SQL函数

1，字符串截取拼接

CONCAT(LEFT(c.id_card,LENGTH(c.id_card)-4),'****');
SUBSTRING_INDEX(c.context,'}',1);
SUBSTRING_INDEX(a.task_context,':',-1) as context;
c.mobile LIKE CONCAT('%', '${mobile}', '%')

2，Case函数

(
        CASE a.type_code
        WHEN 'bp' THEN
            '血压'
        WHEN 'hr' THEN
            '心率'
        WHEN 'fbg' THEN
            '空腹血糖'
        WHEN '2hpbg' THEN
            '餐后两小时血糖'
        WHEN 'tc' THEN
            '总胆固醇'
        WHEN 'bua' THEN
            '血尿酸'
        WHEN 'bmi' THEN
            '体质指数'
        ELSE
            '其他'
        END
    ) AS type_code

3，字符串拼接

CONCAT_WS('/',b.value1,b.value2,b.value3);
CONCAT(a.value1,'mmol/L') AS value1

4，日期加减

DATEDIFF(DATE(MAX(take_time)),DATE(MIN(take_time))) as days

5，年龄计算

YEAR (NOW()) - YEAR (b.birthday) AS birthday

6，IF函数

IF(b.gender=0,'女','男') AS gender

7，COUNT函数

count(DISTINCT(a.called_user_id) & a.begin_calltime>0)

SELECT
    a.user_id,
  a.realname,
  (SELECT COUNT(*) from ut_pic b where b.user_id=a.user_id) as picTotal,
  (SELECT COUNT(*) from ut_pic c where c.user_id=a.user_id and is_show=1) as pic,
  (SELECT COUNT(*) from ut_video d where d.user_id=a.user_id) as videoTotal,
  (SELECT COUNT(*) from ut_video e where e.user_id=a.user_id and is_show=1) as video
FROM
    ut_user a

with ta as
(
   select nvl(v.update_time,v.create_time) sj,v.housekeeper_id hid from t_crd_video v
   union all
   select nvl(a.update_time,a.create_time) sj,a.housekeeper_id hid from t_crd_album a
)
select
h.id ID,h.real_name realName,
(select count(1) from t_crd_album t where t.vaild=1 and t.housekeeper_id=h.id and t.is_show=1) picShow,
(select count(1) from t_crd_album t where t.vaild=1 and t.housekeeper_id=h.id) picTotal,
(select count(1) from t_crd_video v where v.vaild=1 and v.housekeeper_id=h.id and v.is_show=1 ) vidShow,
(select count(1) from t_crd_video v where v.vaild=1 and v.housekeeper_id=h.id ) vidTotal,
to_char((select max(sj) from ta where ta.hid=h.id),'yyyy-mm-dd ') updateTime
from t_crd_housekeeper h
where h.vaild=1 

8，外联

SELECT
    f.id,
    f.store_name,
    f.vendor_name,
    f.area_name,
    f.store_phone,
    h.realname,
    h.mobile,
    f.created_time
FROM
    (
        SELECT
            a.id,
            a.store_name,
            c.vendor_name,
            b.area_name,
            a.store_phone,
            a.created_time
        FROM
            pd_store a,
            pd_vendor_area b,
            pd_vendor c
        WHERE
            a.vendor_area_id = b.id
        AND a.vendor_id = c.id
        AND a.vendor_id = 1
        AND b.area_type = 1
        AND b.area_name LIKE '%%'
        LIMIT 0,5
    ) f
LEFT OUTER JOIN pd_clerk g ON f.id = g.store_id
AND g.user_role = 2
LEFT OUTER JOIN pd_user h ON g.user_id = h.uid

