I thought of this after a recent bug I found. I use Vivaldi browser and recently it updated. After the update my mouse cursor was not visible when within the browser window. Other programs worked fine. I tried visual studio and steam and epic game store all had my mouse, Vivaldi didn’t.
I closed all instances of Vivaldi via task nanager(was unable to click the x) and restarted it. That fixed the bug and I haven’t been able to replicate so I don’t have anything to submit for a bug report. Just a really strange thing.
What have been your weirdest bugs?
Not a bug I experienced but read about https://www.bleepingcomputer.com/news/microsoft/microsoft-fixes-word-bug-that-deleted-documents-when-saving/
A bug in a Microsoft product? Impossible!
Not sure if it fully fits, but one of my old HTC android phones has a bug where it would randomly fail any network request, by just never finishing it. No timeouts either.
It never got fixed. Loading a page in the browser, got stuck, and would need a new refresh. Some apps like Baconreader would get stuck doing the same, any app could have it happen. But sometimes the stars align, Facebooks Messenger had (has? don’t see it anymore) the chat bubbles that showed over whatever else on the screen. And i clicked one of those, and they got stuck loading. And there was literally no way to close it. There was no timeouts, so i eventually had to hold the power button and reboot…
Other notable mentions, Spotify on android did not handle duplicated files (since you could add your own music to it) with the same filename or at least artist - title metadata. It would always only play one of them. It was very interesting because Spotify on pc worked, so i could play the second track fine there, but if i told it to continue play through my phone then it would continue form the same timestamp on other track… Might be related to my phone being stupid, but there was a whole range of infuriating spotify issues. So i just moved to use poweramp and store and sync local music since i had a lot there to begin with .
Why would you have 2 files with the exact same name? Windows can’t even handle that if in the same folder and I would assume Spotify just dumps all files into a folder hence why it didn’t work.
They had different names like (1) if they were duplicates. But all my files were named something like “Artist - title.mp3”, and the Metadata tag for artist and title were generated based on the filename. I don’t know how they would look after being synced over to my phone from Spotify though, but it is probably how Spotify names the files that broke it on my phone.
All the files were in the same folder on my pc too, for reference. So they have unique filenames, but two files could have the exact same artist and song title. I give that’s it’s an obscure edge case, but really frustrating.
Why did you have multiple copies of the same song and artist? Only reason I can think of is like jam band concert versions but i would assume it would be like artist - song - year. Mp3 in that case
I have shitty music taste, and ended up downloading nightcore songs from YouTube. And the videos were usually named “Nightcore - song title”. So chances of duplicate songs end up a lot higher when the only unique thing is the title of the song…
Ah ok that makes sense
Chrome sometimes scrambles json like this. If I copy the scrambled json, it’s fine when I paste it
That looks unformatted and not scrambled. Last time I used Chromium it didn’t come with a JSON formatter so I had to install an extension for it.
The text is nonsense though. Look closer, some of the fields are jumbled.
Chrome does come with a formatter now.
For example, towards the end there isdItsopwith is supposed to bepostIdOh I see. That’s very random, lol. It’s completely backwards.
Maybe this Chrome view of the JSON encounters a right-to-left character?
My understanding is this is googles attempt to prevent you from accessing their stolen data
I don’t see why they would do that when it’s just an ordinary response from a random API: https://app.wafrn.net/api/v2/dashboard?level=2
I was thinking more scrapping off a browser, like using selenium. I haven’t checked in a while but previously almost every element on a Google page had randomly generated id/names so you couldn’t like automatically select the input box, type in a few terms and then click on the search button. If you got all the id’s and tried, on the next loading of the page all the id’s would be different.
It’s not a bug per se, and we’re gonna see a lot of this moving forward.
Did a code review, hefty code review, and as I got to the end of it where “the magic happened” I discovered that the logic was actually a case statement for every permutation.
Because they’d asked the AI to write logic to go through each type, the AI wrote a case statement for EACH TYPE.
… it was so fucking fast though. So wrong but so so fast compared to the actual running code.
deleted by creator
Fast to run or fast to make?
I’m a big fan of if it’s stupid but works it’s not stupid, especially if you get a runtime speed improvement.
Yeah it might be theoretically difficult to make an update but if by “each type” you mean like the primitive data types (string, int, char etc) I doubt the language is going to change how, for example, an int works. If you they do so many programs will break.
It ran super fast but only over that one constrained input, it didn’t allow for new use cases.
Ah ok, i was thinking you meant the previous solution was so complicated so the basic switch sped it up
The LLM basically just did the work once for the sample data set, so all the computation happened up front.
