I have a MySQL table where rows are inserted in a given order by the user. If the user then forgets to add a row and inserts it later it will come after the others that are already inserted, also when performing simple query. The order of insertion is the order I want the elements to be retreived. The table in question is a simple table resolving a many-to-many relationship with two IDs (recipe_id and ingredient_id).
mysql> select * from ingredient_in_recipe where recipe_id = 7;
+-----------+---------------+------+----------+------------+
| recipe_id | ingredient_id | unit | quantity | group_name |
+-----------+---------------+------+----------+------------+
| 7 | 71 | g | 300.00 | NULL |
| 7 | 34 | stk | 3.00 | NULL |
| 7 | 72 | stk | 1.00 | NULL |
| 7 | 73 | stk | 0.50 | NULL |
| 7 | 45 | stk | 6.00 | NULL |
| 7 | 74 | stk | 0.50 | NULL |
| 7 | 23 | g | 15.00 | NULL |
| 7 | 78 | ts | 2.00 | NULL |
| 7 | 75 | ts | 3.00 | NULL | <--- This is where I want the last element to be.
| 7 | 76 | ss | 1.00 | NULL |
| 7 | 77 | stk | 1.00 | NULL |
| 7 | 79 | g | 195.00 | NULL |
| 7 | 38 | ss | 5.00 | NULL | <--- This is inserted later. Should be "higher up".
+-----------+---------------+------+----------+------------+
13 rows in set (0.00 sec)
Is there a simple way to achieve this?
As pointed out by #C4ud3x and #Hanno Binder, there is no guarantee that MySQL returns the data in the same order as they were inserted. Normally, this is the case, but it is not a robust way to handle ordering of the data. Thus I solved the problem by adding a column order_id to be able to use the ORDER BY clause to ensure that the order will be maintained properly.
I found a good answer to a related question over at dba.stackexchange.com. The main point from the accepted answer there is:
If, on the other hand, you intend to rely on this order for anything, you must specify your desired order using ORDER BY. To do anything else is to set yourself up for unwelcome surprises.
Related
I read that the output of the subquery doesn't matter and only its existence matter. But, when I change the code in the subquery, why is my output changing?
These are the tables:
mysql> select * from boats;
+------+-----------+-------+
| bid | bname | color |
+------+-----------+-------+
| 101 | Interlake | blue |
| 102 | Interlake | red |
| 103 | Clipper | green |
| 104 | Marine | red |
+------+-----------+-------+
mysql> select * from sailors;
+------+---------+--------+------+
| sid | sname | rating | age |
+------+---------+--------+------+
| 22 | Dustin | 7 | 45 |
| 29 | Brutus | 1 | 33 |
| 31 | Lubber | 8 | 55.5 |
| 32 | Andy | 8 | 25.5 |
| 58 | Rusty | 10 | 35 |
| 64 | Horatio | 7 | 35 |
| 71 | Zorba | 10 | 16 |
| 74 | Horatio | 9 | 40 |
| 85 | Art | 3 | 25.5 |
| 95 | Bob | 3 | 63.5 |
+------+---------+--------+------+
10 rows in set (0.00 sec)
mysql> select * from reserves;
+------+------+------------+
| sid | bid | day |
+------+------+------------+
| 22 | 101 | 1998-10-10 |
| 22 | 102 | 1998-10-10 |
| 22 | 103 | 1998-10-08 |
| 22 | 104 | 1998-10-08 |
| 31 | 102 | 1998-11-10 |
| 31 | 103 | 1998-11-06 |
| 31 | 104 | 1998-11-12 |
| 64 | 101 | 1998-09-05 |
| 64 | 102 | 1998-09-08 |
| 74 | 103 | 1998-09-08 |
+------+------+------------+
select sname from sailors s where exists(select * from reserves r where r.bid=103);
+---------+
| sname |
+---------+
| Dustin |
| Brutus |
| Lubber |
| Andy |
| Rusty |
| Horatio |
| Zorba |
| Horatio |
| Art |
| Bob |
+---------+
10 rows in set (0.00 sec)
mysql> select sname from sailors s where exists(select * from reserves r where r.bid=103 and r.sid=s.sid);
+---------+
| sname |
+---------+
| Dustin |
| Lubber |
| Horatio |
+---------+
Also, I am not able to understand what r.sid=s.sid is doing here. All the sid in reserves are already from sailors table. Please someone explain it to me.
