[Cuis-dev] Bug in Integer>>#raisedTo:modulo:

[Juan Vuletich]

Hi Folks,

Thanks Luciano and Nicolas for finding the problem and the solution.

I just pushed an update to GitHub. Just removed senders _and_ 
implementors of #\\\. That method doesn't carry its own weight. I also 
wrote a test with Luciano's example.

Cheers,

On 6/5/2023 2:17 PM, Luciano Notarfrancesco via Cuis-dev wrote:
> Makes sense, thank you!
>
> On Mon, 5 Jun 2023 at 19:11 Nicolas Cellier 
> <nicolas.cellier.aka.nice at gmail.com 
> <mailto:nicolas.cellier.aka.nice at gmail.com>> wrote:
>
>     I replaced ALL sends, not just this one, because it's putting too
>     much responsibility for senders for a very weak gain
>
>     Le lun. 5 juin 2023 à 19:09, Luciano Notarfrancesco
>     <luchiano at gmail.com <mailto:luchiano at gmail.com>> a écrit :
>
>         Ah, good to know, you replaced all calls to \\\ with \\ in all
>         methods or only this one? We should do the same in Cuis.
>         Still, the randomness mystery remains…
>
>         On Mon, 5 Jun 2023 at 19:04 Nicolas Cellier via Cuis-dev
>         <cuis-dev at lists.cuis.st <mailto:cuis-dev at lists.cuis.st>> wrote:
>
>             Ah OK... I expunged the sends of \\\ operator from Squeak.
>             This is exactly like \\ but omits normalization of
>             LargeInteger result...
>             Below is the comment that i let in \\\ definition:
>
>
>             "A modulo method former used in DSA."
>
>             "Notes: this method used to be a faster than \\ for
>             LargeIntegers, but this advantage is fainting:
>             - it always was slower for SmallInteger because of the
>             indirection below
>             - a new LargeInteger primitive makes \\ faster up to 64
>             bits operands
>             - even above 64 bits, its advantage has become marginal
>             thanks to revised \\ primitive fallback code
>             Moreover, \\\ behaviour is questionable for these reasons:
>             - for a negative receiver xor argument, it behaves like
>             rem: for LargeInteger and \\ for SmallInteger
>             - it may answer a not normalized LargeInteger (with
>             leading null digits) which breaks some invariants
>             For example, check (SmallInteger maxVal + 1 \\\ 8) isZero.
>             So beware if you ever think using this method."
>
>
>             Le lun. 5 juin 2023 à 18:56, Luciano Notarfrancesco via
>             Cuis-dev <cuis-dev at lists.cuis.st
>             <mailto:cuis-dev at lists.cuis.st>> a écrit :
>
>                 Doing
>                     pow := pow * pow \\ m
>                 instead of
>                     pow := pow * pow \\\ m
>                 at the beginning of the loop in
>                 Integer>>#slidingLeftRightRaisedTo:modulo: seems to
>                 fix it. But still I’m not sure why it is random
>                 otherwise, and I’m not sure that this really fixes the
>                 problem.
>
>
>                 On Mon, 5 Jun 2023 at 18:42 Luciano Notarfrancesco
>                 <luchiano at gmail.com <mailto:luchiano at gmail.com>> wrote:
>
>                     The following:
>                         5445648384 raisedTo: 3 modulo: 1 << 34
>                     Returns 0, which is right. But
>                         5445648384 raisedTo: 4 modulo: 1 << 34
>                     Returns different numbers every time I call it (it
>                     should return 0).
>                     It’s very weird… I don’t see where the randomness
>                     is coming from, perhaps from LargeIntegerPlugin?
>
>
>                 -- 
>                 Cuis-dev mailing list
>                 Cuis-dev at lists.cuis.st <mailto:Cuis-dev at lists.cuis.st>
>                 https://lists.cuis.st/mailman/listinfo/cuis-dev
>
>             -- 
>             Cuis-dev mailing list
>             Cuis-dev at lists.cuis.st <mailto:Cuis-dev at lists.cuis.st>
>             https://lists.cuis.st/mailman/listinfo/cuis-dev
>


-- 
Juan Vuletich
cuis.st
github.com/jvuletich
researchgate.net/profile/Juan-Vuletich
independent.academia.edu/JuanVuletich
patents.justia.com/inventor/juan-manuel-vuletich
linkedin.com/in/juan-vuletich-75611b3
twitter.com/JuanVuletich

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cuis.st/mailman/archives/cuis-dev/attachments/20230606/7106d153/attachment.htm>