dzaima
1e53078324
make match functions take ux instead of u64 for length
2025-05-23 00:47:04 +03:00
dzaima
2a57cb30b8
fix eequal on different NaN patterns
...
+ rename compareFloat→floatCompare to be more consistent
2025-05-23 00:42:07 +03:00
dzaima
a7897ddcde
get rid of NEEQUAL_NEGZERO
...
as compatibility checking now relies on eequal, it can no longer do funky things
2025-05-23 00:23:13 +03:00
dzaima
b0877bfad9
use reflexive match for (2‿3⥊0÷0)⌾(0‿0⊸⊏) 2‿3⥊1 fix
2025-05-22 23:58:04 +03:00
dzaima
d1855f14e9
refactor match function direct calling
2025-05-22 23:56:02 +03:00
dzaima
aab1a15c39
better eequal impl
2025-05-22 23:45:41 +03:00
dzaima
7947f86833
basic fix for (2‿3⥊0÷0)⌾(0‿0⊸⊏) 2‿3⥊1 erroring
2025-05-22 23:45:37 +03:00
dzaima
81ce43e77f
complete comment
2025-05-22 02:28:50 +03:00
dzaima
36ba160477
fix 𝕨/𝕩 doing unchecked multiplications on +´𝕨
...
both that explicit one, and ones in the allocation functions
2025-05-22 01:45:56 +03:00
dzaima
3b5497e329
fix use-after-free in a⌾(i⊸/)𝕩
...
not actually ever problematic I believe as `w` in *_ucw always happens to have other held references, but still a bug and could become problematic in the future;
and the change is actually an improvement, bringing the SGetU computation out of paths that don't need it!
2025-05-22 01:18:04 +03:00
dzaima
04d0191d1f
handle v¨⌾(l⊸/)x with non-boolean l with fast path
2025-05-22 01:14:09 +03:00
dzaima
185a18d478
fix a⌾(b⊸/)x comparing during noAlloc
2025-05-22 00:38:47 +03:00
dzaima
15bf932757
make atom decomposition equality check use eequal
...
gets rid of compound functions having observable identity if they have NaN as a component
2025-05-21 00:54:52 +03:00
dzaima
723ca38816
make native ⌾(is⊸⊏) & ⌾(m⊸/) compare NaNs as equal
2025-05-21 00:36:08 +03:00
dzaima
c17280d96d
cleanup
2025-05-15 00:43:53 +03:00
dzaima
b9ccafcf3f
any_squeeze → squeeze_any
2025-05-13 17:04:38 +03:00
dzaima
f1cb141f56
new squeeze functions
2025-05-13 17:04:38 +03:00
dzaima
9d11fc0cde
use C1 & C2 helpers in grade.h
...
incl. making them capable of expanding a macro F
2025-05-13 17:04:38 +03:00
dzaima
ff406c9586
better int_squeeze_sorted
2025-05-13 17:04:38 +03:00
dzaima
8146040c15
helpers for creating empty arrays
2025-05-13 17:04:32 +03:00
dzaima
6c4eae5e28
rearrange squeeze.c and make num_squeeze & chr_squeeze more sane
2025-05-13 03:32:32 +03:00
dzaima
62366d10fa
•internal.Validate
2025-05-13 00:12:59 +03:00
dzaima
41c9a51ff3
fix unshare on heap-allocated atoms
2025-05-11 20:28:23 +03:00
dzaima
516205321c
fix error message of 𝔽⌾(a⊸⊏)𝕩 more
2025-05-09 01:17:51 +03:00
dzaima
f74bdf6142
improve/fix error messages around units in ⊏ & ⊑ some more
2025-05-09 01:00:41 +03:00
dzaima
8d757363e9
improve error message of 𝔽 in 𝔽⌾(a⊸⊏) returning an atom
2025-05-09 00:32:54 +03:00
dzaima
d2c27fd880
fast 𝔽⌾(num⊸⊏)arr
2025-05-09 00:21:42 +03:00
dzaima
1ecc6b3918
more proper error message for 1⌾((<⋈0)⊸⊑)↕5
2025-05-05 19:37:02 +03:00
dzaima
13906efe44
fix ⊑˘ on rank>2 inputs
2025-05-02 01:28:46 +03:00
dzaima
e32d41eb61
improve •ParseFloat error message on argument with codepoints ≥256
2025-05-01 23:09:48 +03:00
dzaima
f9b3aba234
fix wrong integer types in varargs
...
problematic in 32-bit builds
2025-04-24 00:00:32 +03:00
dzaima
e1b62b95b6
fancy Singeli accumulator; use for integer sum
2025-03-21 04:32:26 +02:00
dzaima
108379f121
fix UB shift
2025-03-16 23:47:01 +02:00
dzaima
f1021580ad
silence warning on non-Singeli builds
2025-03-16 20:22:17 +02:00
dzaima
82ebd37b3f
group statistics test
2025-03-13 23:36:02 +02:00
Marshall Lochbaum
233c88120d
Singeli function for initial group statistics pass since auto-vectorization isn't great
2025-03-13 14:21:38 -04:00
dzaima
72c4eca6c0
fix float ⌾(l⊸⊏) doing wrong comparisons
2025-03-12 17:23:17 +02:00
dzaima
62b095be89
oneElt⊏𝕩 improvements & fixes
2025-03-12 06:29:26 +02:00
dzaima
3ae7d134d2
more uniform F¨ fill behavior
2025-03-12 05:52:52 +02:00
dzaima
bfce527096
fix 𝕗⌾(-2⋆63)𝕩
2025-03-12 03:01:11 +02:00
dzaima
a6202c8ca6
fix error message of ⌾(oob⊸⊏)
2025-03-12 01:41:52 +02:00
dzaima
0848faf8c7
don't leak memory in TEST_RANGE
2025-03-12 01:14:24 +02:00
Marshall Lochbaum
dcecc70c25
More careful fill handling with characters in insert_scal
2025-03-10 21:58:15 -04:00
dzaima
97e0210c41
remove unnecessary F˜˘ acceptance from F˘
...
scan_rows_bit only checks frtid, which is 0 for such
2025-03-11 00:17:11 +02:00
dzaima
bcf32838f4
some more ˝ tests, use COPY_TO
...
not much of a difference, but prettier
2025-03-11 00:15:45 +02:00
Marshall Lochbaum
f1972fc408
Faster stride-1 check
2025-03-10 12:57:25 -04:00
Marshall Lochbaum
e42273bfe8
Generic stride-1 𝔽`˘ implementation
2025-03-10 12:50:54 -04:00
Marshall Lochbaum
bbe4549aff
Scalar implementation of 𝔽˝ for generic arithmetic 𝔽 and short rows
2025-03-10 12:48:06 -04:00
Marshall Lochbaum
e898a81d68
Use a possibly-overlapping word and constant-length loop in k/bool, like ⊣`
2025-03-09 08:44:42 -04:00
Marshall Lochbaum
302c2f926f
AVX2 k/bool for odd 64<k<256 using shift-based masks
2025-03-09 08:44:42 -04:00