Prev Next whats_new_2017

@(@\newcommand{\B}[1]{ {\bf #1} } \newcommand{\R}[1]{ {\rm #1} }@)@
Cppad Swig Changes During 2017

02-16
Advance to cppad-20170226 This avoids a warning about WIN32 during the run_cmake.sh command (while getting current version of Cppad on Cygwin systems).

02-25
  1. Add calculation of sparse Jacobians and Hessians; see sparse_jac and sparse_hes .
  2. Fix some documentation references to model_ref that should have been module_ref .


02-24
  1. Add calculation of Jacobian sparsity jac_sparsity and Hessian sparsity hes_sparsity .
  2. Improve the documentation and examples for vectors . In addition, add vec_bool to the list of vector types


02-23
Fix Cygwin version of make check_lib_lang for lang equal to python and perl. (The execution path needs to include the directory where the cppad_swig_lib Dll is located.) This make command is still not working on Cygwin when lang is octave.

02-22
The command make check_lib__perl was not working on Cygwin because of an error in lib/example/perl/CMakeLists.txt. This has been fixed (but is still not working on Cygwin).

02-21
Add the supporting packages heading to aid the install procedure.

02-20
  1. Change all the size_t variables in the interface to by int (this seems to work better in Swig). For example p is not an int instead of size_t.
  2. Added class for sparsity patterns sparse_rc (not yet hooked up to sparsity calculations).


01-18
Fix a problem with the perl Swig example int_array_class . To be specific, 
 
     @array_obj[@i] = 2 * i;
was changed to 
 
     pl_swig_xam::int_array_ptr_setitem(@array_obj, @i, 2 * @i);

01-17
  1. Change vector_double -> vec_double and vector_ad -> vec_a_double; see vector_ctor .
  2. Improve documentation: Change name of some sections; e.g. cppad_swig_lib -> library. Fix capitalization of Cppad Swig, M4, and other names in titles and headings. Include the optimize example in the documentation. Move M4 macro documentation inside of library section.


01-16
Add a_double conditional assignment , a_fun hessian , a_fun property , and a_fun optimize .

01-15
  1. Add the C++98 standard math functions to the a_double class; see a_double_unary_fun .
  2. Add near_equal for testing purposes.


01-14
Change the Swig example directory, and file names, from example to swig_xam. This distinguishes them from the Cppad Swig examples in lib/example. Group the library into one section and improve documentation of module names .

01-13
  1. Change error_msg to push and pop error messages so it handles multiple exceptions and deletes old messages.
  2. Move error_msg to error_message .
  3. Map calls to the Cppad error handler to calls to error_message.


01-12
Change xam.m4 macros to all start with Capital letters and use Camel case This is relevant to understanding on Cppad Swig generates examples for all languages from a single source.

01-11
  1. Add exception handling; see error_message .
  2. Fix the Perl examples / tests (exception handling revealed they were not working).


01-10
First uploading of this documentation.
Input File: whats_new/2017.omh