The EXISTS is a Boolean Operator which indicates that if there is ANY row in the sub-query you passed to it. When you execute this:
EXISTS(SELECT * FROM reserves r WHERE r.bid=103)
It will return TRUE after finding the FIRST row which has the condition bid = 103 in Reserves table. The first part of the query doesn't matter, it does not matter what you SELECT in Exists and MySQL engine will ignore it, just the WHERE clause is the part which makes the difference, you can use Exists even like this:
EXISTS(SELECT 1 FROM reserves r WHERE r.bid=103)
In the query above, nothing depends on the values in main query, nothing depends on Sailors table, and if there is ANY row in the Reserves table with bid = 103, then it always will return TRUE.
In the second sub-query with EXISTS, you have a different WHERE clause, and it depend on the value of the fields of the main Query, so it will have different result per each row:
EXISTS(SELECT * FROM reserves r WHERE r.bid=103 AND r.sid=s.sid)
In the above query, per each row in Sailors table, MySQL uses sid value to produce the WHERE condition of the sub-query in EXISTS operator, so it will returns TRUE for a row in Sailors table if there are ANY rows in Reserves table which has a bid = 103 and sid = Sailors.sid, and it will returns False for those that has not such a record in Reserves table, and finally you will get a different result
I think I got that. Exists is used to check if the subquery is existing for the main query. I didn't give any link for the main query and subquery in the first query.
For every name in sailors, independently, the subquery is existing. Hence, I got all the names. In the second query, I added s.sid=r.sid which links the main query and subquery. It checks if for a sname, if bid=103, and also, if s.sid=r.sid.
Please comment if I got that right.
I have following tables with data as:
1.Table follow_up as :
mysql> select * from follow_up;
+--------------+----------------+--------------------------------------------------+-------------------+---------+---------------+-----------+---------------+----------+
| follow_up_id | feedback_close | feedback_open | is_email_required | is_Open | reminder_date | client_id | conclusion_id | stage_id |
+--------------+----------------+--------------------------------------------------+-------------------+---------+---------------+-----------+---------------+----------+
| 1 | NULL | dsffsdfsdfsd | 1 | 1 | 2017-09-20 | 101 | 96 | 72 |
| 2 | NULL | FSGDFHFGHFG | 1 | 1 | 2017-09-28 | 101 | 251 | 72 |
| 3 | NULL | Tender stage fb | 0 | 1 | NULL | 101 | 98 | 163 |
| 4 | NULL | Call back tender stage update date from 28 to 30 | 1 | 1 | 2017-09-28 | 101 | 96 | 163 |
| 5 | NULL | Metting follow up for next meeting | 1 | 1 | 2017-10-02 | 101 | 96 | 73 |
+--------------+----------------+--------------------------------------------------+-------------------+---------+---------------+-----------+---------------+----------+
2. Table logs as :
mysql> SELECT * from logs where transaction = 'FLWUP';
+---------+---------+---------------------+---------+-------------+
| user_id | menu_id | logs_time | tran_id | transaction |
+---------+---------+---------------------+---------+-------------+
| 84 | 69 | 2017-09-19 19:31:04 | 1 | FLWUP |
| 84 | 69 | 2017-09-19 19:31:25 | 2 | FLWUP |
| 84 | 69 | 2017-09-20 19:10:41 | 2 | FLWUP |
| 84 | 69 | 2017-09-21 12:35:01 | 3 | FLWUP |
| 84 | 69 | 2017-09-21 12:35:26 | 4 | FLWUP |
| 84 | 69 | 2017-09-21 12:36:16 | 4 | FLWUP |
| 84 | 69 | 2017-09-21 12:38:30 | 5 | FLWUP |
+---------+---------+---------------------+---------+-------------+
7 rows in set (0.00 sec)
3. table allcode as :
mysql> select * from allcode where code_type like 'MARK%';
+------------------+---------+------+----------------------+
| code_type | code_id | srno | code_name |
