|
pokeyman posted:the whole awful.app posting flow needs reworking, it is janky as gently caress tbf its an app on a phone, so it sucks but its less painful than website which is a plus. mad props etc but i cant figure out how to zoom a posted picture without "unpinching". or are goons just posting links instead now
|
#
?
Apr 20, 2024 09:29
|
|
|
|
| # ? May 3, 2024 12:59 |
|
|
images don’t zoom for me reliably either. have to do the firm tap just so
|
|
#
?
Apr 20, 2024 14:17
|
|
|
have to hold longer than you think. no, longer than that if this baby shower this afternoon doesn’t go forever I plan to try to figure that out again (and hopefully not get lost in VisionKit or whatever again)
|
|
#
?
Apr 20, 2024 14:22
|
|
|
Carthag Tuek posted:"pregenerated_posts" is a bit clunky. any better ideas? pregenerated_posts is an excellent name for a property containing pregenerated posts
|
|
#
?
Apr 21, 2024 21:03
|
|
|
"oh no having descriptive names is cringy, like enterprise java framework class names"
|
|
#
?
Apr 21, 2024 21:04
|
|
|
Wheany posted:"oh no having descriptive names is cringy, like enterprise java framework class names" I feel like the designers of rust believe this unironically
|
|
#
?
Apr 21, 2024 21:33
|
|
|
The idiomatic Go variable name would be p
|
|
#
?
Apr 21, 2024 21:53
|
|
|
The Fool posted:I feel like the designers of rust believe this unironically they're just metalang weenies. ml was an academic project by real mathematicians so obviously their variable names are absolute dog poo poo
|
|
#
?
Apr 21, 2024 22:49
|
|
|
bob dobbs is dead posted:they're just metalang weenies. ml was an academic project by real mathematicians so obviously their variable names are absolute dog poo poo turns out the metal in metalang was iron
|
|
#
?
Apr 22, 2024 00:09
|
|
|
lmao the UK bit of the company that I work for merged with an acquired US company and one of the questions was "is there going to be any pay alignment between UK and the US since our devs are now split half and half and work together" the answer was, verbatim, "that ain't never gonna happen" I mean fair enough but it's pretty funny, at least to me
|
|
#
?
Apr 23, 2024 17:03
|
|
|
there was another question, if anyone can move to the US the answer was "no"
|
|
#
?
Apr 23, 2024 17:07
|
|
|
sounds like it might if they said “ain’t never”.
|
|
#
?
Apr 23, 2024 17:08
|
|
|
Using double negatives around programmers seems like a dangerous game to play, tbh.
|
|
#
?
Apr 23, 2024 17:15
|
|
|
i think they were trying to sound jovial about it
|
|
#
?
Apr 23, 2024 17:16
|
|
|
they’ll equalize UK and US pay…to Philippines rates
|
|
#
?
Apr 23, 2024 17:17
|
|
|
Or just cherry pick a region of US & UK, like Fargo and Bangor.
|
|
#
?
Apr 25, 2024 13:11
|
|
|
The Fool posted:awful app does this on the last word if there is s "pending" correction in ios google hangouts on ios used to apply autocorrect to the last word when you hit "send" (same behavior as imessage) but they changed it in chat so now all my hangouts messages end with typos
|
|
#
?
Apr 25, 2024 13:28
|
|
|
Plank Walker posted:but they changed it in chat so now all my hangouts messages end with typos better than sending with a word you didn't intend
|
|
#
?
Apr 25, 2024 13:30
|
|
|
yeah... sort of... like the keyboard will show the correction highlighted and every other app behavior is to replace it when you hit space or return, so i got used to relying on that
|
|
#
?
Apr 25, 2024 13:40
|
|
|
https://www.tomshardware.com/tech-i...o-be-fraudulentquote:"We stand behind our performance claims and are excited for consumers to get their hands on Snapdragon X Elite and X Plus devices soon.” Absolutely nothing is going to change with fact checking or even remote cynicism on fantastical claims, at least it’s being tested post announcement: quote:After OEMs got their hands on the X Series chips, SemiAccurate claims no OEM has been able to replicate Qualcomm's touted test numbers. In fact, the site reports OEMs first saw numbers "far sub-50%" of the numbers Qualcomm claimed. It would be funny if Qualcomm are nitrogen cooling and overclocking beyond 6Ghz, but Occam’s razor clearly pointing at bad testing. For those not aware of the humble advertising, quote:Snapdragon X Plus features the state-of-the-art Qualcomm Oryon™ CPU, a custom-integrated processor that delivers up to 37% faster CPU performance compared to competitors, while consuming up to 54% less power https://www.qualcomm.com/news/releases/2024/04/qualcomm-continues-to-disrupt-the-pc-industry-with-the-addition- ie ARM64 running faster than Intel i9, Apple M3, and AMD Ryzen 9, with “days” of battery life.  More realistic numbers of 16%, of Snapdragon to M3, which is pretty much a rounding error and not great leap for almost 3 quarters trailing release. https://wccftech.com/snapdragon-x-elite-tested-in-surface-laptop-6-high-multi-core-score/ MrMoo fucked around with this message at 14:32 on Apr 25, 2024 |
|
#
?
Apr 25, 2024 13:43
|
|
|
I'm shocked that a processor manufacturer would cheat on benchmarks
|
|
#
?
Apr 25, 2024 18:38
|
|
|
What is the endgame for Qualcomm here? Did they hope that not a single independent reviewer on earth would run benchmarks themselves to verify their numbers???
|
|
#
?
