20250627

No change since 20250616 except that configure.ac now requires libc++-17 (was libc++-14). There are currently two places where the STL function transform_reduce() is used, and this requires version 17 at least.

20250616

Minor change for compatibility with FLINT-3.3.0

20250530

Added various utility functions and functions to find elliptic curves with fixed j-invariant and conductor or primes of bad reduction.

20250122

Only trivial changes compared with 20241112:

• minor reversion to avoid 32-bit test eror
• rename one variable to avoid clang complaint
• explicitly include gmp.h before flint.h.

20241112

• vector/matrix/subspace templates now used for multiprecision integer versions mvector, mmatrix, msubspace instead of these being coded separately.
• vec and mat template classes now based on vector for scalar=int, long, bigint
• improve precision handling for special L-value computations
• added some utilities
• improvements to Compute_Euler()
• many (mostly minor) changes suggested by cppcheck
• switched more C-style arrays to STL vectors
• much code simplification using newer C++ idioms e.g. auto

20240408

Fixed a bug in curve pointers in Point; small change in putput to h1bsd and h1bsdcurisog; more use of STL idioms and auto, removing almost all explicit iterators.

20231212

Minor update since 20231211, removing use of std::ptr_fun

20231211

• Better use of FLINT when FLINT3 is available
• Elliptic curve torsion computation no longer uses floating point

20230424

Implement class bigcomplex for multiprecision complex floating point based on NTL's RR, instead of using the template comlex.

No changes in functionality.

20221012

Enhanced cubic forms code, including bug fixes in reduction.