Hacker News: pjtr

New comment by pjtr in "Why is F# code robust and reliable?"

pjtr — Tue, 24 Sep 2024 06:02:21 +0000

Is "let-over-lambda" (variable capture / closure) not possible in F#? Is that not a form of implicit dependency injection?

New comment by pjtr in "Why is F# code robust and reliable?"

pjtr — Tue, 24 Sep 2024 06:00:01 +0000

This seems fine within a file, but isn't this problematic across files? Now the file order is significant, so not only is the Visual Studio XML project file is an essential part of the language semantics, you also can't organize your files in subdirectories freely? Or did they fix that at some point? How does that scale to larger projects?

New comment by pjtr in "JAX – NumPy on the CPU, GPU, and TPU"

pjtr — Fri, 29 Sep 2023 05:31:15 +0000

Is there a specific reason why Windows is not supported?

New comment by pjtr in "Why did InterlockedIncrement/Decrement only return result’s sign? (2004)"

pjtr — Mon, 13 Jun 2022 06:29:55 +0000

I don't think Microsoft invented it:

> On the CDC CYBER, an interlock register is available ... Operation of this interlock register is similar to TEST AND SET

Concurrency in Operating Systems, October 1976 https://www.computer.org/csdl/magazine/co/1976/10/01647182/1...

New comment by pjtr in "OBS Studio: Open-source software for video recording and live streaming"

pjtr — Thu, 02 Apr 2020 17:06:31 +0000

OBS is very good software, thanks for working on it!

Recording the screen works much better than any other screen recording software I tried. For this use case the preview can be a bit confusing. If the resolution does not match (because of OS 200% scaling for example) going to the settings each time to adjust it is a bit cumbersome; the interactive resizing handles in the preview somehow never helped me. Sometimes one of the reset zoom context menus helps.

Also the "Window source" would be awesome, but is a bit cumbersome to set up every time, and doesn't capture things like menus unfortunately.

It's probably really difficult to improve these things, so they work automagically for dummies like me that know very little about OBS and use only a tiny feature set, without taking making things worse for power users, which are probably happy with things as they are.

New comment by pjtr in "Terrain rendering algorithm in less than 20 lines of code"

pjtr — Fri, 03 Jan 2020 18:16:47 +0000

A short article titled "Voxel Landscapes 2D and 3D By Scout/C-Lous" described most of the tricks, but I can't find the text anywhere. A similar article by the same author on another topic I found here https://github.com/ggnkua/Atari_ST_Sources/blob/master/Docs/...

New comment by pjtr in "Terrain rendering algorithm in less than 20 lines of code"

pjtr — Fri, 03 Jan 2020 17:58:27 +0000

The one I remember was "voxel.pas by Steven H Don". A Delphi port survived here: http://tothpaul.free.fr/sources.php?dprgrp.voxel

Searching for "voxel" in https://github.com/nickelsworth/swag/ also finds some very similar tiny Turbo Pascal programs.

New comment by pjtr in "Terrain rendering algorithm in less than 20 lines of code"

pjtr — Fri, 03 Jan 2020 17:50:26 +0000

Even drawing front-to-back you don't need the y-buffer, if you switch the loops: for x ... for y ...

Another trick I now remember was to interpolate the color values on such a y-segment, to reduce the pixelated look.

New comment by pjtr in "Terrain rendering algorithm in less than 20 lines of code"

pjtr — Fri, 03 Jan 2020 13:29:57 +0000

Same with leaning. Add one more line of code and tweak the draw call:

    var ylean = (input.leftright*(i/screenwidth-0.5) + 0.5) * screendata.canvas.height / 4;
    DrawVerticalLine(i, heightonscreen+ylean, hiddeny[i]+ylean, map.color[mapoffset]);

It adds a lot to the "feeling" IMO :)

New comment by pjtr in "Terrain rendering algorithm in less than 20 lines of code"

pjtr — Fri, 03 Jan 2020 12:27:45 +0000

There were also tricks to extend this simple rendering algorithm to allow limited rotations around the other two axes, to look up / down slightly (just move everything up / down; also implemented in the demo here) and to "lean" when steering left / right (just move everything up / down proportional to the distance from the center of the screen; not implemented in the demo here, but visible in the 1992 NovaLogic Comanche example GIF).

There were Turbo Pascal versions of this on websites in the 90s I think, but it seems they were lost.

New comment by pjtr in "How to undo almost anything with Git (2015)"

pjtr — Tue, 24 Dec 2019 07:55:57 +0000

You are correct, thanks.

New comment by pjtr in "How to undo almost anything with Git (2015)"

pjtr — Tue, 24 Dec 2019 07:16:56 +0000

purge is a third-party extension; it removes untracked files as you wrote.

But ~purge~ prune is also a command of the evolve extension, that is conceptually similar to strip, as GP wrote.

New comment by pjtr in "How to undo almost anything with Git (2015)"

pjtr — Tue, 24 Dec 2019 07:11:25 +0000

Fortunately "transplant" is just an old extension, so you can just forget entirely about it.

"Graft" copies; "Rebase" moves.

Alternatively just forget about graft also, and use "rebase --keep" to copy.

"Rebase with evolve" is conceptually still rebase.

Mercurial has strip (removes changesets from repository, by default stores a backup). Mercurial does not have purge.

Evolve has ~purge~ prune (marks changesets as obsolete).

Evolve is similar to a git reflog.

Evolve stores more contextual information than the reflog, so Mercurial+Evolve it is safer and easier to undo things than in Git+reflog. Mercurial also pushes some of this contextual information, so even collaborative history rewriting is possible in a safe and easy way, unlike git.

New comment by pjtr in "ECMAScript proposal: String.prototype.replaceAll"

pjtr — Sun, 22 Dec 2019 10:34:24 +0000

I don't understand the reasons. What is the actual problem in this "even-odd problem"?

How do all the other languages that have an equivalent function (Perl quotemeta, PHP preg_quote, Python re.escape, Ruby: Regexp.escape, Java Pattern.quote, C# Regex.Escape, Go QuoteMeta, Rust regex::escape) avoid or deal with this problem?

New comment by pjtr in "Unicode Is Awesome"

pjtr — Tue, 03 Dec 2019 06:58:12 +0000

And WORD 16 bits: https://docs.microsoft.com/en-us/openspecs/windows_protocols...

(QWORD 64 bits)

New comment by pjtr in "More MS-DOS Games Playable at the Internet Archive"

pjtr — Mon, 14 Oct 2019 06:01:59 +0000

Targhan maybe?

https://archive.org/details/msdos_Targhan_1989

New comment by pjtr in ".NET Core gRPC"

pjtr — Tue, 24 Sep 2019 06:32:37 +0000

Are there options for machine local IPC over pipes in .NET Core?

New comment by pjtr in "String Lengths in Unicode"

pjtr — Mon, 09 Sep 2019 06:00:12 +0000

"Python 3’s approach is unambiguously the worst one"

No rationale for this seems to be included in the article.

New comment by pjtr in "A Beginner’s Introduction to Python Web Frameworks (2018)"

pjtr — Wed, 21 Aug 2019 05:06:39 +0000

TurboGears 2 https://turbogears.org/

New comment by pjtr in "Fork: A fast and friendly Git client for Mac and Windows"

pjtr — Wed, 24 Jul 2019 07:44:27 +0000

I did see the screenshots. That's what triggered my comment. I did not see any branch names. Now I see "nh-scroll-fix" and "origin/fix-simulate-na..." in the conflict dialog, but not in the merge dialog. Branch names seem insufficient anyway though, but it would be a slight improvement I guess.