[GiNaC-devel] Release 1.6.3

Hi everybody,

Alexei and I have prepared a new release and are happy to announce that
GiNaC 1.6.3 is out and available. These are the changes:

* New expand_options::expand_transcendental. GiNaC will apply additional
  rules when expanding expressions if this option is specified, i.e.
  log(a*b) -> log(a) + log(b)
  exp(a + b) -> exp(a)*exp(b)
* Functions can define a custom expand() and info() now.
* Automatically rewrite (x^{-1})^c -> x^{-c} for positive c.
* Transform abs(x)^n -> x^n if x is real and n is even.
* Improved power::info(): power with a real base and even exponent
  reports that it's non-negative.
* Fixed spurious match failure due to expairseq::match() bug:
  > match(sin(y)*exp(b)+sin(x)*exp(a), sin($0)*exp(a)+exp(b)*sin($1))
  FAIL
* Fixed power::is_polynomial() and mul::is_polynomial() computation.
* factor_univariate correctly handles polynomials over rationals (no
  segfaults any more).
* Fixed an infinite loop in factor_multivariate due to incorrect
  handling of polynomials having a leading coefficient being a sum
  with a non-trivial integer content.
* Fixed numeric evaluation of generalized multiple polylogarithms for
  real and negative argument.
* Avoid calling log(1-x) in numeric evaluation of classical
  polylogarithm.
* Fixed numeric evaluation of multiple polylogarithms (correctly set
  the signs of the imaginary parts in the intermediate expression).
* Fixed numeric evaluation of Nielsen polylogs for arguments close to
  the 6th root of unity or its conjugate.
* Avoid infinite recursion in series expansion of logarithm.
* numeric::info(info_flags::nonnegative) correctly handles complex
  numbers with non-zero imaginary part.
* Fixed several build issues:
  - libreadline and dlopen() misdetection.
  - multiple definition of symbol on MinGW.
  - auto* tools and texinfo being unhappy in various ways.
* GiNaC can be built with CMake now.

Regards,
   -richy.
-- 
Richard B. Kreckel
<http://in.terlu.de/~kreckel/>