[Cryptography Team] PrimesFinder issue

[Chris Muller]

Wow, I did not expect to get such good mileage from that email!  :-)  I
loaded it and the issue appears to be fixed.  Thank you very much, Rob,
that was a nice interaction.

Best,
  Chris

On Wed, Jan 4, 2023 at 9:29 PM rabbit <rabbit at callistohouse.org> wrote:

> I should add it made me absolutely cringe to touch Luciano's code in fear,
> but for the tests...
>
> Luciano, whattup B. You still skatin in Shanghai? Dojay, dojay! Love to
> have you back!
>
> *•* rabbit
>
>
> On 1/4/23 22:20, rabbit wrote:
>
> Hey Chris,
>
> Thanks much for your investigations. I took a look and it is an issue with
> PrimesWheel>>#+= where the satisfied variable is counter-logic.
>
> I defaulted to false, set to true when a Zero is found in the residue
> (such that the pertinent prime candidate residue is added with the provided
> term (+ 2) modulo the prime candidate is ZERO).
>
> I merged with your CryptographyRandom change to SecureRandom.
>
> *••• **Config Maps **•••*
>
> I updated the following config maps:
>
>    - *ProCrypto.release.2*
>    - *ProCrypto.release.3*
>    - *ESqueak.release.1*
>
> *••• Packages* *•••*
>
> *••• **CryptographyRandom **•••*
>
> Name: CryptographyRandom-rabbt.24
>
> Author: rabbt
> Time: 4 January 2023, 10:06:57.026992 pm
> UUID: c9bd4f75-e822-4b25-96e3-7ff0f0734067
> Ancestors: CryptographyRandom-rabbt.23
>
> extended to #*=
>
> Name: CryptographyRandom-rabbt.23
> Author: rabbt
>
> Fixed PrimesWheel to satify isZero for the residue and allowed for growth
> of the wheel when the primes exceeds the current primes array.
>
> *••• **CryptographyRandomTests **•••*
>
> Name: CryptographyRandomTests-rabbt.2
>
> Author: rabbt
> Time: 4 January 2023, 10:03:17.773586 pm
> UUID: 1f0afa87-8c19-4ba2-9653-e982886dc8b4
> Ancestors: CryptographyRandomTests-rww.1
>
> Added CryptoPrimesTest, testing PrimesFinder, next after 11 is 13 and
> Wheel growth and exceeding the base 2000.
>
> Fixed PrimesWheel to satify isZero for the residue and allowed for growth
> of the wheel when the primes exceeds the current primes array.
> On 1/3/23 20:04, Chris Muller wrote:
>
> I just committed a fix to CryptographyRandom, and while working on it
> noticed another problem which I don't know how to fix.
>
>     (PrimesFinder candidate: 11) nextPrime  "2003"
>
> Obviously, it should be 13.  The issue appears to be with
> PrimesWheel>>#+=, original code since 2002, but I can't quite follow what
> or why the method is trying to accomplish.  "residues"?
>
>  - Chris
>
> _______________________________________________
> Cryptography mailing listCryptography at lists.squeakfoundation.orghttp://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography
>
> --
> •••
> 𝙄𝙛 𝙮𝙤𝙪 𝙖𝙧𝙚 𝙙𝙧𝙞𝙫𝙞𝙣𝙜 𝙖 𝙋𝙤𝙧𝙨𝙘𝙝𝙚, 𝙩𝙝𝙖𝙣𝙠 𝙮𝙤𝙪 𝙛𝙤𝙧
> 𝙢𝙤𝙫𝙞𝙣𝙜 𝙤𝙫𝙚𝙧, 𝙨𝙤 𝙩𝙝𝙖𝙩 𝙄 𝙘𝙤𝙪𝙡𝙙 𝙨𝙖𝙛𝙚𝙡𝙮 𝙥𝙖𝙨𝙨!
> 𝘼𝙧𝙧𝙞𝙫𝙚𝙙𝙚𝙧𝙘𝙞, 𝙧𝙖𝙗𝙗𝙞𝙩 • 𝘿𝙖𝙩𝙨𝙪𝙣 𝟮𝟰𝟬𝙕 • 🐰
>
>
> _______________________________________________
> Cryptography mailing listCryptography at lists.squeakfoundation.orghttp://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography
>
> --
> •••
> 𝙄𝙛 𝙮𝙤𝙪 𝙖𝙧𝙚 𝙙𝙧𝙞𝙫𝙞𝙣𝙜 𝙖 𝙋𝙤𝙧𝙨𝙘𝙝𝙚, 𝙩𝙝𝙖𝙣𝙠 𝙮𝙤𝙪 𝙛𝙤𝙧
> 𝙢𝙤𝙫𝙞𝙣𝙜 𝙤𝙫𝙚𝙧, 𝙨𝙤 𝙩𝙝𝙖𝙩 𝙄 𝙘𝙤𝙪𝙡𝙙 𝙨𝙖𝙛𝙚𝙡𝙮 𝙥𝙖𝙨𝙨!
> 𝘼𝙧𝙧𝙞𝙫𝙚𝙙𝙚𝙧𝙘𝙞, 𝙧𝙖𝙗𝙗𝙞𝙩 • 𝘿𝙖𝙩𝙨𝙪𝙣 𝟮𝟰𝟬𝙕 • 🐰
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/cryptography/attachments/20230105/582b2143/attachment-0001.html>