Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-06-10 Thread Kalyan Reddy
Based on the feedback so far, deadline_time got more votes. So, renaming the column to deadline_time. On 2025/05/28 07:01:55 Buğra Öztürk wrote: > I agree deadline.time could be a naming problems in the future. deadline_at > looks more intuitive to me > > On Wed, 28 May 2025, 08:01 Amogh Desai

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-28 Thread Buğra Öztürk
I agree deadline.time could be a naming problems in the future. deadline_at looks more intuitive to me On Wed, 28 May 2025, 08:01 Amogh Desai, wrote: > If the user doesn't get to see the name directly as part of the interface, > deadline.expiration or deadline.deadline_time doesn't > bother me a

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-27 Thread Amogh Desai
If the user doesn't get to see the name directly as part of the interface, deadline.expiration or deadline.deadline_time doesn't bother me as much. Somehow I like `deadline.deadline_time` better, that said, I am ok with either as long as we document it well :) Thanks & Regards, Amogh Desai On

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-27 Thread Kalyan Reddy
Based on the comment here(https://github.com/apache/airflow/pull/50957#discussion_r2103894496) , we do delete these records after expiration. So, we can call it expiration > if user doesn't see it, then it is much less important (and it can be > private). What is the interface for the user? I

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-27 Thread Daniel Standish
Oh boy, well, as they say, naming and cache invalidation. But expiration, that does not sound good. That sorta suggests "this is the time when this deadline record expires" as in, we delete it after this date -- or something. Deadline time seems better to me. Anyway. Does the user see any of thi

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-27 Thread Aritra Basu
I like expiration, works with and without the table name. +1 from me in that one. -- Regards, Aritra Basu On Tue, 27 May 2025, 11:40 pm Ferruzzi, Dennis, wrote: > Ive been tossing around a few ideas and I think the best alternative I > came up with was `deadline.expiration`. It's clear what it

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-27 Thread Ferruzzi, Dennis
Ive been tossing around a few ideas and I think the best alternative I came up with was `deadline.expiration`. It's clear what it means and doesn't feel too awkward in use. I'm also good with deadline_time I guess, but `deadline.deadline_time` "feels" awkward IMHO - ferruzzi _

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-27 Thread Daniel Standish
Generally speaking, it's not a huge problem when you use a reserved word in a column name. You add backticks or whatever. And we don't write the sql manually anyway. That said, plus one for deadline_time. Unless there's some other modifier better than "deadline_". Don't love the "_at" naming.

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-26 Thread Kalyan Reddy
Thanks Jens. In light of this, I'm proposing to rename it to "deadline_at" . I've updated the PR as well. But "deadline_time" also sounds good. We can go with that as well. I can update PR with what we agree here. On 2025/05/27 01:47:57 Wei Lee wrote: > Not sure whether "deadline.deadline_time"

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-26 Thread Wei Lee
Not sure whether "deadline.deadline_time" is an ok idea 🤔 I also worry about what Jens said. Best, Wei > On May 27, 2025, at 4:26 AM, Jens Scheffler > wrote: > > Hi Denis, > > from point of content of the name matching to the meaning I agree - but I > have doubts because the column name con

Re: [Lazy Concensus] Naming tweak in Deadlines table

2025-05-26 Thread Jens Scheffler
Hi Denis, from point of content of the name matching to the meaning I agree - but I have doubts because the column name conflicts with the SQL type that has the same same. As we use ORM this is mostly fine but other logic running SQL on the DB might get into conflict if column name "time" is