profile
viewpoint
If you are wondering where the data of this site comes from, please visit https://api.github.com/users/Artoria2e5/events. GitMemory does not store any data, but only uses NGINX to cache data for a period of time. The idea behind GitMemory is simply to give users a better reading experience.

AOSC-Dev/fontconfig 3

Experimental Fontconfig branch with Michal Srb's optimisation work.

AOSC-Dev/chntpw 2

Chntpw fork that can operate on Microsoft accounts

Artoria2e5/163player 0

Command-line Netease (163) music player with graphical lyrics implemented in pure Bash

Artoria2e5/996.ICU 0

Repo for counting stars and contributing. Press F to pay respect to glorious developers.

Artoria2e5/acbs 0

A re-implementation of the original ABBS utilities.

Artoria2e5/AFLplusplus 0

The fuzzer afl++ is afl with community patches, qemu 5.1 upgrade, collision-free coverage, enhanced laf-intel & redqueen, AFLfast++ power schedules, MOpt mutators, unicorn_mode, and a lot more!

PR opened xiaolitongxue666/xiaolitongxue666.github.io

base.scss: add fallback color before image loads

背景还没加载的时候默认是白色,什么都看不到。挑一个相近的颜色吧。

(其实是应该 var(--background-color),但是那个 #000 纯黑不是很近……我又不想改两个文件)

+1 -1

0 comment

1 changed file

pr created time in a day

push eventArtoria2e5/xiaolitongxue666.github.io

Mingye Wang

commit sha 18f9fdf8e42d891d6e530aecd5a0064a335fee6a

base.scss: add fallback color before image loads 背景还没加载的时候默认是白色,什么都看不到。挑一个相近的颜色吧。 (其实是应该 var(--background-color),但是那个 #000 纯黑不是很近……我又不想改两个文件)

view details

push time in a day

issue commentguillaumepotier/gettext.js

Wrong (or not so accurate) documentation

It's in #50.

cnaaq

comment created time in 2 days

issue commentcea-hpc/wi4mpi

Compilation fails with GCC 11

This might have to do with the linker too -- I'm guessing the developer's environment might be using a more conservative distro using ld.bfd instead of ld.gold.

Passing allow-multiple-definition should silence it regardless of linker version. I know having multiple definitions is bad, but I can't figure out where it got defined a second time for now, let alone in somewhere as odd as asm_jit.o. This library itself is full of symbol magic already...

vchuravy

comment created time in 3 days

PR opened cea-hpc/wi4mpi

README: make env var list a list

The list currently renders as a paragraph without line breaks. Let's fix that.

Also link the see man thing.

+18 -16

0 comment

1 changed file

pr created time in 3 days

push eventArtoria2e5/wi4mpi

Mingye Wang

commit sha 0ee623f0c10d19e75c5bf39218228b71152ac325

README: make env var list a list The list currently renders as a paragraph without line breaks. Let's fix that. Also link the see man thing.

view details

push time in 3 days

fork Artoria2e5/wi4mpi

Wrapper interface for MPI

fork in 3 days

startedCaffeineMC/lithium-fabric

started time in 5 days

pull request commentitorr/homo

修复 1e50 产生 x.yyyyyye+zzz 错误触发小数检测的问题

咦好像网页没更新?

Artoria2e5

comment created time in 5 days

issue closedtommyettinger/anim8-gdx

Separate formats from dithering?

I am a bit interested in getting a somewhat simple description of the cool new "scatter" algorithm, but the way it's duplicated in AnimatedGif and PNG8 is... not nice, I guess. Would it make more sense to make a shared function with a signature somewhat like the following to do all the dither instead?

public byte[] dither(int[] image, int width, PaletteReducer pr, DitherAlgorithm da) {
  // ...
  return indexedPixels;
}

closed time in 6 days

Artoria2e5

issue commenttommyettinger/anim8-gdx

Separate formats from dithering?

Hmmm, they are actually in PaletteReducer already. I guess it's a case of inlining? Okay.

Artoria2e5

comment created time in 6 days

issue openedtommyettinger/anim8-gdx

Separate formats from dithering?

I am a bit interested in getting a somewhat simple description of the cool new "scatter" algorithm, but the way it's duplicated in AnimatedGif and PNG8 is... not nice, I guess. Would it make more sense to make a shared function with a signature somewhat like the following to do all the dither instead?

public class DitherPass {
    public byte[] indexedPixels;

created time in 6 days

issue commentkilljoy1221/HD-Font-Generator

Font Fallback Stack?

Likely -- that's what a Stack is, right?

