Not sure but maybe it was meant to be something like this:

SELECT
  SUM(
    CASE
      WHEN int_datahora_entrada <= '2020-06-30' AND int_datahora_saida >= '2020-06-01'
      THEN date_part('day', age(greatest(int_datahora_entrada, '2020-06-01'), least(int_datahora_saida, '2020-06-30')))
      ELSE 0
    END
  ) AS "06"
FROM
  int_internacao
;

Andry, you are a SQL GOD!!!!!

Morning

@matmany That's awfully kind of you to say, I'm just glad to help

Morning

is @mustaccio about? i kind of want to challenge your answer but didn't want to throw shade. by doing it in comments or not giving you ample chance to reply engage. i don't know how to read your answer without it requiring that there exist a maintained piece of meta documentation that provides a reference for which platforms use which verbiage; which seems... like maybe a pain in the butt and disingenuous to expect all reviewers to know about

for example, it looks like this edit should be accepted based on these docs (with an edit to that first instance of primary/secondary to say source/replica), but requiring everyone to look that up every time seems... annoying?

also - TA has spoiled me for the edit timer 😅

@PeterVandivier I wouldn't worry about the requiring part. It seems to me there will always be people willing to fight the fights that few, if any, care about. Therefore, you don't need to put that as a requirement to them, they'll insist on volunteering to do that job

@AndriyM yea... but... i mean... the question is meant to establish consensus so far as we can about what is appropriate in that situation. "requiring" is just a shorter word to say that. what "should" be done with that edit - reject or approve? if the policy is "look it up in the docs" then... we require/encourage/politely-suggest-you-consider going to an external resource first

@PeterVandivier If that edit is an example of what you are asking about then the specific scenario you have in mind is, "There's an old answer, someone spotted an offensive (to them) terminology and wants to change it – what should they do? Should they be required/encouraged/etc. to look up the docs or just get on with it?" Is that what you are asking about?

@AndriyM yes, exactly. the edit does nothing but change "sensitive words". i figured those kind of edits have started feeling a bit spammy to me and i wondered if they had for anyone else.

that said, i figured there's merit to considering them "acceptable", hence the meta question

I mean, we are talking about a person who's offended by terminology that's been in use for decades – and we are expecting them to be reasonable about that?

but re: @mustaccio's answer, i feel like it's a little disingenuous to say "yes, leave it to the review queue" but also "do it on a case by case basis". if the policy is subtle, then we can do better - e.g. a bulk edit or something as has been suggested in the past

@AndriyM not sure what your question is there? i feel like the edit i linked is a perfect example. it provides no "improvement" to the post beyond changing the master/slave terminology. My question is - do we accept those "by policy" - or do we require other improvements

perhaps i haven't articulated my concern properly? i accepted the edit based on mustaccio's answer (it follows the edits posted in the mysql docs i found). but if we're saying that edit should have been rejected if it used "leader/follower" instead of "source/replica", then that feels a little too subtle to be broadly helpful as a policy for the review queue. seems like it could get confusing and we should suggest a bulk edit from a more comprehensive lookup to remove the issue

hope that makes more sense?

The wording in mustaccio's answer might be a little too generic to cover the edit in question. I don't believe that edit to be warranted. Even though the docs have changed their terminology, they also provide you with the old terms for reference, the terms that the edited post was using. There was no need to edit the post just to replace those terms, because the docs still allow you to understand what the post was talking about.

It's like changing a post to replace a comma join with the explicit join syntax. Someone's been taught to use that kind of terminology or that kind of syntax. You don't need to edit a post if the meaning is clear as it is.

i mean... if you accept mustaccio's answer as the policy, then i feel like you kind of have to support the edit in question... if you disagree with the edit, i think you kind of have to say that "reject" is the policy, or at the very least make mustaccio's answer more restrictive to say "(1)follows the docs and (2)the previous edit was unclear" - in which case why even bother with the (1) when (2) is sufficient by itself to inform your decision

Yes, as I said, his wording might be somewhat lacking. And especially after your explanation, it's easy to see that there's room for improvement. But for what it's worth, I personally agree with his suggestion in spirit.

fair enough :)

agree to disagree, i suppose 🤷‍♂️

Always happy to be civil about it

incorrect, i'm already planning to fight you if we ever meet in person

prepare yourself

there can be only one

What should I do? Sign up for some MMA courses or something?

probably a good idea

I'm not very experienced in this, to be honest

All right, will keep that in mind, thanks :)

well you have time to practice, with the covid i imagine we won't have much opportunity to meet for some time

You are probably right, there's my chance

Morning

looking at the edit-in-question again, the edit was made only to the question, the answer still has master/slave verbiage 😂😅😭

/headdesk

Afternoon all - holiday here in Ireland - going for a spin with the financee - later dudes!...

Morning y'all

@PeterVandivier y'all realize I can read you complaining, right?

Wouldn't have touched it if it hadn't been bumped, but if master/slave is the hill you're willing to die on, in 2020 go off

@bbaird changing the verbiage in the question but not in the answer makes the thread as a whole more ambiguous. I think you might be reading chat a bit defensively though as I don't think there's been complaining here. in fact your edit has been accepted. i voted to accept based on the answer in meta that i have structural concerns about. but i still think if you're going to make the edit to the question, you should edit the answer as well to preserve readability

@PeterVandivier +1

:) aw shucks whicho plus wun

I must not have seen the verbiage in the answer (I was focused fixing the title of the question based on how it's currently described in the MySQL docs). I'll edit now.

