LeetCode 196: Delete Duplicate Emails [SQL]

Solution and walkthrough of leetcode database problem 196: Delete Duplicate Emails. I'm using MySQL but this solution should work in any SQL dialect such as PostgreSQL SQL Server, etc.
Link to the problem: leetcode.com/problems/delete-...

[@frederikmuller]

If you want to have access to more free SQL problems, check out StrataScratch: stratascratch.com/?via=frederik

[@tongkezhao2188]

I like the way that you always give interview tips relate to each question at the end of a video. Good job!

[@andrew5291]

Thank you! You are the best! Very helpful : )

[@ahmedboutaraa8771]

thank you

[@stellaueda4565]

Thx again

[@pranitmhatre8924]

I got exactly the same problem in my coding Interview.

[@joshuaperez2271]

Thanks for explaining this concept!

[@frederikmuller]

Glad it was helpful!

[@sheetalm5959]

Thank you for the video, I am new to SQL coding. Can you please explain how did you get the 'Intuition' behind using self join here? Thanks.

[@frederikmuller]

Sure, the intuition behind using self join is that we compare the table to another copy of itself to see if we find any matching email addresses with higher ID values than the current row we're looking at (these will be the duplicates). We're then deleting these rows from our data base by using the DELETE statement since by the problem definition we only want to keep the entry with the lowest ID value. Hope that makes sense!

[@sheetalm5959]

@@frederikmuller Got it. Thank you🙂

[@utsavsagar6665]

Please let me know where i am wrong in this in syntax 🙏 delete from person where email in(select email from person group by email order by id having count(email) >1);

[@satyamgavane9345]

what mistake i have done in my code please can you tell me with cte as(select *,row_number() over (partition by email order by id) as r from Person) delete from cte where r>1;

[@shriyamjalan247]

Hi Frederik, I was trying to write a query by selecting the duplicate emails and then deleting it. Can you help me explain what I am doing wrong ? Also, if it is possible to write a query in this way. I selected the duplicate emails in following way: select email from Person group by email having count(Email)>1 I put this in where clause and then tried deleting it.

[@inzamamsafi5261]

you need a minimum id while grouping. so take that into consideration as well

[@mickyman753]

you can do this select temp.email from (select email from Person group by email order by id) as temp having count(temp.Email)> 1;

[@shuchivardan375]

can we use row_number and partition by email function?

[@frederikmuller]

actually yes, and then just delete everything with row_number > 1 if you order by Id in ascending order.

[@frederikmuller]

good idea!

[@christophercastro6078]

I am assuming this wouldn't work if the table had more than 1 duplicate email. Only works if there is at most 1 duplicate email. The statement in the problem descriptions states: Write a solution to delete ALL duplicate emails... Which could be interpreted two ways.

[@christophercastro6078]

Never mind, this works with several identical emails since there is only one id that is the smallest from the bunch!

[@frederikmuller]

Thanks for bringing this up, this will help people.

[@utsavsagar6665]

Thank you for the help but still I am unable to understand what is wrong with this syntax.

[@uddipangoswami8427]

Why this is not working in Oracle?

[@kwoktungwong9486]

Hi, why do we delete p2 here not p1?

[@frederikmuller]

Because we’re supposed to keep the email entry with the lowest id. Since we have a filter on p1.Id < p2.Id, p2 will always be the email duplicate entry with the higher id. There can be more than one duplicate but this query deletes every p2 that can be joined based on the conditions: same email and bigger id as another entry (our p1). Hope that makes sense.

[@paper-studio]

this code not working, not even the discussion code working, there is some issue I guess

[@ekdumfarzi8039]

delete p2 from... why did u use p2, we dont write delete statement like that...where can i learn more about that way of writing..

[@frederikmuller]

I just replied this to another comment asking the same question: Because we’re supposed to keep the email entry with the lowest id. Since we have a filter on p1.Id < p2.Id, p2 will always be the email duplicate entry with the higher id. There can be more than one duplicate but this query deletes every p2 that can be joined based on the conditions: same email and bigger id as another entry (our p1). Hope that makes sense. The DELETE statement works just like the SELECT statement but instead it deletes rows instead of selecting them for output. It doesn’t show up often in interviews so I don’t think you need to read up more on it. The trick with this question is to find what you want to delete by using a JOIN which makes the question tricky and non-standard.

[@vasiliyosipov43]

I was dissapointed about that too, because we can't write anything after delete except from. But I guess it's the MySql special option. Didn't work for PostgreSQL btw

[@sungwonjudycho]

Hi Fredrik! Thank you so much for your videos; it's been really helpful while trying to study SQL by myself. I had some questions about your explanation and would be great if I could get some pointers on what I'm missing here. Q1. why "WHERE" is not accepted? [My Attempt] DELETE p2.id FROM Person p1 JOIN Person p2 ON p1.email = p2.email WHERE p1.id < p2.id Q2.I'm not sure I understand how the [AND] clause could be used on this Join table. Shouldn't ON only accept when the two columns are identical? [ACCEPTED ANSWER] DELETE p2.id FROM Person p1 JOIN Person p2 ON p1.email = p2.email AND p1.id < p2.id