Apr 25, 2024 18:48
|
|
|
they probably dont care. theres probably so many inter-company management promises going around that the only thing that matters is the thing launches by the deadline so everyone gets their bonuses. whether anyone buys the product doesnt matter, just so long as they hit their deadline.
|
|
#
?
Apr 25, 2024 23:01
|
|
|
lmao i wrote a half dozen small java class files in intellij then closed and reopened it and it had helpfully replaced the contents of all of them with one line of the form "package lmao;class SucksToBeYouIdiot {}" ayfkm with this poo poo? edit: yeah idk my install is hosed somehow, I hit Save and none of my changes appear in the filesystem even though intellij remembers them. Sapozhnik fucked around with this message at 23:25 on Apr 26, 2024 |
|
#
?
Apr 26, 2024 23:14
|
|
|
we have a web app. it's a pretty straightforward web store type of thing that we customize for different clients. there are no unit tests and no integration tests and naturally it's all tightly coupled spaghetti code. there have been some regressions with recent changes so my boss says we need tests. his solution: something like cypress or selenium running through the checkout process. when i asked if we were just testing the frontend the response was "no, the whole app". ok whatever, it might be difficult to mock things like payment processors and db connections out of the backend for testing... that's not what he means either. he wants us to run actual credit card transactions with a real card in automated tests. what about the charges? naturally we write another test for the refund page 
|
|
#
?
Apr 26, 2024 23:30
|
|
|
Does your processor not have test env + cards to use to test transaction flows??
|
|
#
?
Apr 26, 2024 23:36
|
|
|
we do for some stuff but there are other things that go through a third-party before they get to the real payment processor (long story that's just as dumb as everything else) and those don't work with anything but a real card
|
|
#
?
Apr 26, 2024 23:40
|
|
|
Joe Chip posted:we do for some stuff but there are other things that go through a third-party before they get to the real payment processor (long story that's just as dumb as everything else) and those don't work with anything but a real card “if” you can make those full coverage tests fast and reliable they are the best kind. there isn’t like a CAP theorem for tests that says you can’t have it all. still a big “if” tho
|
|
#
?
Apr 27, 2024 00:13
|
|
|
Joe Chip posted:we do for some stuff but there are other things that go through a third-party before they get to the real payment processor (long story that's just as dumb as everything else) and those don't work with anything but a real card ... does the third party not have a test environment?
|
|
#
?
Apr 27, 2024 00:57
|
|
|
i asked a similar question my first week here a few months ago. the "test" environment is actually their dev/staging server and not only is it frequently down it can behave completely different from day to day as changes are pushed. once i found out why all their api paths started with WS_* i was less surprised
|
|
#
?
Apr 27, 2024 02:48
|
|
|
Joe Chip posted:we have a web app. it's a pretty straightforward web store type of thing that we customize for different clients. there are no unit tests and no integration tests and naturally it's all tightly coupled spaghetti code. there have been some regressions with recent changes so my boss says we need tests. his solution: something like cypress or selenium running through the checkout process. when i asked if we were just testing the frontend the response was "no, the whole app". ok whatever, it might be difficult to mock things like payment processors and db connections out of the backend for testing... belize is looking pretty good these days, my friend
|
|
#
?
Apr 28, 2024 11:31
|
|
|
fwiw I think your boss has the right idea. if i had the option between choosing having unit tests or integration tests like the one your boss is proposing, i’ll take the integration ones every day of the week. we do something similar at my job, where we are every 10 minutes or so doing “black box” end to end (in so much as it is possible) testing of our product in production. our product going down could be hugely impactful to the integrity of the business, so we want an immediate phone call if it’s ever broken.
|
|
#
?
Apr 28, 2024 14:16
|
|
|
i’ve never done it at the level of UI though, because our UI being down is much less scary to the business than the core backend functionality being down. that may not be the case for you though
|
|
#
?
Apr 28, 2024 14:20
|
|
|
yeah we have alerts on spikes of traffic to our status page and support search/connections E2E testing is fine, and often necessary, as a backstop but pretty much every time one of them finds an issue I feel like it’s pointing out either a missing class of unit tests or a bad architectural decision that should be remedied
|
|
#
?
Apr 29, 2024 02:13
|
|
|
assuming you don’t want to test interactions with payment providers can’t you have a test only promotional code that gives 100% discount. Which would bypass needing to get paid. Make sure to not have that active in prod, or do 
|
|
#
?
Apr 29, 2024 13:22
|
|
|
Aren’t there specifically test credit card numbers for this very thing?
|
|
#
?
Apr 30, 2024 00:03
|
|
|
yeah, OP said that one of the systems they work with doesn’t support them (which is insane, I thought the card networks would crush you if you didn’t support that) time to find a big-boy supplier for whatever that poo poo is, not that OP probably gets a choice
|
|
#
?
Apr 30, 2024 00:28
|
|
|
my two questions if you did go with non-test e2e billing tests (with refunding) are 1) which cc are you testing with and 2) how’s that going to look to fraud departments?
|
|
#
?
Apr 30, 2024 01:16
|
|
|
I have trouble believing your processor doesn't have test numbers and customer data. Did your coworker tell you that or does the system really not have test numbers you can use. I'd suspect you talked to the wrong person/people
|
|
#
?
Apr 30, 2024 01:33
|
|
|
|
| # ? May 3, 2024 12:59 |
|
|
also just move to Shopify
|
|
#
?
Apr 30, 2024 01:34
|
|