Nenad Rakocevic
Nenad Rakocevic
1. That's expected, because they share the same underlying R/S runtime code. 2. Try using a `[cdecl]` in the R/S function passed to `redRoutine`.
I also think that it was a mistake to allow `absolute` on `char!` as it makes no sense. The `mod` issue should be resolved in a different way.
> on-deep-change* is useless unless complete Most of the use-cases in View don't require an accurate `index` or `part` value (which are most of the issues you've reported), so it...
@hiiamboris Could you also provide the set up code (the full owner object)?
> react?/target only works after is, not after react That's how it's intended, the documentation should be improved to reflect that.
I don't get what you mean.
@hiiamboris On ARM (RPi), it gives: ``` true true true true 0 1 10 ``` It seems the NaN comparisons handling is wrong at: https://github.com/red/red/blob/master/system/targets/ARM.r#L1931 Moreover, it seems the method...
> Does `system/tests/source/units/float-test.reds` succeed for you? I don't have access to @rebolek's Pi anymore. It does pass. I re-tested your code above and I've found the source of the wrong...
> Shows that indeed these opcodes should return false given any NaN argument, That's the part I've missed as I wrongly assumed that they would return TRUE even on a...