|
Corla Plankun posted:can you post a snippet that triggers this? I've never seen this before I hadn't either, and no doubt the original regex was absolutely bonkers, but I never would have thought it could lock up the entire process replicating case (tested up to 3.9): code:my homie dhall fucked around with this message at 14:48 on Mar 25, 2021 |
#
?
Mar 25, 2021 14:40
|
|
|
|
| # ? Apr 20, 2024 03:16 |
|
|
lol
|
|
#
?
Mar 25, 2021 14:48
|
|
|
my homie dhall posted:pretty cool “serious” language the only good gil is selling colecovisions in springfield
|
|
#
?
Mar 25, 2021 15:43
|
|
|
its what i buy my potions with
|
|
#
?
Mar 25, 2021 15:47
|
|
|
this is in the nature of regexes that they can gently caress you up. the subset of regex that doesnt have possibly exponential runtime is basically globs (this is why no cloud dealio allows you to grep their logs w real regexes) if you want nonblockin python you need separate processes the theory sez: regex poo poo is linear time. backrefs gently caress it up to npcomplete. the peeps who actually make regex systems say: lol shove those backrefs in there bob dobbs is dead fucked around with this message at 18:18 on Mar 25, 2021 |
|
#
?
Mar 25, 2021 17:50
|
|
|
Menacer posted:It has been proven by history that creativity lies in the individual mind and not in the team. The team is the nemesis of progress and change. The team is a convenient hideaway for the mediocre and the weak. ... "The team is stronger and more capable than the one" is the mission statement of collaboration. It is also the motto of communism. a couple years ago i spent a week at stanford playing b-school games. one of them was this wilderness survival exercise, where you have to rank 10 items by their usefulness in the described survival situation. you do the ranking by yourself, then get together with your team and discuss, to come up with a final team ranking by consensus. at the end, your personal and team rankings are compared to "the right answer", as given by a navy seal. the point was to demonstrate that in virtually all cases, your team's consensus ranking was closer to correct than your personal ranking, so you benefitted by being in the team, and this is indeed what the data showed. however, there was like one case where a guy beat his team. turns out he was an expert outdoorsman, so in his case the team would have done better if they all just shut up and listened to him. so i asked the prof, "well what happens if there is someone on the team who acts like a subject-matter expert, in which case we really should listen to him, but in fact he is not an expert, yet the rest of us are too inexperienced ourselves to realize that he is full of poo poo?" and she gave me this wide-eyed look like i had asked a research question. she started talking about the "confidence-competence paradox" which i guess is another name for dunning-kruger. in any case, i wonder about the frequency of this. i bet for every 1 time there is a real expert who the team should defer to, there are 10 times that you actually have some rear end in a top hat like the quoted author, who is no genius, just a misanthrope
|
|
#
?
Mar 25, 2021 17:55
|
|
|
Speaking of regexes, I just had to fix a bug my predecessor left me where a bit of code was using a regex to look for things like "{foo}" in a template and replace them with the value of map["foo"]. This worked find for ages until one of the replacement values, which are user-supplied, contained a dollar sign and suddenly the calls to replace() were complaining about invalid group references. This caused email reminders to fail silently for a month because my predecessor also didn't set up log monitoring on any of the servers. Fortunately, this particular web application was written with a new enough version of (ugh) Grails that it supports Java (ugh) 8, so I can move it to one of the servers I created, which does have log monitoring.
|
|
#
?
Mar 25, 2021 19:00
|
|
|
DELETE CASCADE posted:a couple years ago i spent a week at stanford playing b-school games. one of them was this wilderness survival exercise, where you have to rank 10 items by their usefulness in the described survival situation. you do the ranking by yourself, then get together with your team and discuss, to come up with a final team ranking by consensus. at the end, your personal and team rankings are compared to "the right answer", as given by a navy seal. the point was to demonstrate that in virtually all cases, your team's consensus ranking was closer to correct than your personal ranking, so you benefitted by being in the team, and this is indeed what the data showed.
|
|
#
?
Mar 25, 2021 19:34
|
|
|
that's why you gotta interview well and hire well and avoid single points of failure (or incompetence) in your organization
|
|
#
?
Mar 26, 2021 01:40
|
|
|
DELETE CASCADE posted:a couple years ago i spent a week at stanford playing b-school games. one of them was this wilderness survival exercise, where you have to rank 10 items by their usefulness in the described survival situation. you do the ranking by yourself, then get together with your team and discuss, to come up with a final team ranking by consensus. at the end, your personal and team rankings are compared to "the right answer", as given by a navy seal. the point was to demonstrate that in virtually all cases, your team's consensus ranking was closer to correct than your personal ranking, so you benefitted by being in the team, and this is indeed what the data showed.
|
|
#
?
Mar 26, 2021 01:42
|
|
|
lmao i jjust inserted a bunch of garbage rows in one of the production dbs and i dont have delete privs on it
|
|
#
?
Mar 26, 2021 02:30
|
|
|
that generally makes sense, delete flag only
|
|
#
?
Mar 26, 2021 02:41
|
|
|
i dont disagree, but that would require major changes to the uniqueness constraints and backend & we dont havee a time budget for that ill just have to whine at the dba
|
|
#
?
Mar 26, 2021 02:46
|
|
|
hmm is there a way to recursively delete empty "folders" in S3?
|
|
#
?
Mar 26, 2021 03:43
|
|
|
there is not!!!!!! folders in s3 aren't real
|
|
#
?
Mar 26, 2021 03:52
|
|
|
i know, hence the quotes but if you go "aws s3 ls" it will show like A/ B/ C/ as if they are folders. but only C/ contains something, A & B are just empty objects or whatever i want to detect & delete A/ and B/ with as little manual work as possible because there are a whole lot of them like find -empty -delete u know
|
|
#
?
