Nicolas Boichat

Results 115 comments of Nicolas Boichat

Implement pow/pow_with_context (exponentiation with integer parameter)

Fixed the CI, updated margin logic to add multiplications/10 as extra margin (not that I'm fully convinced by the argument in #147...) Still some rare errors with this logic.

Raise to the power of an _integer_

Thanks! Going in a very different direction from what you're saying, but you made me think ,-P I first converted the function to the iterative version (from Wikipedia again ,-P)....

Raise to the power of an _integer_

Argh, something is wrong is my "proof", things can still go wrong (that is, the last digit can be different), when we end up on a fraction that's really code...

Raise to the power of an _integer_

(went away for a few days ,-)) Thinking about this a bit more, I agree it does appear to make sense to think in terms of bits, instead of digits,...

Raise to the power of an _integer_

Of course, by running the "fuzzer" for long enough I can still find bad examples (https://github.com/drinkcat/bigdecimal-rs/blob/pow_test_2/pow_test/src/main.rs) -- maybe 1 in a million runs... ``` Compute/compare -4172329935948502689779088291680609380122797322250695253563263937250989869257152712036874041648218112**379158740 100d 0123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789 good 86470343637125948448136380601729659949966946434276499998821385023022743721505439654781179021761139018861188336468850821984876567949079743066361544933970630233636304347127572456482862798516531791539586102717310413091103392280962568747164976365671994663786377738925512913604967306041390296342664992544228188175500676759699672701333688210922723963085573265023914046006633132367500955909913528603137437905755746561986532622280261164182478467219842355310069910375404127805415361287351374751928193739032318e+37771936745...

printf: trim leading whitespace when parsing numeric values

1. That change should actually change `printf` behaviour (`seq` currently uses totally different parsing code, I'm trying to unify them...). Not sure why you don't see any change, but you...

printf: trim leading whitespace when parsing numeric values

> > 1. That change should actually change `printf` behaviour (`seq` currently uses totally different parsing code, I'm trying to unify them...). Not sure why you don't see any change,...

printf: trim leading whitespace when parsing numeric values

Don't forget to run `cargo fmt` and `cargo clippy`. Otherwise, LGTM! (turns out I'll need this one to pass `seq` tests with the common parsing methods!)

tests/cp/preserve-gid: `cp -p` fails to preserve group

I think the problem is that `canonicalize` fails if the user doesn't have read access to the entirety of the path (which is what that `chroot` thing causes). This was...

Update MSRV to 1.88

redoxer build is not happy, even after update to 0.2.54 (https://github.com/drinkcat/coreutils/actions/runs/16870133030/job/47783201524): ``` error: rustc 1.86.0-nightly is not supported by the following packages: [email protected] requires rustc 1.88.0 [email protected] requires rustc 1.88.0...