9, 外联

SELECT
    k.operater_id,
    k.realname,
    k.mobile,
    k.pdNum,
    l.operater_id,
    l.realname,
    l.mobile,
    l.dslNum
FROM
    (
        SELECT
            a.operater_id,
            b.realname,
            b.mobile,
            COUNT(DISTINCT a.user_id) AS pdNum
        FROM
            pd_indicator_values a,
            pd_user b,
            pd_user_vendor c
        WHERE
            a.operater_id = b.uid
        AND a.vendor_id = 12
        AND a.type_code <> 'hr'
        AND a.user_id = c.user_id
        AND a.take_time >= '2016-10-26 00:00:00'
        AND a.take_time <= '2016-11-30 23:59:59'
        AND c.vendor_id = 12
        AND c.vendor_member_id LIKE '202%'
        GROUP BY
            a.operater_id
    ) k
LEFT OUTER JOIN (
    SELECT
        a.operater_id,
        b.realname,
        b.mobile,
        COUNT(DISTINCT a.user_id) AS dslNum
    FROM
        pd_indicator_values a,
        pd_user b,
        pd_user_vendor c
    WHERE
        a.operater_id = b.uid
    AND a.vendor_id = 12
    AND a.type_code <> 'hr'
    AND a.user_id = c.user_id
    AND a.take_time >= '2016-10-26 00:00:00'
    AND a.take_time <= '2016-11-30 23:59:59'
    AND c.vendor_id = 12
    AND c.vendor_member_id NOT LIKE '202%'
    GROUP BY
        a.operater_id
) l ON k.operater_id = l.operater_id
UNION
    SELECT
        k.operater_id,
        k.realname,
        k.mobile,
        k.pdNum,
        l.operater_id,
        l.realname,
        l.mobile,
        l.dslNum
    FROM
        (
            SELECT
                a.operater_id,
                b.realname,
                b.mobile,
                COUNT(DISTINCT a.user_id) AS pdNum
            FROM
                pd_indicator_values a,
                pd_user b,
                pd_user_vendor c
            WHERE
                a.operater_id = b.uid
            AND a.vendor_id = 12
            AND a.type_code <> 'hr'
            AND a.user_id = c.user_id
            AND a.take_time >= '2016-10-26 00:00:00'
            AND a.take_time <= '2016-11-30 23:59:59'
            AND c.vendor_id = 12
            AND c.vendor_member_id LIKE '202%'
            GROUP BY
                a.operater_id
        ) k
    RIGHT OUTER JOIN (
        SELECT
            a.operater_id,
            b.realname,
            b.mobile,
            COUNT(DISTINCT a.user_id) AS dslNum
        FROM
            pd_indicator_values a,
            pd_user b,
            pd_user_vendor c
        WHERE
            a.operater_id = b.uid
        AND a.vendor_id = 12
        AND a.type_code <> 'hr'
        AND a.user_id = c.user_id
        AND a.take_time >= '2016-10-26 00:00:00'
        AND a.take_time <= '2016-11-30 23:59:59'
        AND c.vendor_id = 12
        AND c.vendor_member_id NOT LIKE '202%'
        GROUP BY
            a.operater_id
    ) l ON k.operater_id = l.operater_id;

N，其他

SELECT
  b.store_name,
  a.take_time,
  c.realname,
  CONCAT(LEFT(c.id_card,LENGTH(c.id_card)-4),'****'),
  c.mobile,
  (
        CASE a.type_code
        WHEN 'bp' THEN
            '血压'
        WHEN 'hr' THEN
            '心率'
        WHEN 'fbg' THEN
            '空腹血糖'
        WHEN '2hpbg' THEN
            '餐后两小时血糖'
        WHEN 'tc' THEN
            '总胆固醇'
        WHEN 'bua' THEN
            '血尿酸'
        WHEN 'bmi' THEN
            '体质指数'
        ELSE
            '其他'
        END
    ) AS type_code,
  a.value1,
  a.value2,
  a.value3,
  d.realname as clerkname,
  d.mobile as clerkmobile
FROM
    pd_indicator_values a,
    pd_store b,
    pd_user c,
    pd_user d
WHERE
    a.store_id = b.id
AND a.user_id = c.uid
AND a.operater_id = d.uid
AND a.store_id = 164
AND a.take_time >= '2016-09-01 00:00:00'
AND a.take_time < '2016-11-01 00:00:00';

SELECT
    a.realname,
  a.gender,
  c.realname as clerkname,
  c.mobile,
  d.store_name,
  b.type_code,
  b.take_time,
  CONCAT_WS('/',b.value1,b.value2,b.value3)
FROM
    pd_user a,
    pd_indicator_values b,
    pd_user c,
    pd_store d
WHERE
    a.uid = b.user_id
AND b.operater_id = c.uid
AND b.store_id = d.id ORDER BY b.user_id ASC;

SELECT
    COUNT(*)
FROM
    (
        SELECT
            user_id,
            COUNT(user_id) AS coun,
            DATEDIFF(
                DATE(MAX(take_time)),
                DATE(MIN(take_time))
            ) AS days
        FROM
            pd_indicator_values
        WHERE
            take_time >= '2015-11-01 00:00:00'
        AND take_time <= '2016-11-24 23:59:59'
        AND vendor_id = 1
        GROUP BY
            user_id
        HAVING
            coun > 1
        AND days > 1
    ) k;

