[libmath-prime-util-perl] annotated tag upstream/0.59 created (now ed3959b)

[Salvatore Bonaccorso]

This is an automated email from the git hooks/post-receive script.

carnil pushed a change to annotated tag upstream/0.59
in repository libmath-prime-util-perl.

        at  ed3959b   (tag)
   tagging  ea498dbd7c69b3de062e2b7268a036ee9a077112 (commit)
  replaces  upstream/0.58
 tagged by  Salvatore Bonaccorso
        on  Thu Aug 4 07:58:38 2016 +0200

- Log -----------------------------------------------------------------
Upstream version 0.59

Dana Jacobsen (92):
      Fix bad Ramanujan prime count table entry
      Compiler warnings
      Add test for RT 110888
      Merge branch 'master' of https://github.com/danaj/Math-Prime-Util
      vecsum: 64-bit inputs resulting in 128-bit results handled internally
      New addmod, mulmod, divmod, powmod functions
      Tests and docs for addmod, mulmod, divmod, powmod
      add sqrtmod
      Tests, doc, and cleanup for sqrtmod
      Workaround for bugs in bigint packages
      Add is_euler_pseudoprime; let addmod/mulmod/divmod/powmod take neg first args
      register function in release test
      Add composite modulus support to sqrtmod
      znlog updates; prev_prime(2) and nth_prime(0) return undef; is_quasi_carmichael
      Allocation changes for cluster sieve
      Work around bugs in older Math::BigInt
      is_primitive_root for non-primes; better tables for sum_primes
      Add is_primitive_root; lucasu/lucasv return bigints
      Update AKS to Bernstein 2003 Theorem 4.1
      Add is_quasi_carmichael
      Don't print debug info for big vecsums
      Initial hclassno
      hclassno updates, PP ramanujan_tau using it
      Work around broken Perl 5.6
      Uninit variable in new DLP code
      Faster sqrtmod, fix even composite modulus
      is_primitive_root in PP+GMP done with znorder and totient
      Speedup for znprimroot.  Add two from_digit functions.
      sumdigits base change.  from_digits rewrite.
      PP sumdigits
      initial sieve_range
      sieve_range tests, doc, perf, edge cases
      Remove docs for binary()
      Edge cases for sieve_range
      Fix ispower issue on some compilers with large negative input
      Update for release
      Negative overflow is different on old Perls
      Bump Devel::CheckLib version
      Comments for AKS.  Turn off PFGW by default for example gap verify
      Mertens less memory and small speedup
      Shortcuts for GMP bernreal and harmreal
      Verify short M is enough for all 64-bit n
      Fixes for quadmath
      Turn back on one threading test
      Restricted Perrin pseudoprimes; Montgomery for psp, esp, Perrin; uint128_t fix
      Update restricted Perrin: encapsulate Mont exp, add uncompiled signature code
      Add filtered is_perfect_cube/fifth/seventh predicates
      Add is_prime_power.  Speedups for power detection.
      Simple Ramanujan sum prototype.  Docs.
      Add ipow() internal function
      Updates for Perrin
      make-perrin-data.pl use is_prime_power
      Add logint
      Speedup PP forcomposites.  Export and doc logint
      Rewrite integer root function
      Add Ramanujan sum
      PP ramanujan sum.  Ramanujan Tau with non-bigints.
      Speedup for PP tests on slow machines
      32-bit unsigned type for some constants
      Fix fromdigits string error for large array inputs
      fix vecsum with 2 args that overflowed
      vecsum shift correctly on overflow with 32-bit Perl
      Update for latest Mersenne results
      Update for release
      is_euler_plumb_pseudoprime export issues
      Tests and PP version of euler_plumb
      update test for prev_prime returning undef instead of 0
      Remove comment
      Add rootint
      Perrin restrict: 0, 1 minimal restrict, 2 Adams/Shanks, 3 Arno/Grantham
      Redefine rootint like Pari's sqrtnint.  n >= 0, k > 0 or croak.
      jacobi variable should be int
      Fix is_primitive_root with n=0
      Call GMP is_prime_power and rootint if possible
      Redo the way we see if a GMP function is available from PP code
      foroddcomposites error: 2200488465591,2200488465591+100000 made evens
      Switch Perrin test from matrix exp to Adams/Shanks doubling rule.  3x faster.
      Move Montgomery math functions to separate file
      Use some macros to make Montgomery operations tidier
      Reorder some code in Perrin signature function
      Documentation for restricted Perrin tests
      Docs for perrin test
      Speedup for Catalan test
      Tests for Perrin restrictions; deepify small pseudoprime tests
      Don't call GMP perrin if we have restrictions.
      Move macro mont API to montmath.h
      Add PP Perrin restrictions; fix GMP ispower with big negative roots.
      Ask for GMP 0.40 or newer
      znprimroot with negative args
      Use GMP version number in XS-GMP calling decision
      Include znlog test
      Add surround_primes to TODO

Salvatore Bonaccorso (1):
      Imported Upstream version 0.59

-----------------------------------------------------------------------

No new revisions were added by this update.

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/pkg-perl/packages/libmath-prime-util-perl.git