+------------------+---------+------+----------------------+
| MARKETING_STAGES | 72 | 1 | Enquiry |
| MARKETING_STAGES | 73 | 3 | Meeting |
| MARKETING_STAGES | 74 | 4 | Presentation |
| MARKETING_STAGES | 163 | 2 | Tender |
+------------------+---------+------+----------------------+
11 rows in set (0.00 sec)
I have invoked a query and got result as :
mysql> select f.follow_up_id,f.feedback_open, f.feedback_close, f.reminder_date,
ast.code_name as stage, ac.code_name as conclusion, max(l.logs_time)
from follow_up f
join logs l on l.tran_id = f.follow_up_id
join allcode ast on ast.code_id = f.stage_id
join allcode ac on ac.code_id = f.conclusion_id
where l.transaction='FLWUP' and f.client_id = 101
group by ast.code_name order by ast.srno;
+--------------+------------------------------------+----------------+---------------+---------+------------+---------------------+
| follow_up_id | feedback_open | feedback_close | reminder_date | stage | conclusion | max(l.logs_time) |
+--------------+------------------------------------+----------------+---------------+---------+------------+---------------------+
| 1 | dsffsdfsdfsd | NULL | 2017-09-20 | Enquiry | Call Back | 2017-09-20 19:10:41 |
| 3 | Tender stage fb | NULL | NULL | Tender | Next | 2017-09-21 12:36:16 |
| 5 | Metting follow up for next meeting | NULL | 2017-10-02 | Meeting | Call Back | 2017-09-21 12:38:30 |
+--------------+------------------------------------+----------------+---------------+---------+------------+---------------------+
3 rows in set (0.00 sec)
But I want result as :
+--------------+-----------------------------------------------------+----------------+---------------+---------+------------+---------------------+
| follow_up_id | feedback_open | feedback_close | reminder_date | stage | conclusion | max(l.logs_time) |
+--------------+-----------------------------------------------------+----------------+---------------+---------+------------+---------------------+
| 2 | FSGDFHFGHFG | NULL | 2017-09-20 | Enquiry | Call Back | 2017-09-20 19:10:41 |
| 4 | Call back tender stage update date from 28 to 30 | NULL | NULL | Tender | Next | 2017-09-21 12:36:16 |
| 5 | Metting follow up for next meeting | NULL | 2017-10-02 | Meeting | Call Back | 2017-09-21 12:38:30 |
+--------------+-----------------------------------------------------+----------------+---------------+---------+------------+---------------------+
3 rows in set (0.00 sec)
I'm not able to JOIN and group by to get required result.
column conclusion_id and stage_id of table follow_up are referring to code_id of table allcode.
Question :
the result I want is to be
group by stage_id,
order by srno of allcode and
last/recent follow_up_id of follow_up table
DEMO Includes my answer, original question with full group by needed, and Reupal's answer in demo. You were missing the values in your sample data for conclusionID so I just created them based on ID (now updated to ISO, Callback but missing 98.)
and my results don't match yours in this column; but I believe your expected results are in error.
Seems like you want the max follow_up_ID for each stage_ID when multiple stage_ID's exist
This can be handled by a derived table/inline view getting that max follow_UP_ID grouped by the stage_ID and a joining it back to your set. to limit results to include only the max follow_Up_ID by stage_Id.
I'm also not a fan of mySQL's extended group by and prefer including all columns not aggregated in the select in the group by. Using the extended group by tends to hide potential problems. In this case grouping by just the ast.code_name allowed the engine to select a non distinct value from the other columns. You ended up not getting the desired results and furthermore it hide the fact you would get multiple records in your query were it not for the extended group by use/misuse.