If that was the only data set we’d ever need, and the data set didn’t change realtime based on sensors then … hell yeah that gross switch woulda worked 😂
Ah ok… That makes sense. I was thinking like generate a stream of data, data comes in some format like (string dataType, float dataValue) and a switch like case int{do this} case string {do that} then get new data and repeat while(true).
If it only was calculating based on the test case yeah that’s bad.
I told this before but I had a project which had an xml configuration of many sections each targeting something different. I wanted each section as an object. This was a system of many combined systems like one dmm, one custom automated shield box, one bluetooth tester, etc. Each needed specific data to connect to it and configure it for the tests we wanted to run. This was all I needed in the xml file.
So I coded 1 function to read 1 section of the xml file and map it to an class object. I then copied that into Ai, copied configuration file into Ai, and told it to replicate it for each section. That worked. I think Ai can be used in specific circumstances where you are doing like repeated operations but it gets grossly overused. You also have to know the domain and the language to understand what it’s doing.
Another example is I was asked to make a sql query which was basically a join(serial number and test guid in one table, test guid and test results in another table and only given serial number). My boss sent me a list of like 150 serial numbers separated by spaces over teams. I wrote the query for the first one and told the Ai to replicate it on the other serial numbers. Yes I could of gone through the list, backspace the space, add common, press enter and repeat. I knew how the query worked. I know the domain I was working in. The Ai output was basically the same as my original just with more serial numbers in the where clause.
I worked at an IoT platform startup. All of our embedded device demos stopped working August 1st. I was told the same thing happened last year, but it was fine, things would start working in September. I decided to go fix it anyway. Eventually I figured out the culprit was a custom HTTP library. Instead of doing anything sensible, the way it found the Content-Length header was to loop over the bytes of the response until it found the first ‘g’ add 5 to that pointer and then assume that whatever was there was the number of bytes it should read. Unfortunately, HTTP responses have a Date header which includes the month and August has a ‘g’ in it.
There were a bunch of these demo devices already flashed and shipped out. The ‘fix’ to get them to work, even in August, was to downgrade requests to HTTP 0.9 which didn’t require a Date header in the response.
That’s “I wrote my first shell script” type of bad. I remember Ubuntu+GNOME used to be full of such crap in their early days. There are still some helper scripts around from those days (the naughts), you can still catch a whiff of it.
Is there any reason they couldn’t just look for the complete “Content-length” string?
Yeah, to be fair, there was an issue getting
string.hto work (so i could just usestrstr) with the vendor’s shitty toolchain, that took me talking to an engineer at the vendor, and the dev who wrote that was out of our Taiwan office. But also, my first fix was just doing a sort of sliding-window check, manually checing fors[0] == '\n' && s[1] == 'C' && s[2] == 'o' &&..., which was gross, but much more correct.Wait…not being able to use a basic library like string.h opens up so many other questions
🤷, embedded device manufacturers were really bad at software back then. I honestly don’t remember the details anymore.
The only words I don’t believe there are… Were and back then
I joined a team and was assigned a bug that had been bugging them for two years. Randomly files they saved would be corrupted.
Eventually isolated it to the third party library they used to serialize the data. For some reason this library corrupted file names that were an odd length. So “cat” would get corrupted while “cats” would save and load just fine. It was a black box library we didn’t control so was told to just program a workaround to check filename length and append a character if it was odd and move on.
I still want to know what that library was doing.
I still want to know what they library was doing.
Haunting
Notepad on windows, I think around the XP and earlier versions, had a similar bug. If you typed
this app can breakand saved it, the file would be corrupted. Something to do with the number and placement of characters.ASCII was interpreted as UTF because the function that checked whether the given text was Unicode checked the difference between bytes at even and odd positions. Many of the common phrases used to trigger this were in the 4-3-3-5 format (by letters), e.g., Bush hid the facts However, there was never any reason that this format of character placement was necessary for the bug (though even length was necessary)
Hmm yeah that’s really weird… I can’t think of a purpose for a check like that, plus even it was a valid reason throw an exception or return a result indicating failure, don’t just corrupt the file
I’m curious now. Can you describe “corrupted”? One char off? Non-ascii gibberish?
This was 15 years ago and I was only on that team for a year so some details are fuzzy. I just remember the odd length filename distinctly. We never found a pattern, not that we looked super hard. We did check the usual suspects like Base64 encoding and big vs little endian and such but nothing lined up. The file would just be garbled nonsense. No whitespace and every character under the sun in a massive block.