I honestly don't see much point editing out the old terminology when a) it's been around for years (so it's unlikely for someone to stay confused for a long time when lots of people know what it's about), and b) the docs clearly give you a reference which new term corresponds to which old term.

I mean, editing an old post solely for that doesn't seem to have much point for me

It was bumped. I don't think going back and finding old posts to edit is anything I want to undertake, but when it came up I cringed.

And it's not like the terminology has been out of favor recently... it's been out of favor for at least 5 years at this point.

Well, it's a new thing for me

New as in the last several months

I think there is (justifiably) renewed interest in removing the last vestiges of it, but I remember MS getting rid of the language prior to 2016.

@PeterVandivier thank you for the clarification.

Microsoft Style Guide for master/slave, master/subordinate

Also from e.g. docs.microsoft.com/en-us/azure/mysql/howto-data-in-replication

I'm not taking a particular position here, just adding information

The MySQL docs omit the language too, except for in the code which seems to defeat the purpose somewhat.

@PeterVandivier This seems to be a grey area. I did reject the edit, as you can see, the reasoning being that 1) MySQL commands themselves still refer to "source"/"replica" as master and slave, and 2) the answer uses "master"/"slave". It would seem confusing to me to see the question asking about the "source" and "replicas" and reading the answer that doesn't have those terms in it.

You're welcome to challenge or throw shade or what not. I expressed my opinion, I don't require everyone to agree with it.

@mustaccio well if throwing shade is allowed... :-P

you have summoned @billinkc

Let's throw shade at Paul, he's always a worthy target

That's useless, he's in the dark at the moment I surmise. We'll have to wait for few hours

And it would probably be the wrong angle too

Anyway, signed on this morning to say hello, not stir controversy around edits to a bumped post

@bbaird the meta question predates your edit - it's an ongoing issue. the pot stirs itself just fine :-P you've caused no drama

Not the first time it's stirred in a similar way either, or the first pot to be stirred like that in the first place

Is there a 'causing a ruckus' badge I can get out of this?

if there were, i'd have it golded out probably

I might have given you mine

oh the irony, by policy mysql uses "allowlist", but peep this URL 😂 dev.mysql.com/doc/refman/8.0/en/…

@mustaccio while i'm still not a huge fan of the per-platform policy on principle, i'd genuinely like to include a lookup grid for what's acceptable/rejectable if that's the preference of the community. would you be amenable to adding one to your answer or switching it to a CW post? something like...

it feels super cumbersome (part of why i'm not a fan), but if that's to be the stated community consensus, i feel like it's sort of necessary to provide that kind of reference

hopefully in a less-clunky format

(plaintext copy of that image)

I don't want to spend more time on this topic than I already did, but you're welcome to -- I changed my answer to CW

fair enough, ty

ah boooooo... i forgot you can't do table markup... laaaaame

edited to include references

plz add more as appropriate at your convenience :) (all, not just mustaccio)

and an upvote is quickly removed, :-p i'm sensing the edit was not welcome by all

Within SQL server specifically I think primary/secondary are used in AGs, publisher/subscriber for replication.

given the assertion that there's official preferences given in documentation, i think that specific post should only list replacements where there's a first-party reference link (e.g. no wikipedia links)

as for AG and replication, i don't think those have ever used master/slave in broad common parlance so it's probably a non-issue. at least for my part i learned primary/secondary and pub/sub when i learned both of them 🤷‍♂️

I agree

Ireland's fearless struggle against Covid-19 continues unabated! The HSE (our Health and Safety Executive) is advising "young people to have sexual contact online or over the phone to stop the spread of Covid-19." - hmmm, wonder how that'll work as a strategy! :-)

Surprised that was ruled as scandalous by the ASAI

Reference for the comment irishtimes.com/culture/tv-radio-web/…

@billinkc I have no problem with a company (and this one in particular) advertising its products and instructing its clients on how to use the product (and this one in particular). However, this ad was crass and crude - more importantly, a significant majority of the complainants were women, so it's not a I-have/don't-have periods issue!

Good to know. Being a crass American, yes, I repeat myself, saw no issue with it

@billinkc My original comment about Covid-19 was more aimed at stimulating (pardon the pun :-) ) debate about the "online or phone sex" issue?

Plus, the HSE is not "advserting" a product and therefore doesn't fall under the ASAI purview!

I guess I was just surprised that they'd make any sort of advise on sexuality outside of traditional marriage

And the my mind linked that to the advertising thing. Was not trying to rile things up though

@billinkc They're issuing health advice - they would/have issue(ed) similar advice to homosexuals - indeed sexuality (or marital status) wasn't specified by the HSE! Ireland has changed a lot in the last 25 years - divorce, gay marriage referendum passing! Anyway, it's good to shoot the breeze! :-)

Ok, just wanted to make sure you didn't feel like I was taking pot shots at y'all. Lord knows there's a few things going on in our country that are a wee disgraceful