Mar 26, 2021 03:55
|
|
|
you get to write a script. hope you don't need to invoke that slow rear end aws command line python nonsense too many times! (you will)
|
|
#
?
Mar 26, 2021 03:58
|
|
|
uuuhgghghhhhhhhh
|
|
#
?
Mar 26, 2021 04:04
|
|
|
but thx im gonna do a recursive ls & build a list i guess
|
|
#
?
Mar 26, 2021 04:05
|
|
|
not gonna lie i realized embarassingly recently that i can avoid having "index.html" in the address bar of my site's pages if i just navigate to directory instead of directory/index.html of course that knowledge is now of no use to me as my site redesign just uses page hashes and a script to change an iframe src depending on which one it is
|
|
#
?
Mar 26, 2021 04:36
|
|
|
well that wasnt too bad i went through a listing & added everything with size > 0 (files) to a trie and everything else ("folders"/prefixes) to a set. then i removed all the elements from the set that were present as prefixes in the trie. presto, the set now contains only empty prefixes!
|
|
#
?
Mar 26, 2021 07:42
|
|
|
hbag posted:not gonna lie i realized embarassingly recently that i can avoid having "index.html" in the address bar of my site's pages if i just navigate to directory instead of directory/index.html in before Shaggar says "use asp.net routing and cshtml" it's actually good though
|
|
#
?
Mar 26, 2021 08:09
|
|
|
effective engineering organisations do x at rate y -> if we improve our x we will become an effective engineering organisation
|
|
#
?
Mar 26, 2021 09:17
|
|
|
oh no the sample output data from the ai is dogshit who could possibly have predicted this
|
|
#
?
Mar 26, 2021 09:25
|
|
|
DELETE CASCADE posted:a couple years ago i spent a week at stanford playing b-school games. one of them was this wilderness survival exercise, where you have to rank 10 items by their usefulness in the described survival situation. you do the ranking by yourself, then get together with your team and discuss, to come up with a final team ranking by consensus. at the end, your personal and team rankings are compared to "the right answer", as given by a navy seal. the point was to demonstrate that in virtually all cases, your team's consensus ranking was closer to correct than your personal ranking, so you benefitted by being in the team, and this is indeed what the data showed. the team is less competent at assessing the problem so naturally a worse outcome is expected. when poo poo like this happens everyone suffers and the bamboozler person in particular, his reputation is damaged. others in the team may say "I told you so" or "if we had done it my way,", these people failed too but in a different worse way.
|
|
#
?
Mar 26, 2021 12:05
|
|
|
Powerful Two-Hander posted:oh no the sample output data from the ai is dogshit who could possibly have predicted this An ai predictive model
|
|
#
?
Mar 26, 2021 14:17
|
|
|
Powerful Two-Hander posted:oh no the sample output data from the ai is dogshit who could possibly have predicted this you're supposed to feed the output back in until you reach a fixed point
|
|
#
?
Mar 26, 2021 15:22
|
|
|
Carthag Tuek posted:i know, hence the quotes if you're planging, transforming the S3 key into a pathlike object and then checking for the basename is pretty quick way to do this, should be o(n) edit: i assume you're already faking it with unix dir separators so you should be good there, remember to paginate over the s3 response since they block that off to 1000 objects at a time Share Bear fucked around with this message at 16:11 on Mar 26, 2021 |
|
#
?
Mar 26, 2021 16:07
|
|
|
Carthag Tuek posted:i know, hence the quotes What if you just used s3fs and did that
|
|
#
?
Mar 26, 2021 16:10
|
|
|
thx but i got it.. made the list of empty prefixes per my prev post and just iterated over those while deleting & everything is nice now 
|
|
#
?
Mar 26, 2021 18:55
|
|
|
Anyone here ever used Postman? I feel like it might the the singular worst program I've ever used. Takes 15 seconds to start up what is essentially just a wrapper around javascript fetch api, each panel in the app has tabs nested within tabs nested within tabs, and somehow they forgot to make it possible to right click -> copy location on urls
|
|
#
?
Mar 26, 2021 18:56
|
|
|
postman sucks rear end lol i'm in the market for a reasonable graphical web rest client if anyone has one. zero fancy things. basically looking for httpie with a ui. support for pasting in an HAR file copied from browser dev tools a plus. just need mac support for work
|
|
#
?
Mar 26, 2021 19:07
|
|
|
abraham linksys posted:postman sucks rear end lol https://github.com/pashky/restclient.el
|
|
#
?
Mar 26, 2021 19:17
|
|
|
abraham linksys posted:postman sucks rear end lol https://insomnia.rest it's pretty good
|
|
#
?
Mar 26, 2021 19:21
|
|
|
just man postman man
|
|
#
?
Mar 26, 2021 20:10
|
|
|
Blinkz0rz posted:https://insomnia.rest yeah I've liked using this more than postman, though it's fundamentally the same thing, just a little bit less poo poo
|
|
#
?
Mar 26, 2021 20:49
|
|
|
Insomnia is like an even worse version of a Git GUI. Just grow up and use cURL like a professional.
|
|
#
?
Mar 26, 2021 20:51
|
|
|
does insomnia mean youll be up all night trying to get it to work
|
|
#
?
Mar 26, 2021 20:55
|
|
|
pokeyman posted:you're supposed to feed the output back in until you reach a fixed point So what you're saying is eventually all data will be dogshit? 
|
|
#
?
Mar 26, 2021 21:04
|
|
|
|
| # ? Apr 20, 2024 03:16 |
|
|
*looks at data* well that wont take so many passthrus
|
|
#
?
Mar 26, 2021 21:10
|
|