SELECT
    h.vendor_name,
    h.take_time,
    h.realname,
    h.created_time,
    h.gender,
    h.birthday,
    h.address,
    h.mobile,
    h.id_card,
    h.type_code,
    h.value1_status,
    h.value1,
    h.value2_status,
    h.value2,
    h.value3_status,
    h.value3,
    h.clerkname,
    h.clerkmobile,
    h.store_name,
    j.realname as stoname,
    j.mobile as stomobile
FROM
    (
        SELECT
            e.vendor_name,
            a.take_time,
            f.created_time,
            b.realname,
            (
                CASE b.gender
                WHEN '' THEN
                    '女'
                WHEN '' THEN
                    '男'
                ELSE
                    ' '
                END
            ) AS gender,
            YEAR (NOW()) - YEAR (b.birthday) AS birthday,
            g.address,
            b.mobile,
            b.id_card,
            (
                CASE a.type_code
                WHEN 'bp' THEN
                    '血压'
                WHEN 'hr' THEN
                    '心率'
                WHEN 'fbg' THEN
                    '空腹血糖'
                WHEN '2hpbg' THEN
                    '餐后两小时血糖'
                WHEN 'tc' THEN
                    '总胆固醇'
                WHEN 'bua' THEN
                    '血尿酸'
                WHEN 'bmi' THEN
                    '体质指数'
                ELSE
                    '其他'
                END
            ) AS type_code,
            (
                CASE a.value1_status
                WHEN '' THEN
                    '正常'
                WHEN '' THEN
                    '风险'
                WHEN '' THEN
                    '危险'
                WHEN '' THEN
                    '偏小风险'
                WHEN '' THEN
                    '偏小危险'
                ELSE
                    '其他'
                END
            ) AS value1_status,
            a.value1,
            (
                CASE a.value2_status
                WHEN '' THEN
                    '正常'
                WHEN '' THEN
                    '风险'
                WHEN '' THEN
                    '危险'
                WHEN '' THEN
                    '偏小风险'
                WHEN '' THEN
                    '偏小危险'
                ELSE
                    '其他'
                END
            ) AS value2_status,
            a.value2,
            (
                CASE a.value3_status
                WHEN '' THEN
                    '正常'
                WHEN '' THEN
                    '风险'
                WHEN '' THEN
                    '危险'
                WHEN '' THEN
                    '偏小风险'
                WHEN '' THEN
                    '偏小危险'
                ELSE
                    '其他'
                END
            ) AS value3_status,
            a.value3,
            c.realname AS clerkname,
            c.mobile AS clerkmobile,
            d.store_name,
            d.id,
      q.cou
        FROM
            (
                SELECT
                    o.user_id,
                    COUNT(o.user_id) AS cou
                FROM
                    pd_indicator_values o
                WHERE
                    o.take_time >= '2015-06-01 00:00:00'
                AND o.take_time <= '2016-11-22 23:59:59'
                GROUP BY
                    o.user_id
                HAVING
                    cou > 2
            ) q,
            pd_indicator_values a,
            pd_user b,
            pd_user c,
            pd_store d,
            pd_vendor e,
            pd_user_vendor f,
            pd_user_info g
        WHERE
            q.user_id = a.user_id
        AND a.user_id = b.uid
        AND a.operater_id = c.uid
        AND a.store_id = d.id
        AND a.vendor_id = e.id
        AND a.user_id = f.user_id
        AND a.user_id = g.user_id
        AND a.take_time >= '2015-06-01 00:00:00'
        AND a.take_time <= '2016-11-22 23:59:59'
    ) h,
    pd_clerk i,
    pd_user j
WHERE
    h.id = i.store_id
AND i.user_role = 2
AND i.user_id = j.uid ORDER BY h.cou DESC;

1,将查询结果存入表
insert into pd_temp select * from pd_other;

2,将表数据导出xls文件最大1048576记录数
select * from pd_temp limit 1048576,364387 into outfile 'c:\\20.xls';
3,格式转换
记事本另存为ASCII格式
或
iconv -futf8 -tgb2312 -otest 21.xls 20.xls

