Changes since 2.99.14:

 . Added 'native' IDL keyword. The meaning is slightly different from that of
   the OMG IDL. GenoM 'native' keyowrd defines a new opaque type, that can be
   used only in codels (or IDS). The type is manipulated as a pointer by GenoM
   and the templates, and codels are responsible for the actual type
   definition or memory management. This is typically used to store types in
   the IDS that cannot be represented as IDL (external types, opaque types,
   etc.).

 . Renamed #pragma native to #pragma masquerade. This is to avoid confusion
   with the new 'native' IDL keyword. The #pragma purpose is also hopefully
   clearer.

 . Updated mappings for the IDL 'char' type. 'char's are now mapped to
   uint8_t in C/C++, so that the C 'char' type is reserved for strings.

 . Fixed grammar to allow compound initializers with just a doc string and no
   value.

 . Improved syntax error handling in the parser. In particular, missing
   semicolons or syntax erros in codels parameters should not spew loads of
   errors anymore.

 . Updated skeleton template to generate $component-genom.pc instead of
   lib$component_codels.pc, because the top level package is most often used in
   other genom components to include a part of the interface (rather than
   linking with the codels).

 . Fixed skeleton '-l c++' option for compiling codels with a C++ compiler.

 . Temporary disabled support for C++ codels interface. This will not be
   supported in 3.0, but remains on the TODO list. C++ codels must use the C
   interface in the meantime.

 . Fixed a bunch of minor bugs and added respective regression tests.

The patch includes explicity <sys/types.h> to get the definition of
ssize_t. It is equivalent to commit cd27fb49d in mainstream.

Following the author's advice, re-enable this package but disable parallel
jobs, because some files make gcc grow up to 1GB during compilation. This is an
issue with the current setup for bulk builds.

This raises too many issues with the current bulk + virtualbox setup.

gcc eats more than 1GB, especially for a couple of source files, and it either
segfaults after reaching the datasize/stacksize limit or compiles for hours,
preventing the rest the bulk from being properly tested.

py-morse is the Python bindings for the MORSE simulator.
It uses underneath the MORSE socket interface.

Changes since 0.5:

  - More sensors are exportable through POM_SENSOR_POS interface
  - Various bug fixes in the tclserv protocol handling
  - Add various overlays for genom modules (platine, viam, rflex, ...)
  - Adapt to new data structures (viman in particular)

. Add a -config.cmake in SYSTEM_SEARCH

. Export GENLISP_BIN in depend.mk

. Add a -config.cmake in SYSTEM_SEARCH

. Export GEN{MSG,SRV}_PY_BIN in depend.mk

. Add a -config.cmake in SYSTEM_SEARCH

. Export GENCPP_BIN path in depend.mk

. Add indirect dependency on roscpp-core in depend.mk

. Add a bunch of %-config.cmake in SYSTEM_SEARCH

. Add rospack as an indirect dependency of catkin

. Add %-config.cmake in SYSTEM_SEARCH, so that cmake sysdep will
  add -Dcatkin_DIR et al.

While here, remove old patch that is not relevant anymore.

Compilation tested with options: pr2 spark tcl

Fix a fatal issue in server/configure.ac.

Bump PKGREVISION.

Remove an unneeded cslib.h dependancy.

Cleanup the configure.ac

Fix a bug in the EncodeDecodeOpenprs.c template which would screw up a
test (a trailing ; in a test).

Make sure to pass all variables defining the proper python version as found by
robotpkg, otherwise ros CMakeFiles may happily find python-3.2 and use it
instead of python<3, although this does not work in the end (at least ros-rx
does this and breaks on F17).

This is done in the common ros Makefile as this affects all packages.

This template produces encoding and decoding functions in a module
(which can be loaded by an Openprs kernel), as well as basic OPs to
have OpenPRS to be a client of a genom3 module (using the
{ros,pocolibs}/client/c library.

tcl-genomix package provides a TCL package that interacts with the genomix HTTP
server and can control GenoM3 components.

It can prompt interactively for arguments of services. Services can be invoked
synchronously or asynchronously and callbacks can be triggered whenever
services complete.

tcl-genomix is best used in combination with the interactive eltclsh shell.

The genomix HTTP server is a generic interface between clients and genom
components (using the generic genom C client template). Control is done be the
mean of specific HTTP GET requests (not yet documented...).

The genom3-ros template (see meta-pkgs/ros-base) provides a GenoM3
template for generating ros-based components.

The genom3-pocolibs template (see middleware/pocolibs) provides a GenoM3
template for generating pocolibs-based components.

The Generator of Modules GenoM is a tool to design real-time software
architectures. It encapsulates software functions inside independent
components. GenoM is more specifically dedicated to complex on-board systems,
such as autonomous mobile robots or satellites, that require:

. The integration of heterogeneous functions with different real-time
  constraints and algorithm complexities (control of sensors and actuators,
  data processings, task planification, etc.).

. An homogeneous integration of these functions in a control architecture which
  requires coherent and predictable behaviors (starting, ending, error
  handling), and standard interfaces (configuration, control flow, data
  flow).

. The management of parallelization, physical distribution and portability of
  the functions.

. Simple procedures to add, modify or (re)use the functions by non-specialists

GenoM generates the source code of components by using:

. A generic template, common for all components. This guarantees that all
  components share the same consistent behaviour. The template itself is not
  part of GenoM, so that different template kind can be developped easily.

. A formal description of the components interface. This description is based
  on a simple language using OMG IDL for data types definitions and a custom
  syntax for the description of a more detailed component model.

The project is released under an open-source, BSD-like license.