Programming thread

  • Want to keep track of this thread?
    Accounts can bookmark posts, watch threads for updates, and jump back to where you stopped reading.
    Create account
brother tinfoilman said:
His point is that the "next generation" won't care about source control because it's just another opaque and agent-driven system. They won't even know it's happening under the hood. Git chads will continue to feast and rule the Earth, but do you trust Microshit (Github) or the G*rmans (Codeberg) with your source code? How about Drew "gooner? I hardly knew her!" DeVault (SourceHut)?

The actual lesson this faggot youtuber should be teaching is that setting up your own git takes 20 minutes on a cheap VPS.
Click to expand...
Cgit enjoyers unite.
 
Whatevermancer said:
I thought that if you disable GC in D, a lot of the standard library becomes unavailable cause it relies on it.
Click to expand...
The answer to this is sort of, the standard library does use the GC but you can override it temporarily but I think it's only really practical to do that with arena allocators but I think arenas are very flexible so it works out (mostly).

There's @nogc which is fairly limiting, the better option is to keep it enabled and either don't use it or replace it when you need to, there's also a compiler flag to show any potential uses of it so it's fairly easy to manage. I do think they should make more of the standard library nogc though, I think I saw something they made recently where they said the emphasis on GC was a mistake but not sure if they're planning on changing anything.

Rango Dango said:
unless im using the different definition of templates it does allow you to have templated functions and structs https://odin-lang.org/docs/overview/#parametric-polymorphism

this guy has some great videos on odin
https://youtube.com/watch?v=X2Fy2zcfRhM
Click to expand...
To me the idea of templates is more than just generics but I'm gonna be honest I'm not sure if there's a proper definition or what it is. D lets you do static if on anything known at compile time in basically any context (global scope, function scope, struct/class declarations, etc.) to conditionally include code and it's great, honestly don't know why this isn't in every compiled language at this point.

There's quite a lot of compile time functionality in it and I really dislike the whole "purity" thing over avoiding this in Odin. The creator says shit like "do you really need this?" and it's like yeah nigga I don't want to write more code when I have the alternative of not doing that, the problem is not templates, it's bad implementations of templates. Templates being implemented poorly in the past is not a reason to just decide that we should never have innovations in language design ever again, it's an insanely retarded argument.
 
Enne said:
It's okay, has some nice things (defer
Click to expand...

Oh, I forgot about defer, yeah. Zig was the first language I've played with that had that and I was really glad to see Odin had the same when I started to look at it. That's definitely one of the little things I miss when I have to return to webslopland for the day job.

I'll take another look at D. I remember looking at it before but I don't remember why I passed it over.

brother tinfoilman said:
The actual lesson this faggot youtuber should be teaching is that setting up your own git takes 20 minutes on a cheap VPS.
Click to expand...

The value in GitHub and its clones isn't hosting the code so much as it is all the other tools around it, most especially issues but also automated test runners, notifications and API stuff, Pages, easy user and access management, and so on. You can't set up self-hosted tools to handle that stuff so easily.

I've looked at Fossil a few times, a DCVS from the SQLite team that does have a lot of that stuff built into itself, but the network effect of Git is too strong to get me to switch even for my own little personal projects. Plus the way you have to manually set up a separate data directory outside of the repo rather than just doing the equivalent of a .git directory feels stodgy.
 
I resumed tweaking some obscure and completely impractical numerical code for demonstration purposes. I spent all day, and it works for everything except a handful of powers of two. It's not even really the primitive I needed for something else, so I've given up. I'll resume work on a better interface to the same behaviour within the month.
:bossmanjack: :bossmanjack: :bossmanjack:
 
I need to vent. How hard is it to limit your changes in any code versioning system to one specific problem or implementation? I am tired of shooting down 400+ file change pull requests that touch every corner of the solution. People can't be bothered to actually read the diffs when they create merge conflicts. This is why people are being replaced by AI, holy shit.
 
analrapist said:
SWI-Prolog is the package that's best documented and most accessible for learners. There are other Prologs for other specific uses, but SWI is both the most generalist and the easiest to learn about.
Click to expand...
Do you know any good guides/tutorials for it? Are the ones on their website any good? Ik it’s probably more relevant to GNU Prolog, but I’ve had good experience with GNU programming manuals before and I know they have one for Prolog. Is it any good?
 
August Levasseur said:
Do you know any good guides/tutorials for it? Are the ones on their website any good?
Click to expand...
SWI's documentation is fantastic. From a quick glance, as I'm not as familiar with GNU's docs, SWI's is perhaps more comprehensive. Both look like they do a decent job as an independent tutorial. Given your general levels of programming fluency, I'd reckon both or either are more than adequate for you as a tutorial, though perhaps you want other sources. When I'm doing SWI dev, I rarely need to look elsewhere. I'm not sure what a good tutorial would be as I learned Prolog first at university as part of a course.
 