JapanYoshi

comment created time in 7 days

issue closedQuerz/mcaselector

Feature request: section-level operations for 1.18 merges

The upcoming 1.18 version provides extra underground space, and I am one of these people very interested in shoehorning the new underground into my existing chunks by splicing at y=15 or something to get the extra deep space.

As a result, I am going to need copy-and-paste on a more granular level than the chunk -- the 16x16x16 sections in the chunk file format might be just the level I need.

closed time in 7 days

Artoria2e5

issue commentQuerz/mcaselector

Feature request: section-level operations for 1.18 merges

So it does regenerate empty sections! Awesome.

Artoria2e5

comment created time in 7 days

startedRoadrunnerWMC/libimagequant-python

started time in 8 days

PR opened RosettaCommons/RoseTTAFold

README: more format improvements
  • Set sh highlighting on code blocks to de-highlight... lowlight(???) the comments
  • Indent code blocks so they are treated as part of the list
  • Use inline code for filenames

[The README changes are getting boring, I know...]

+41 -42

0 comment

1 changed file

pr created time in 8 days

create barnchArtoria2e5/RoseTTAFold

branch : Artoria2e5-patch-1

created branch time in 8 days

fork Artoria2e5/RoseTTAFold

This package contains deep learning models and related scripts for RoseTTAFold

fork in 8 days

issue openedQuerz/mcaselector

Feature request: section-level operations for 1.18 merges

The upcoming 1.18 version provides extra underground space, and I am one of these people very interested in shoehorning the new underground into my existing chunks by splicing at y=15 or something to get the extra deep space.

As a result, I am going to need copy-and-paste on a more granular level than the chunk -- the 16x16x16 sections in the chunk file format might be just the level I need.

created time in 8 days

issue commentQuerz/mcaselector

Filter unmodified chunks

Even with the same seed ore and stuff are generated differently, since a separate RNG and seed source is used there. I believe the InhabitedTime filter mostly meets your use case.

judebert

comment created time in 8 days

pull request commentRosettaCommons/RoseTTAFold

Update README.md

You can further improve the formatting by indenting the fenced code blocks. This way they should be treated as part of the numbered list.

runitralph

comment created time in 9 days

create barnchArtoria2e5/Pillow

branch : cache

created branch time in 11 days

fork Artoria2e5/Pillow

The friendly PIL fork (Python Imaging Library)

https://python-pillow.org

fork in 11 days

issue commentpython-pillow/Pillow

quantize() to a palette often chooses unideal color matches

A simple but far-from-optimal way to "fix" the cache would be extending the ImagingPaletteInstance.cache member to a tagged union allowing multiple values. Sparse cells can keep their one-value encoding, while the multi-value ones can just use a linear search -- 64 entries to look through at most, so nothing fancy there I think. [Ideally the whole cache becomes some proper search tree, but that hurts my head. Ayy, double space use can't be that bad, right?]

(Sorry for the previous message -- I didn't even do a getpalette to check it out.)

ericpauley

comment created time in 11 days

PR opened ImageOptim/libimagequant

colordifference_stdc: use float as in SSE

The SSE implementation of the function uses single-precision float, whereas this one goes for.... double all over the place.

Extremely unscientific comparisons on godbolt (https://gcc.godbolt.org/z/oa3hP5ffs) shows that both Clang and GCC do much better generating x64 code when float is used. Other SIMD systems should act similarly.

+4 -4

0 comment

1 changed file

pr created time in 11 days

push eventArtoria2e5/libimagequant

Mingye Wang

commit sha 6aba29f14cd3ddf2d04f13077d9e8b04028e559d

colordifference_stdc: use float as in SSE The SSE implementation of the function uses single-precision float, whereas this one goes for.... double all over the place. Extremely unscientific comparisons on godbolt (https://gcc.godbolt.org/z/oa3hP5ffs) shows that both Clang and GCC do much better generating code when float is used.

view details

push time in 11 days

fork Artoria2e5/libimagequant

Palette quantization library that powers pngquant and other PNG optimizers

https://pngquant.org/lib

fork in 11 days

issue commentpython-pillow/Pillow

quantize() to a palette often chooses unideal color matches

I am seeing something a bit more odd from modifying @radarhere's example, to the point that I think we may have something else going on:

from PIL import Image
i = Image.new('RGB', (1,1), (0, 25, 255))
p = Image.new('P', (1,1))
i2 = i.quantize(palette=p)
print(i2.convert('RGB').load()[0, 0])  # (85, 85, 85)

It's somehow deciding it should use a grey ramp -- what's that about? (And it happens with the original putpalette too?!)

ericpauley

comment created time in 11 days