-011-01’),
(6, 3, ‘20111-011-02’),
(21, 3, ‘20111-011-03’),
(22, 3, ‘20111-011-04’),
(75, 4, ‘20111-011-01’),
(87, 4, ‘20111-011-02’),
(101, 11, ‘20111-011-01’),
(10, 11, ‘20111-011-02’),
(11, 6, ‘20111-011-01’),
(12, 6, ‘20111-011-02’),
(13, 7, ‘20111-011-01’),
(94, 7, ‘20111-011-02’),
(18, 8, ‘20111-011-01’),
(156, 8, ‘20111-011-02’);