[omniORB] omniORB's future (and mine too)

[Stefan Seefeld]

Duncan Grisby wrote:
> On Thursday 25 April, Stefan Seefeld wrote:
> 
> 
>>I'd suggest splitting omniidl into a generic IDL parser (a simple
>>python module) and the executable / backends. This would make it
>>even more appealing for others to use it, I think.
> 
> 
> What do you mean?  omniidl is already a generic IDL parser with
> separate back ends.

ok, I wasn't very clear. All I meant to suggest was to find a way
to make omniidl as a generic IDL parser more popular. May be that
just means to release suitable packages (deb, rpm, etc.) for it,
and advertising its existence on www.python.org.

>>Second, I'd like to add a 'make doc' target to the build system
>>to use the synopsis tool to generate a reference manual for the
>>entire code base of omniORB (IDL, C++, python). Unfortunately,
>>I haven't been able to figure out how the build system works, with
>>all the include and macro / canned command sequence machinery.
> 
> 
> A "make doc" would be great. I think it should be easy enough to add
> rules to each of the dir.mk files to create documentation. Any common
> requirements could go in beforedir.mk for the old-style build and
> beforeauto.mk.in for the autoconf build. You can probably safely
> ignore most of the make nastiness.

ok, will look into it again.

Regards,
		Stefan