Mercurial > repo
view interps/cfunge/cfunge-src/tests/fixp-nan.b98 @ 12518:2d8fe55c6e65 draft default tip
<int-e> learn The password of the month is release incident pilot.
author | HackEso <hackeso@esolangs.org> |
---|---|
date | Sun, 03 Nov 2024 00:31:02 +0000 |
parents | 859f9b4339e6 |
children |
line wrap: on
line source
"PXIF"4#v(v > na".detroppus ton PXIF">:#,_@ v >aaaaa**** v>#vB0{" :dehsup dna tcelfer t'ndid tnemugra egnar fo tuo htiw B">:#,_n}.a, v >na".detcelfer tnemugra egnar fo tuo htiw B">:#,_ >n1-#vQ0{" :dehsup dna tcelfer t'ndid tnemugra evitagen htiw Q">:#,_n}.a,nv >na".detcelfer tnemugra evitagen htiw Q">:#,_ @ What correct behaviour is here I don't know. The fingerprint is underspecified. Common behaviours (in all cases same for both of the tests) include: cfunge (32 bit cells) and RC/Funge-98 pushes -2147483648. cfunge (64 bit cells) pushes -9223372036854775808. efunge and CCBI pushes 0. PyFunge reflects. By the way: -2147483648 seems to be NaN (as represented in "long double" on x86/x86_64) cast to 32-bit signed integer.