SQL Server : Breaking Comma Delimited String and matching values in a table

The name of the pictureThe name of the pictureThe name of the pictureClash Royale CLAN TAG#URR8PPP



SQL Server : Breaking Comma Delimited String and matching values in a table



I am using SQL Server and have a "tags" column that has a string of of comma-separated tags. Is there a way to break the string up and place in another table and have them match to be able to easily see liked tags?



Row 1:


3years_andmore,access_ccc,access_sdl,associate_iii,ccc_tickets,desoto_counter,phone_call__property_tax,ticketing,trainer



Row 2:


3years_andmore,access_ccc,access_sdl,associate_iii,ccc_tickets,desoto_counter,phone_call__dmv,ticketing,trainer



Row 3:


5_minutes,access_ccc,access_sdl,associate_ii,ccc_tickets,desoto_counter,lessthan_3years,phone_call__operations___title_by_mail_inquiry,trainer



Row4:


access_ccc,access_sdl,associate_ii,ccc_customer_request_manager_other,ccc_tickets,desoto_counter,lessthan_3years,phone_call__associate_requesting_manager__customer_requesting_mr_brierton,trainer



There is no real order and not the same tags in field but is there a way to at least sort into a new table and break them up and match to see what tickets have the same tags?


SELECT
[id],
[url],
[external_id],
[type],
[subject],
[description],
[priority],
[status],
[recipient],
[requester_id],
[submitter_id],
[assignee_id],
[organization_id],
[group_id],
[collaborator_ids],
[forum_topic_id],
[problem_id],
[has_incidents],
[due_at],
[tags],
[via],
[custom_fields],
[satisfaction_rating],
[sharing_agreement_ids],
[followup_ids],
[ticket_form_id],
[created_at],
[updated_at],
[channel]
FROM
[Brierton].[dbo].[Tickets]
WHERE
created_at BETWEEN '2017-11-01' AND '2018-08-23'
AND ',' + tags + ',' LIKE '%,' + 'ccc_tickets' + ',%'





Is there a finite number of possible tags? Ie do they come from some sort of existing list of possible tags, or are they some sort of input value and new tags are always possible?
– mdsimps99
1 hour ago





The reason you are struggling here is because that design is really bad. It violates 1NF by shoving a bunch of values into a single tuple. You don't want a query to view this, you need to fix the design and then fix the data. There are tons of ways to split strings. Here is one. sqlservercentral.com/articles/Tally+Table/72993 Several other excellent choices can be found here. sqlperformance.com/2012/07/t-sql-queries/split-strings It looks like you have several of these in your table.
– Sean Lange
1 hour ago






In all honesty, I'd probably consider a trigger and secondary table containing [ID], [TagID], and one tag as the data elements, indexed on ID, tag, TagID in order (TagID to ensure that no duplicate entries in the tag list cause an issue), Otherwise, the string splitting functions are your best bet. This assumes, of course, that you are allowed by your licensing to put a trigger on the underlying table.
– Laughing Vergil
1 hour ago





1 Answer
1



Since version 2016 there's a built in string_split() function. To get the tags per ticket in a "more relational way" you could use:


string_split()


SELECT
t.id,
x.value
FROM
tickets t
CROSS APPLY
(SELECT value
FROM string_split(t.tags, ',')) x;



From there you could aggregate to get the most used tags.


SELECT
x.value,
COUNT(*)
FROM
tickets t
CROSS APPLY
(SELECT value
FROM string_split(t.tags, ',')) x
GROUP BY
x.value
ORDER BY
COUNT(*) DESC;



db<>fiddle






By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Popular posts from this blog

Firebase Auth - with Email and Password - Check user already registered

Dynamically update html content plain JS

Creating a leaderboard in HTML/JS