[CMake] Controlling Doxygen using CMake

[Robert Dailey]

This is more or less a design question. I use CMake to generate visual
studio projects for a collection of 3 subsystems (components). Each
subsystem has any number of separate projects (DLLs, Executables, etc).
Each subsystem defines its own doxygen configuration file which is used to
generate tagfiles and separate HTML output for documentation contained in
only that subsystem. Subsystems may be dependent on each other (library A
uses library B, both in different components), so doxygen configuration
files are configured to link to external tagfiles for other component's
documentation.

It seems complex to let CMake handle this. I don't know whether I should
have custom targets setup for each component/subsystem that controls
generating individual documentation for each. I could call project() and
include ONLY custom targets in it for documentation generation, with
appropriate dependencies setup. I also thought of having a separate script
(python or batch file, something of that sort) to handle generating
documentation. A third option would be to have CMake invoke doxygen itself
during the configuration process, but this would cause huge delays and
doesn't seem to be the right place to do this.

Does anyone have a similar setup? What recommendations would anyone have
for me in this situation?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.cmake.org/pipermail/cmake/attachments/20120504/90cd8ac9/attachment.htm>