« first day (3334 days earlier)      last day (1520 days later) » 

12:27 AM
Oxygen toxicity is a condition resulting from the harmful effects of breathing molecular oxygen (O2) at increased partial pressures. Severe cases can result in cell damage and death, with effects most often seen in the central nervous system, lungs, and eyes. Historically, the central nervous system condition was called the Paul Bert effect, and the pulmonary condition the Lorrain Smith effect, after the researchers who pioneered the discoveries and descriptions in the late 19th century. Oxygen toxicity is a concern for underwater divers, those on high concentrations of supplemental oxygen...
 
 
6 hours later…
6:44 AM
Morning
 
 
1 hour later…
7:48 AM
Morning
 
8:44 AM
@PeterVandivier got it!
 
@PeterVandivier Folowwing
 
felt the need to @ someone just in case, now that we hit the vote count, i got all paranoid that the jira-bot wouldn't pick it up 😬
@McNets lol ty :p
 
9:24 AM
Hate to be that guy, but there's 2139 items in the backlog. I bet their sprint planning meetings are hell! :D
 
9:45 AM
@Philᵀᴹ where did you find that number? Is there a list?
 
2141 now 😬
@Philᵀᴹ yea, i don't expect it to get done any time soon, but the fact that it's not fast closed as "will-not-do" is the real win for me
 
I guess they have quick wins too
 
now any subsequent dupe requests get linked to an open item rather than a closed one
tbh, i kind of hope it scope-creeps too. like - please make that setting apply to any string-linted text with a newline escape regardless of the language (or more likely - make it user settings configurable per-language )
although that might suck sometimes when it comes into contact with regex
 
10:19 AM
@PeterVandivier couldn't you write an extension that does that?
I remember way back when I started using visual studio I had an extension that implemented Vim keystrokes
 
11:04 AM
@TomV-trytopanswers.xyz i'm sure it's possible :P i don't have any familiarity with contributing to VSCode and i don't expect i've got the literacy to do so
if i could do it quickly myself, i absolutely would. alas that i am but a lowly data shoveler unschooled in the subtleties of app development
 
11:18 AM
@TomV-trytopanswers.xyz Did it allow to exit ?
;)
@PeterVandivier do you need a shoulder where to cry?
 
usually
:p
 
 
1 hour later…
12:30 PM
@PeterVandivier Isn't it an option already? XD
 
Hmm I think it's only 1 character each time, not "\n" and obviously it's not limited to JSON (though the latter might be desirable behaviour)
 
i think thats single character breakpoints
yea
 
Maybe the limitation of "only do this when it's a .json file" isn't useful and could make the implementation quite a bit more complicated
 
🤔 that setting is actually "editor.wordSeparators": "~!@#$%^&*()-=+[{]}\\|;:'\",.<>/?"` not wordWrapBreakAfterCharacters
idk 🤷‍♂️, happy to have the ticket closed as "already implemented" too - as long as they show me how to do it
 
12:40 PM
wordseparators returns results when you search in the settings, wordwrapbreakaftercharacters doesn't
But if they drop the JSON requirement it's just adding \n here github.com/microsoft/vscode/blob/…
and allowing for more than 1 character of course
 
1:15 PM
@TomV-trytopanswers.xyz lol yea, i think that's likely the hard bit
 
@PeterVandivier The combination of that and the fact that the issue only wants it for one file extension break current logic yes
 
@TomV-trytopanswers.xyz naw... file type (or really language type) filters are alllllll over vscode
and if it were me actually defining scope of work for that, i wouldn't feel bound to take it literally as written for only json files
but maybe ¯\_(ツ)_/¯
who knows
life is a mystery
 
@PeterVandivier Yeah I think it could be useful in any string variable
 
2:05 PM
@Philᵀᴹ I spent far too much time playing it back when we had a BBC, ca. 1985.
 
2:30 PM
I added an extended example
Just to give the newbie an idea what we prefer.
 
haha, comment deleted
you're a good man, charlie brown
 
2:50 PM
🤔 should i have left it for the OP?
sry, thought you were prompting me to delete it
 
Pearls before swine most likely
But you never know, sometimes such an act produces a very valuable contributor
I remember Randi was too eager to get an answer in before anybody else at first, and look at his answers now
 
i mean, he's still eager af though
he just goes in and does some doctorate-level revisions and gets 10 upvotes per post
the game is the game
 
@PeterVandivier no
leave it
You could undelete your comment.
I guess a majority of the newbies are generally forum users and don't understand the true value of DBA.SE. If we can uphold the quality or encourage others to increase their own quality, then the community as a whole benefits.
That's why I (very rarely) sometimes increase the quality instead of posting my own answer.
 
@Johnakahot2use i closed the window too fast, cant undelete
 
I mean who needs unicorn points?
 
3:02 PM
Peter does
 
lol, i really do
 
Yeah, you so you can cast close votes
 
i'd have access to the close vote queues by now if i hadn't had a fit of madness and granted a couple bounties :p
 
What comes after isn't really interesting until "view deleted posts"
 
a couple?
I've got a long way to go.
 
