TRIP 1.3.0 release announcement

[gast...@imcce.fr]

It is with great pleasure that I am able to announce the major release of TRIP version 1.3.0

You may download it from http://www.imcce.fr/trip/

Here's a list of the new features available in TRIP 1.3.0 :

The enhancements are :
- New numerical matrices  with double, quadruple and multiple precision real or complex floating-point numbers.
  new  keywords: matrixR and matrixC (see the tutorial http://www.imcce.fr/trip/doc/tutorialnumericalmatrix.pdf)
- New functions : inf, sup : get the lower and upper bounds of the arrays, numerical vectors or matrices.
- The functions sum and integral use the compensated summation if the option is set.
- The FFT quadruple-precision is supported.
- The command stat reports in the statistics the minimal and maximal degree of each variable if the option is set.
- The loop for supports the OpenMP statement introduces by /*!trip omp parallel for */
  New OpenMP statements will be introduced in the next release.
  
The bug fixes are :
- The format %f was fixed in quadruple precision floating-point numbers.
- The format %d in the function str and writes may produce an overflow.
- The function read with an format specifier supports now the letter D for the exponent delimiter.
- The function read with an format specifier hangs when it reports an error.
- The function readbin hangs when a vector of complex number is supplied.
- A regression was fixed for the selection of elements in an array of vectors.
- The function integnum supports array of variables 
  and returns the output of the numerical integration in an array of numerical vectors.
- The function writes hangs in some cases.
- The function det hangs if the argument is an empty function.

Mickael Gastineau