SELECT f.follow_up_id,f.feedback_open, f.feedback_close, f.reminder_date,
ast.code_name as stage, ac.code_name as conclusion, max(l.logs_time)
from follow_up f
join logs l on l.tran_id = f.follow_up_id
join allcode ast on ast.code_id = f.stage_id
join allcode ac on ac.code_id = f.conclusion_id
JOIN SELECT max(follow_up_ID) MFID, stage_ID
FROM follow_up
GROUP BY stage_ID) Z
on f.follow_up_ID = Z.MFID
and F.Stage_ID = Z.Stage_ID
WHERE l.transaction='FLWUP' and f.client_id = 101
GROUP BY f.follow_up_id,f.feedback_open, f.feedback_close, f.reminder_date,
ast.code_name , ac.code_name
ORDER BY ast.srno;
Try below, notice ordering and group by sequence.
select f.follow_up_id,f.feedback_open, f.feedback_close, f.reminder_date,
ast.code_name as stage, ac.code_name as conclusion, max(l.logs_time)
from follow_up f
join logs l on l.tran_id = f.follow_up_id
join allcode ast on ast.code_id = f.stage_id
join allcode ac on ac.code_id = f.conclusion_id
where l.transaction='FLWUP' and f.client_id = 101
group by follow_up.stage_id order by ast.srno, follow_up.follow_up_id DESC;
This should works, and if its not then you should search like how to set ordering on multiple column.
Ref. article- SQL multiple column ordering
I'm trying to break up a SQL table that needs to take a users name and find the unique user ID's from up to 4 systems.
The data is currently like this:
| Name | User_ID |
-----------------
| A | 10 |
| A | 110 |
| A | 1500 |
| A | 4 |
| B | 20 |
| B | 100 |
| B | 2 |
| C | 10 |
I need to pivot it around the user's name to look like this (the id's don't need to be in numerical order as the SYS#_ID for each doesn't matter):
| Name | SYS1_ID | SYS2_ID | SYS3_ID | SYS4_ID |
------------------------------------------------
| A | 4 | 10 | 110 | 1500 |
| B | 2 | 20 | 100 | NULL |
| C | 10 | NULL | NULL | NULL |
This is the code I have tried on MySQL:
PIVOT(
COUNT(User_ID)
FOR Name
IN (SYS1_ID, SYS2_ID, SYS3_ID, SYS4_ID)
)
AS PivotedUsers
ORDER BY PivotedUsers.User_Name;
I'm unsure if PIVOT works on MySQL as I keep getting an error "PIVOT unknown". Is there a way to find the values that each user has and if they do not appear in the table already add them to the next column with a max of 4 values?
I have a search section for looking up products which has a navigation bar for filtering purposes that shows the total results of each product feature. For example:
TOTAL RESULTS 60
New (32)
Used (28)
Particular (10)
Company (50)
In mysql I have the following queries (one per feature)
Type
SELECT a.id_type, whois.name as whoisName, COUNT(a.id_type) as countWhois
FROM (published a
INNER JOIN types whois ON whois.id = a.id_type)
GROUP BY id_type
+---------+------------+------------+
| id_type | whoisName | countWhois |
+---------+------------+------------+
| 0 | Company | 50 |
| 1 | Particular | 10 |
+---------+------------+------------+
Condition
SELECT a.id_condition, cond.name as condName, COUNT(a.id_condition) as countCondition
FROM (published a
INNER JOIN conditions cond ON cond.id = a.id_condition)
GROUP BY id_condition
+--------------+---------------+----------------+
| id_condition | conditionName | countCondition |
+--------------+---------------+----------------+
| 0 | New | 32 |
| 1 | Used | 28 |
+--------------+---------------+----------------+
I want to summarize the two queries in a single one but can´t figure out how. I was thinking something like this:
+---------+------------+------------+--------------+---------------+----------------+
| id_type | whoisName | countWhois | id_condition | conditionName | countCondition |
+---------+------------+------------+--------------+---------------+----------------+
| 0 | Company | 50 | NULL | NULL | NULL |
| 1 | Particular | 10 | NULL | NULL | NULL |
| NULL | NULL | NULL | 0 | New | 32 |
| NULL | NULL | NULL | 1 | Used | 28 |
+---------+------------+------------+--------------+---------------+----------------+
Is this possible?
