(martin.wojtczyk.de)

631 points wojtczyk | 4 comments | 31 Aug 24 04:29 UTC | HN request time: 0s | source

Show context

rgovostes ◴[31 Aug 24 18:30 UTC] No.41410786[source]▶

A friend doing homework for a university assignment, circa Leopard or Snow Leopard, noticed that Calculator produced negative values when raising a negative number to an even power.

The bug turned out to be in CFNumber, in Core Foundation. CFNumber does a lot of fiddly stuff at the bit level for performance, and one of their optimizations for exponentiation was incorrect. Somehow it was never found by tests or due to buggy behaviors it created in other apps, but by someone clicking buttons and thinking critically about the output.

replies(5): >>41410865 #>>41411004 #>>41411294 #>>41411316 #>>41412437 #

FabHK ◴[31 Aug 24 18:41 UTC] No.41410865[source]▶

>>41410786 #

In Excel (and some other languages), unary minus has higher precedence than exponentiation, so that -x^2 = (-x)^2 = x^2. That can bite you (for example when calculating the normal probability density function “manually”).

replies(2): >>41413220 #>>41418382 #

secondcoming ◴[01 Sep 24 00:45 UTC] No.41413220[source]▶

>>41410865 #

Are there languages where this isn’t the case?

replies(3): >>41413237 #>>41413617 #>>41414823 #

1. grumpyprole ◴[01 Sep 24 06:42 UTC] No.41414823{3}[source]▶

>>41413220 #

Standard mathematical notation?

replies(1): >>41416317 #

2. cubefox ◴[01 Sep 24 12:13 UTC] No.41416317[source]▶

>>41414823 (TP) #

It's arguably ambiguous.