I want to count string separators from a MySQL query, mean if the field value is
like :-
1,2,3,4,5
as the string is comma separated so the separator count will be 4.
any idea then please share
THANKS,
you can try to count the length of string and minus the length of string without commas as follows:
LENGTH('1,2,3,4,5') - LENGTH(REPLACE('1,2,3,4,5', ',', ''))
select length('1,2,3,4,5') - length(replace('1,2,3,4,5', ',', ''))
I suggest the following design :
Table name : USER_HOBBIES
| USER_ID | HOBBY_ID |
1 1
1 2
1 3
2 2
2 4
2 5
And now you can easily count user hobbies for a given user :
SELECT count(*) FROM USER_HOBBIES WHERE USER_ID = <user-id>
although it requires another table it is much clearer and on a long list of hobbies this will be much faster than using a function for manipulating strings.
Related
I am basic on SQL queries and I need some help.
I have to select all string values which contains number e.g. 7 only on specific position in that string.
For example:
I have string: 987654321 and if on position 3 I will have number 7, then it should be selected.
So in example this string will be selected, because on 3rd position I have number 7.
Is there any SQL function for that, or something which could help me?
EDIT:
Example table
TABLE
Numbers Value
987654321 1
123456789 2
789009871 3
654321092 4
847949372 5
Output:
TABLE
Numbers Value
987654321 1
847949372 5
Statement:
SELECT table.numbers
FROM TABLE
WHERE substr(table.numbers,3,1)='7' <--- what to do here? --->
Many thanks in advance.
For a regex option, you may use MySQL's REGEXP operator:
SELECT *
FROM yourTable
WHERE num REGEXP '^[0-9]{2}7';
On Oracle, you could use REGEXP_LIKE:
SELECT *
FROM yourTable
WHERE REGEXP_LIKE(num, '^[0-9]{2}7');
You should use case statement.
select case when substr(stringcol, 3,1) = '7' then stringcol else "not valid" end as stringcol from <Table Name>
I need to get the total number of occurrences by separate server ID like below :
-----------------------------
logID serversID
-------------------------------
1 50,51,51,50
2 51,52
3 50,50
I want a result like this:
ServerID Count
------------ ---------------
50 4
51 3
52 1
Thanks you for your help.
Fix your data model! A string is the wrong way to store multiple values. A string is the wrong way to store numbers. The correct way to represent this data is to use a second table, with one row per logid and serverid.
If you are stuck with this data model and you don't have a reference table for servers, you can split the values . . . painfully:
select substring_index(substring_index(t.serversid, ',', n.n), ',', -1) as server, count(*)
from (select 1 as n union all
select 2 union all
select 3 union all
. . . -- as many as the biggest list
) n join
t
on t.servers like concat(repeat('%,', n.n - 1), '%')
group by server;
Here is a db<>fiddle.
I need output in following order(firstly, group by last 3 letters and then arrange in order based on the first 3 digits)
ColumnA
001_eng
004_eng
002_chn
003_usa
But order by ColumnA gives me
ColumnA
001_eng
002_chn
003_usa
004_eng
This is just sample data. I have hundreds of entries in this format and the values keep changing everyday. So, specifying all the entries inside the field is not a feasible option.
I'm not sure of how to use FIELD() in my case.
You can use FIELD:
select *
from tablename
order by
FIELD(ColumnA, '001_eng', '004_eng', '002_chn', '003_usa')
(please be careful if ColumnA is not in the list the field function will return 0 and the rows will be put on top)
or you can use CASE WHEN:
select *
from tablename
order by
case
when ColumnA='001_eng' then 1
when ColumnA='004_eng' then 2
when ColumnA='002_chn' then 3
when ColumnA='003_usa' then 4
else 5
end
or you can use a different languages table where you specify the order:
id | name | sortorder
1 | 001_eng | 1
2 | 002_chn | 3
3 | 003_usa | 4
4 | 004_eng | 2
then you can use a join
select t.*
from
tablename t inner join languages l
on t.lang_id = l.id
order by
l.sortorder
(with proper indexes this would be the better solution with optimal performances)
You can use SUBSTRING_INDEX in case all ColumnA values are formatted like in the sample data:
SELECT *
FROM mytable
ORDER BY FIELD(SUBSTRING_INDEX(ColumnA, '_', -1), 'eng', 'chn', 'usa'),
SUBSTRING_INDEX(ColumnA, '_', 1)
Demo here
you can use substring() and get order by
SELECT *
FROM table_name
ORDER BY SUBSTRING(ColumnA, -7, 3);
I have fields like this:
-----------------
id | name
-----------------
1 | name123
-----------------
2 | name
-----------------
3 | name456
-----------------
4 | name
I want to extract rows which have digit in name and a field that contains the number like this
------------------------------
id | name | number
-----------------------------
1 | name123 | 123
-----------------------------
3 | name456 | 456
how can we find the records that have digit and extract digit as a new field?