Thanks and sorry if my English is bad, it's not my native language.
I am getting wrong column names when using union.
Here is what i do, i have two very big tables with same structure and different records, so here it is.
mysql> select * from e18 where `15` like '%car%' limit 1;
+------+------+----+------+------+------+------+------+------+------+------+------+------+------+------+-------------+------+------+------+------+------+------+--------+------+------+------+------+------+------+------+------+------+------+------+------+------+------+------+-----------+
| id | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 |
+------+------+----+------+------+------+------+------+------+------+------+------+------+------+------+-------------+------+------+------+------+------+------+--------+------+------+------+------+------+------+------+------+------+------+------+------+------+------+------+-----------+
| 2730 | 2730 | 18 | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | cars: stuff | NULL | NULL | NULL | NULL | NULL | NULL | 5 1 | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | yy |
+------+------+----+------+------+------+------+------+------+------+------+------+------+------+------+-------------+------+------+------+------+------+------+------+--------+------+------+------+------+------+------+------+------+------+------+------+------+------+------+-----------+
1 row in set
mysql> (select * from e8 where `15` like '%car%') union
(select * from e10 where `15` like '%car%') union
(select * from e18 where `15` like '%car%') limit 1;");
+------+------+----+------+------+------+------+------+------+------+------+------+------+------+------+------+-------------+------+------+------+------+------+------+------+--------+------+------+------+------+------+------+------+------+------+------+------+------+------+-----------+
| id | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 15 |
+------+------+----+------+------+------+------+------+------+------+------+------+------+------+------+------+-------------+------+------+------+------+------+------+------+--------+------+------+------+------+------+------+------+------+------+------+------+------+------+-----------+
| 2730 | 2730 | 18 | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | cars: stuff | NULL | NULL | NULL | NULL | NULL | NULL | NULL | 5 1 | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | yy |
+------+------+----+------+------+------+------+------+------+------+------+------+------+------+------+------+-------------+------+------+------+------+------+------+------+--------+------+------+------+------+------+------+------+------+------+------+------+------+------+-----------+
1 row in set
Union all and union return same result in this case.
There is only one row with word part "car" in it and it is in table e18.
For some reason, column names in result that i get from using usion are messed up, looks like i am missing something, any ideas what it is ?
Thanks in advance.
Union works by column position NOT name. But you have not specified the column position because you did * so it's in some order picked by the database, but not picked by you.
The name of the final result set is the name of the columns in the first query in the union.
The fix is easy: Write out the names of all the columns you want, and make sure to keep the order consistent between all three queries.
The columns are NOT sorted by name (so renaming the columns will not help you), the order is some internal order in the database.
Using * is considered poor practice: You don't know what you are getting, and if you only need some of the columns then using * retrieves more data then necessary, making things slower.
BTW Naming columns like this (by number) is very poor programming practice. How in the world do you keep things straight? Your columns have numbers, your tables have numbers. Are you trying to write obfuscated code? To make sure no one else can ever work on your code? Because if you are, this is one way to do it.
It appears that in the first query your field 15 is in numerical order with the other fields. In the 2nd query, it's showing up at the end. If you specify the fields you want (yes I know it's a lot of typing), then you won't have this problem.
Secondarily, given the number of NULLs in your return set and the fact that you're using multiple tables to store the same sort of data, your data probably isn't well normalized. Your database will be much easier to use (as well as faster and more efficient) if you normalize it.
And thirdly, 15 is not a reasonable field name -especially not when sibling fields are named for other numbers.