SELECT
    a.take_time AS take_time,
    b.realname AS realname,
    IF(b.gender=0,'女','男') AS gender,
    b.mobile AS mobile,
    b.id_card AS id_card,
    IF(a.type_code='fbg', '空腹血糖', '餐后两小时血糖') AS type_code,
    CONCAT(a.value1,'mmol/L') AS value1,
    d.realname as clerkname,
  d.mobile as clerkmobile,
    c.store_name AS store_name
FROM
    pd_indicator_values a,
    pd_user b,
    pd_store c,
    pd_user d
WHERE
    a.user_id = b.uid
AND a.operater_id = d.uid
AND a.store_id = c.id
AND a.vendor_id = 1
AND a.province_code = 520000
AND a.city_code = 520100
AND a.take_time >= '2016-08-01 00:00:00'
AND a.take_time < '2016-09-01 00:00:00'
AND (
    a.type_code = 'fbg'
    OR a.type_code = '2hpbg'
)
AND (
    a.value1_status = 3
    OR a.value1_status = 5
);

SELECT
    d.id,
    d.store_name,
    c.realname,
    c.mobile,
    count(a.caller_user_id),
    count(
        DISTINCT (a.called_user_id) & a.begin_calltime > 0
    ),
    count(a.begin_calltime > 0),
    SUM(a.caller_duration)
FROM
    pd_call_records a,
    pd_clerk b,
    pd_user c,
    pd_store d
WHERE
    a.caller_user_id = b.user_id
AND b.user_id = c.uid
AND b.store_id = d.id
AND a.created_time >= '2016-08-01 18:48:45'
GROUP BY
    a.caller_user_id;

SELECT
    c.realname,
    c.mobile,
    SUBSTRING_INDEX(c.context, '}', 1),
    c.callednum,
    c.callnum,
    c.calltime,
    d.realname,
    f.store_name,
    d.mobile
FROM
    (
        SELECT
            b.realname AS realname,
            b.mobile AS mobile,
            SUBSTRING_INDEX(a.task_context, ':' ,- 1) AS context,

        IF (a.caller_duration > 0, 1, 0) AS callednum,

    IF (a.called_duration > 0, 1, 0) AS callnum,
    IFNULL(a.called_duration, 0) AS calltime,
    a.caller_user_id AS clerkid
FROM
    pd_call_records a,
    pd_user b
WHERE
    a.called_user_id = b.uid
AND task_id > 0
    ) c,
    pd_user d,
    pd_clerk e,
    pd_store f
WHERE
    c.clerkid = d.uid
AND d.uid = e.user_id
AND e.store_id = f.id;

SELECT COUNT(k.user_id) from (
SELECT
  a.user_id,
    DATEDIFF(DATE(MAX(a.take_time)),DATE(MIN(a.take_time))) as days
FROM
    pd_indicator_values a
WHERE
    a.vendor_id = 12
AND a.take_time >= '2016-10-26 00:00:00'
AND a.take_time <= '2016-11-30 23:59:59'
GROUP BY a.user_id

)k where k.days>1

邮箱联想匹配：

SELECT
    a.ID,
    b.USER_ID,
    a.PERSON_ID,
    a.TYPE,
    a.MAIL,
    a.CREATE_TIME,
    a.UPDATE_TIME,
    a.DELETE_TAG,
    a.MAIN_USER_ID,
  b.`NAME`,
  b.AVATAR,
  b.LEVEL1_GROUP_ID,
  b.LEVEL2_GROUP_ID,
  b.LEVEL3_GROUP_ID
FROM
    t_contacts_person_mail a, t_contacts_person b
WHERE
    a.PERSON_ID = b.ID
AND    a.DELETE_TAG = 0 AND b.DELETE_TAG = 0 AND locate('@',a.MAIL)>0
-- 子账号
AND (b.USER_ID ='9e5687aa76b74daaae47bdbf9f453e97' OR (b.USER_ID='83fcb7323c9a47de98403be7cedb9433' AND b.IS_OPEN = 1))
-- 主账号
-- AND b.USER_ID in ('83fcb7323c9a47de98403be7cedb9433', '9e5687aa76b74daaae47bdbf9f453e97')
AND substring_index(a.MAIL, '@', 1) LIKE CONCAT('%', 'a', '%')
GROUP BY a.MAIL
ORDER BY (length(substring_index(a.MAIL, '@', 1)) - length('a')) ASC, b.IS_OPEN ASC, b.CREATE_TIME DESC
LIMIT 0,10