Here is another way to do with mysql
SELECT
id,
name,
SUBSTRING(
name,LEAST (
if (Locate('0',name) >0,Locate('0',name),999),
if (Locate('1',name) >0,Locate('1',name),999),
if (Locate('2',name) >0,Locate('2',name),999),
if (Locate('3',name) >0,Locate('3',name),999),
if (Locate('4',name) >0,Locate('4',name),999),
if (Locate('5',name) >0,Locate('5',name),999),
if (Locate('6',name) >0,Locate('6',name),999),
if (Locate('7',name) >0,Locate('7',name),999),
if (Locate('8',name) >0,Locate('8',name),999),
if (Locate('9',name) >0,Locate('9',name),999)
),LENGTH(name)
) as number
from users
having number <> '' ;
you can use MySQL's string conversion on an int to strip out the name like so
SELECT
t.id,
t.name,
REVERSE(REVERSE(t.name)+ 0) AS num,
REPLACE(t.name,REVERSE(REVERSE(t.name)+ 0),'') AS actualname
FROM foobar t
HAVING num <> 0
the trick with this is by adding a 0 mysql is comparing the numeric value in the name... however the name has to start with a number... so I reverse it do the calculation and then reverse again... NOTE all of your names have to start with the name and end with a number for this to work for all of them
FIDDLE DEMO
EDIT:
since you say that some can start with a number and others end with a number.. then try this
SELECT
t.id,
t.name,
REVERSE(REVERSE(t.name)+ 0) AS num,
REPLACE(t.name,REVERSE(REVERSE(t.name)+ 0),'') AS actualname
FROM foobar t
HAVING num <> 0
UNION ALL
SELECT
t.id,
t.name,
t.name + 0 AS num,
REPLACE(t.name,t.name + 0,'') AS actualname
FROM foobar t
HAVING num <> 0
ANOTHER DEMO
Another way, assuming the number you want is at the end of the string. REVERSE() to put the number part in front, then CONVERT() to make it a number and strip off the text, then REVERSE() again WHERE name ends in a number. Feels like a kludge though:
select id, name, reverse(convert(reverse(name),signed ))
from tbl
where name REGEXP '[0-9]+$';
SQL Fiddle Example
SCHEMA / DATA for TABLE :
SubscriberId NewsletterIdCsv
------------ ---------------
11 52,52,,52
We have this denormalized data, where I need to count the number of comma separated values, for which I am doing this :
SELECT SUM(len(newsletteridcsv) - len(replace(rtrim(ltrim(newsletteridcsv)), ',','')) +1) as SubscribersSubscribedtoNewsletterCount
FROM TABLE
WHERE subscriberid = 11
Result :
SubscribersSubscribedtoNewsletterCount
--------------------------------------
4
The problem is some of our data has blanks / spaces in between the comma separated values, if I run the above query the expected result should be 3 (as one of the value is blank space), how do I check in my query to exclude the blank spaces?
EDIT :
DATA :
SubscriberId NewsletterIdCsv
------------ ---------------
11 52,52,,52
12 22,23
I need to get an accumulative SUM instead of just each rows sum, so for this above data I need to have just a final count i.e. 5 in this case, excluding the blank space.
Here's one solution, although their may be a more efficient way:
SELECT A.[SubscriberId],
SUM(CASE WHEN Split.a.value('.', 'VARCHAR(100)') = '' THEN 0 ELSE 1 END) cnt
FROM
(
SELECT [SubscriberId],
CAST ('<M>' + REPLACE(NewsletterIdCsv, ',', '</M><M>') + '</M>' AS XML) AS String
FROM YourTable
) AS A
CROSS APPLY String.nodes ('/M') AS Split(a)
GROUP BY A.[SubscriberId]
And the SQL Fiddle.
Basically it converts your NewsletterIdCsv field to XML and then uses CROSS APPLY to split the data. Finally, using CASE to see if it's blank and SUM the non-blank values. Alternatively, you could probably build a UDF to do something similar.