3:08 PM
25k is boring
 
@Johnakahot2use But between casting close votes and viewing deleted questions, which of those do you actually use?
I mean none of that really excites
 
Yeah that comes after
 
Shouldn't we automatically do that --^ ?
;-)
Would keep the quality pretty high. :-D
 
And I can protect questions, yet the only one I wanted to protect I couldn't because it was on meta
 
3:10 PM
joking
 
 
1 hour later…
4:24 PM
@Johnakahot2use I never "protected" any question
 
 
4 hours later…
8:52 PM
anyone farting in the lift today?
 
9:09 PM
EAGL?
 
guess not
 
i didn't say no...
 
9:26 PM
@PeterVandivier I'm not sure what else I should add. Since you bring up psql commands, perhaps mentioning that my code is in a Go executable, and the query is in a Go package we import would be clarifying. Any other thoughts?
 
@swasheck nope only outside (the lift)
 
9:42 PM
0
Q: CREATE TABLE IF NOT EXISTS fails with privilege exception when table exists

Danek DuvallI discovered that CREATE TABLE IF NOT EXISTS will fail with insufficient_privilege if the role executing the statement does, in fact, not have the necessary privileges, but does so regardless of whether the table exists or not. I would expect that if the table exists, the statement succeeds with...

@DanekDuvall you state both that you cannot change the code that runs and also that you've tried a custom script to swallow the error, hence my confusion
if you've got a script that's trying to create tables using creds without the perms to do so, that's easily an XY problem, just... don't try to do things you shouldn't do... right?
 
Ah. Yes, I was being too dense. The code I "can't change" is in the other Go package (github.com/antonlindstrom/pgstore). I can't change it because I don't own it, though I've filed an issue to get the issue addressed, and if that happens, this problem goes away. I did change it, though, because I just modified the source that's downloaded and compiled as part of the build process, so I know my solution works. But that's not something I can do in production.
 
is golang your orchestration language? cause whatever framework or framework-substitute you're using is gonna be the devil in the details that determines your choice of workaround
 
I could fork, fix, and use the fork, of course, though I'd rather avoid that if at all possible.
 
@DanekDuvall but why not? ops restrictions?
@DanekDuvall yea, why can't you do that?
sounds like you've solved the problem?
 
I can't modify the code we download because the build process doesn't really allow for it (nor should it). I'd rather avoid forking because it's another thing to maintain, and if it's a problem I'm having, then presumably others might run into it as well at some point.
 
9:49 PM
i can see why you'd want to stick to the common distribution if possible, but...i mean... work off the fork that does what you need unless & until the upstream contribution is accepted, right?
 
And yeah, I've "fixed" the problem, I asked the question because I wanted to know a) whether my expectation was flawed, and if so, how; and b) if my expectation wasn't flawed, should I file a bug against Postgres.
@PeterVandivier Yes, that's definitely one of my options. Possibly stronger than opening up the schema for CREATE, which also does the trick.
 
file a PR to update the docs if you feel compelled to do so. the ambiguous wording could for sure be clarified a bit
relaxing security to serve your framework's flawed error handling is an option, but i'll let you decide if it's a good one
 
@PeterVandivier You think the behavior seems correct, though?
@PeterVandivier :)
 
tbh, kind of, yea
seems like it'd be a perms escalation issue otherwise at first glance
 
How?
 
9:54 PM
CREATE IF NOT EXISTS... must be a pseudo-transactional command (at first glance)
1. schema lock identifier
2. schema check identifier
3. schema modify
4. schema unlock identifier
if you don't have the permission to do #2, that doesn't mean you get to skip #1
hence the error
right?
 
But I do have the permission to do #2; it's #3 I don't have perms for. And I would expect to leave that transaction early if #2 comes back with "table exists", completely bypassing the perms issue in #3.
Unless it's #1 that's causing the perms issue?
 
you can get fancy and do #2 first by actually querying the system catalog, but that's a) the workaround i suggested in comments and b) a race condition that doesn't respect the IF EXISTS construct
 
Right; that's why I structured my solution the way I did, to avoid that race.
 
#1 requires perms for #3 afaik
(talking offhand, not based on docs, just assuming)
 
If so, then, that's the answer.
 
10:00 PM
you could do a deep dive in the source code to check all this, but it's late and i'm procrastinating on a migration script i need to button up
 
@PeterVandivier migration script = drink rum, right?
 
@McNets you joke, but literally also yes
 
Would psql-bugs be the right place to send this?
 
@DanekDuvall probs, yea. i'm not active on the psql-hackers distro though
 
@PeterVandivier Okay, thanks. Appreciate your time and brain cells!
 
10:04 PM
you could drop into the postgres slack channel too, there's a hackers subchannel, but it's not super active
@DanekDuvall my pleasure :)
 
10:43 PM
@PaulWhite9 - looks like this edit is wrong. that link was always to the topic index, right?
i've voted to reject in any case, just wanted to check
 
@PeterVandivier I rejected too. That edit would need a lot more explanation.
 
10:57 PM
👍
 

« first day (3334 days earlier)      last day (1520 days later) »