I recently tried Pop!OS liveUSB without a mouse connected to PC, there were some accessibility issues but one was unrelated to mouse and pretty cursed:
I chose system language to be something non-latin-based, and ended up with a keyboard layout that can’t input English text. There was no way to change the layout and adding the US layout didn’t work, because system showed that the US layout is already the one and only available. So I couldn’t input commands to set things up but also couldn’t click in the UI to set things up
Win32 MFC desktop application, should be able to run continuously unsupervised for days or weeks. When I log in via RDP it crashed. Yes, just in that moment. In the end I figured it out.
MFC supports multiple visual styles, such as classic Windows, few Office versions and few more. We used feature for loading (possibly translated) strings (resource) from various DLLs. First from a currently active plug-in DLL, if the string was not there, then from a sub-system DLL, if it was not there then from main EXE. Plus there was some logic about translation and fallback to English in case of missing or not up-to-date translation. The cause was in one of the MFC visual styles, it changed a global variable related to that string loading thing for a short duration of time, then changed it back. It was only in one of the styles and there was no need to change it, the code could just load the value and use it as a parameter to a function that needed it. Our code in a different thread failed to load a string and threw an exception. Cross a COM interface boundary. Every COM method is decorated by a bunch of macros our senior engineer didn’t understand and was not willing to learn. One of the macros was noexcept. Thus an exception being thrown cross an noexcept boundary crashed our app.
The main language of the app was English. The user could change it to German, French, Russian, Chinese, Spanish, etc. But sometimes the translation was not complete. Sometimes some plug-in was not translated. We used string IDs to identify various texts. Sometimes the ID pointed to the translated text, sometimes not, thus English version of that text was used instead.
Also I reported the bug in the visual style to Microsoft. They refused to fix it and said something like “don’t do that”.
But COM interfaces can’t pass exceptions, only HRESULTs. I don’t know how it was working in the first place if the setup was this janky.
Yes, that is correct. We were using C++ and COM and didn’t protect ourselves against accidental exception being thrown across the COM boundary. But it never happened. Except the one case CString::Load (or whatever the function name was) failed because of Office visual style had bug in it.
If you don’t mind answering, what was the primary language if not English?
deleted by creator
deleted by creator
deleted by creator
deleted by creator
We deployed a client software in a Docker container on Windows 10. It could not connect to the backend, even though we saw SYN packages originating from it.
So, we ran WireShark on the Windows host and saw that the SYN-ACK packages from the backend were arriving there, too, but no ACK came through to complete the TCP handshake.Eventually, we rolled out a network debugging container on that Windows host and then could see in the tcpdump, that the SYN-ACK packages, which arrived on the Windows host, just did not show up in the container. Hyper-V or something was quietly dropping them.
Other network connections were working fine, just the SYN-ACK from our backend triggered this.
Load 2TB of data into ram during startup. Prompt user credentials to continue. Whoops, we need to do cryptography on those credentials… Free the heap. Invalid creds force exits. Valid creds goes gray until those 2TB are loaded back into memory… From storage halfway across the country over the 3mbps link, the best available in BFE.
I use a 3D CAD software that once you switch to 3D mode, it makes things partially transparent, but it does so by scaling the opacity of the entire window. The end result is that transparent parts look correct when you have a black desktop background, but otherwise you’ll see what’s on the window behind it or the desktop background. Same issue when you get gaps between two sides or curves. I think no one noticed because they would usually launch it from a terminal, so they always had black backgrounds.
Early in the GNOME /wayland transition, there was a bug I hit where you could drag windows off the edge of the screen, and somehow the desktop would scroll. It was kinda cool conceptually, but completely broken functionally, as it was hard to scroll back.
Edit: Wayland was actually the solution, Xorg was the problem:
https://programming.dev/pictrs/image/bc70285d-88ac-42aa-80be-b8b8f012547b.png
I hit a bug recently in KDE Wayland where the task bar was just slightly offset from the edge of the screen, so there was a gap behind it. Very dumb.
OPAM (OCaml’s package manager) had a bug where it couldn’t find
curlorwgetto download stuff with (don’t ask me why it shelled out to those in the first place) if you were in more than 32 Unix groups. Have fun thinking of a reasonable explanation for that!I’ll hazard a guess… OPAM was using a text-based utility to get the list of groups, and 33 is the point where the list would wrap to a new line.
Opam doesn’t endorse polyamory past a certain threshold
Why are you in 32 unix groups?
Can’t remember tbh. It was a work setup. It wasn’t an unreasonable one though. 32 is not very many!
some people like to party.
Well the venga boys are coming
Once upon a time I worked with a CMS that allowed an admin to delete the CMS itself and also any web server that ran it. Poof: gone. Fun times.
to be real evil, create a test to confirm what that button does cause you know…100% code coverage