Netizennameless said:
I need to vent. How hard is it to limit your changes in any code versioning system to one specific problem or implementation? I am tired of shooting down 400+ file change pull requests that touch every corner of the solution. People can't be bothered to actually read the diffs when they create merge conflicts. This is why people are being replaced by AI, holy shit.
Click to expand...
Now that programming is dEmOCRaTiZeD it takes a rare state of mind to care about the contents of a commit much less actually read a diff. And it takes a lot of beatings to change that in everyone else. Most of them think "coding" is change file -> let editor fuck with all the line endings and indent types -> git add . -> git commit -m 'update'. Better off rejecting pull requests out-of-hand when the solution is 50% whitespace changes.
 
Oh yeah, i just remembered one of the things that annoyed me about that last primeagen video. He acting like how many stars a project has on github was some huge deal. Like the linux kernel vs this ai tools github stars. For the kernel that isn't even their actual repo they literally just mirror the actual repo to github.

Besides that, that way of looking at things I personally just don't like. Judging importance based on github stars? I don't know if I can articulate properly why it's so repulsive to me, but it is. It's definitely shallow, but that's definitely not the whole reason.

Nitro! said:
Now that programming is dEmOCRaTiZeD it takes a rare state of mind to care about the contents of a commit much less actually read a diff. And it takes a lot of beatings to change that in everyone else. Most of them think "coding" is change file -> let editor fuck with all the line endings and indent types -> git add . -> git commit -m 'update'. Better off rejecting pull requests out-of-hand when the solution is 50% whitespace changes.
Click to expand...

I will be sending you 20 pull requests that just amount to me autoformatting the file with my editor shortly, and changing code for "readability", I expect to get see these merged in by the end of the day.
 
prollyanotherlurker said:
Oh yeah, i just remembered one of the things that annoyed me about that last primeagen video. He acting like how many stars a project has on github was some huge deal.
Click to expand...
Its huge in the FAANG community. A few people I know were seriously considering paying for those github-stars-as-a-service things to have something to put on their resume, luckily I convinced them fraud is not the best strategy. I also remember giving a few recommendations to a friend of mine, and he rejected one because it only had 120 stars, unfortunately he is web-dev and so I think its a permanent brain injury.

As an aside I never really like the primeagen's videos. I'm not trying to pull the "I am a based contrarian" take. But every one of his videos where he reviews tools/IDEs seems to entirely come from the point of view of the most web-dev pilled, style over substance takes possible. I normally just ignore these retards, but the primeagen unfortunately poisons many others into ignoring good tools because they don't fit the "brand image."
 
General Zigger said:
Its huge in the FAANG community. A few people I know were seriously considering paying for those github-stars-as-a-service things to have something to put on their resume, luckily I convinced them fraud is not the best strategy. I also remember giving a few recommendations to a friend of mine, and he rejected one because it only had 120 stars, unfortunately he is web-dev and so I think its a permanent brain injury.

As an aside I never really like the primeagen's videos. I'm not trying to pull the "I am a based contrarian" take. But every one of his videos where he reviews tools/IDEs seems to entirely come from the point of view of the most web-dev pilled, style over substance takes possible. I normally just ignore these retards, but the primeagen unfortunately poisons many others into ignoring good tools because they don't fit the "brand image."
Click to expand...
I don't think I've ever seen any of his videos where he talks about an ide. But I'm a vim user and I've never really had any interest in learning about any gui ide. The only experiences I've had with them put me off them them pretty much immediately. (They were slow, very busy and distracting compared to what Im used to, I didn't like it)
 
Last edited:
prollyanotherlurker said:
I will be sending you 20 pull requests that just amount to me autoformatting the file with my editor shortly, and changing code for "readability", I expect to get see these merged in by the end of the day.
Click to expand...
I don't have a sticker for this feeling.

laptop-slam.gif
 
prollyanotherlurker said:
Besides that, that way of looking at things I personally just don't like. Judging importance based on github stars? I don't know if I can articulate properly why it's so repulsive to me, but it is. It's definitely shallow, but that's definitely not the whole reason.
Click to expand...
I subconsciously judge the value and quality of a library or program in a GitHub repository by some first glance metrics, without looking at the code itself.
Here is a list of green and red flags that pattern recognition has instilled into me:
  • Star amount between 100 and 999 (not overhyped slop but not useless either)
  • 100+ commits
  • At the root file listing, many old files (5+ years old) and a few newer ones, indicating stability and being maintained with bug fixes
  • At the root file listing, files associated with more stable languages/protocols (C, Makefile, json)
  • At the root file listing, files associated with faster-changing dependencies (Python, cmake, yaml)
  • Arbitrary logo, images, badges or emojis in the README
  • Inconsistent file/folder naming
  • AI slop
  • Indian contributors
  • COC, or a gay autistic license
 
You must log in or register to reply here.
Back
Top Bottom