Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Search - "invalid url"
-
God damn fucking shit.
Now I know again why I don't do apps.
This is a app as simple as can be:
Enter a link, click a button, do a http request, download a file.
BUT FUCKING HELL WHY ARE YOU SO FUCKING RETARDED ANDROID?!
I'm not familiar with java but i don't care why is this so freaking unintiutiv to get shit done? Why are there thousands of ways and none works or atleast at a easy way? Make an object for this, make an object for that...
THIS IS RETARDED.
In PHP a simple "file_get_contents" would do the job. I were even down for some curl shenanigans if it were an easy implementation. BUT GOD DAMN.
URL url = new URL("http://fuckinghardcoded.com")
Oh no can't compile because that MIGHT be an invalid URL. Ok try catch this or just tell the rest of the Programm to watch out for this bad boy cause he might throw a MalformedURLException.
Ditch that and try volley. Everything is document except how to fire that queue! Does it do that by itself? Do I really have to do an override to a function while declaring? CMON ON I'M A WEBDEV IS THIS TRYING TO DO A FUCKING CALLBACK AND IS THIS TRYING TO BE AN ANONYMOUS FUNCTION??? Why is this so frustrating and confusing? I'm also mad at myself this is dropdead simple shit but I can't get it to work. Fuck this, fuck java , fuck android and fuck myself10 -
So my brother went back to school today. Now, during the 5 years I was there they had the most shit security on their IT systems, but aparently now they have fucked up their ssl. If you try to load the https page it comes up with the warning saying its an invalid certificate, but once you click it, it doesn't even load the school website, it loads this random page. Clicking on the buttons then take you to a page under their domain provided by another school. Going to this schools website, the https seems to be broken in the exact same way. It wouldnt be so bad, but it can confuse the hell out of people who type https before a url, and thos who dont realise and end up on the insecure site will need to provide passwords over an insecure connection. I am so glad im out of that place, they had such crap IT and everything was so easy to break.1
-
Oh great, now garbage websites can even break the browser's reload button. Wanna reload the page that never stops loading or crashes its scripts? Click the button and watch the browser do fuck all and a little bit of nothing.
Nah, you see to refresh that polished rotten turd you gotta close the tab and then restore it, and if that doesn't work open a new one and copy-paste the URL because who the fuck doesn't like some invalid state.
Can't wait for the day I have to shutdown and disconnect the battery because some idiot messed up the layout.3 -
I started building an application for FIDAL (Italian Federation of Athletics) because why not: I was bored and wanted to learn Flutter.
There is no API, but I didn't even expect it. Parsing the HTML is easy enough.
BUT OH MY GOD THE ENTIRE WEBSITE IS SHIT. Take this page: http://www.fidal.it/graduatorie.php, it uses some useless jQuery plugin and uses a buttload of JavaScript that isn't even needed. BUT WAIT. Try entering an invalid "club code" (http://fidal.it/graduatorie.php/...), a FUCKING white page with 200, are you kidding?
I'd also like to mention that all pages that require form input won't load correctly if you don't include "submit=Invia" in the URL.
I am not giving up.3 -
I'm back from the dead to rant again. This time it's punycode.
My job has to do with processing the commoncrawl web archives, and for some reason one in 20.000.000 archived webpages crashed my program. After some debugging I found this issue that seems to be the reason my code crashes https://github.com/servo/rust-url/...
To summarize the issue: Since punycode unicode characters can be encoded into domain names. But not every character is allowed. Not only do these invalid domains get registered, I need an in-depth knowledge about unicode to understand what is wrong here.
How did we turn domain names into something so complicated?3 -
So today we renamed a repo on bitbucket. We changed the remote url on local PCs and kept working. When deploying, our deploying platform threw an error saying invalid repo name, which was expected. Thing is, said platform doesn't have a "change repo remote url" option, so we did it manually over SSH. It didn't work as it now says the bitbucket token is invalid. There is no option to change or set the token. Redeploying will take almost an entire day due to configurations. FML.1
-
!dev
https://epicgames.com/fortnite/...
This url brings to error page with invalid email address that will actually autoreply "We're writing to let you know that the group you tried to contact (do-bug) may not exist, or you may not have permission to post messages to the group. A few more details on why you weren't able to post"
Okay, sure. This may happen to indie stuff etc.
But we are talking about bigass company thats fucking trying to assasinate steam (and so far kindda fails). You dont want to show customers error messages, sure, but at least, if you do it, and you tell your customer to send email to admin, make fucking sure that you provide an actually fucking working email. Is it so hard?
No wonder they cant conquer steam. And thats positive side3 -
Wow, angular is still a pile of shit in 2024, nothing changed.
I renew my https://devrant.com/rants/7582990 previous rant
I've recently switched to angular 17, not because I'm a masochist, but because, unfortunately, we have a huge portal for a super huge multinational enterprise and it's made in angular.
It's 2 years worth of work, and they've suddenly decided it's cool to switch to angular 17, because standards, because it's new etc.
Now that this crap angular 17 came out I prepared my hair pulling room, where there are whips and self torture instruments, and I've typed into browser url they "super new super modern super efficient" angular.dev, which apparently is their new official super 1337 documentation site (spoiler, it's shit as the other if not worse).
Since they realized angular was pigshit, they decided to eviscerate it like a sacrifical lamb in ancient maya age and add lot of stuff that makes it modern and more friendly.
They think they made the big bang of news, but they implemented stuff that exist since 10 years after people were cutting their wrists in their github "request a feature" section for years.
Well, to make it brief, they made a whole clunky obscure way to bootstrap it and didn't even had the decency and modesty to properly document it (they never learn, sigh....)
In any case I put up a .NET minimal API that works well, and a small angular app with a Hello world page that fetches a "hello word" string from a test api route.
The api works everywhere, browser, postman etc etc.
But ta-dahhhh, in angular throws error.
They put various way of using http client. Main 2 are withFetch() and without.
withFetch() says "as error "Invalid self signed certificate" and withoutFetch "Unknown error".
Apparently we have to do shenanigans also to do some dev development3 -
I don't get keycloak. Anyone who has experience with it, please help.
We have what I would think is a common setup: a kubernetes cluster with a Spring boot api-gateway and keycloak as oauth2-provider.
The api-gateway needs an issuer-uri to keycloak for endpoint discovery, i.e. to configure a bunch of endpoints to keykloak for different purposes.
The two main purposes are: 1. to redirect the user to keycloak (must be an url reachable from outside the cluster, i.e. ingress) 2. to authenticate tokens directly with keycloak from within the cluster.
Keycloak can be configured to set some of these discovery endpoints to different values. Specifically it makes a separation between backfacing (system calls in cluster) and frontfacing (user call from browser) urls All seems good.
However, when using this setup, each time spring security authenticates a token against keycloak it says the "issuer" is invalid. This is because the issuer is the host on which the token was generated. This host was the one in the url which the user was redirected to i.e. the ingress.
It feels like there is no way around this except running keycloak outside the Kubernetes cluster, but surely there must be a way to run keycloak in the same cluster. What else is the purpose of keycloak having the concept of back- and frontfacing urls?1