[cmake-developers] Automoc in cmake

[Rolf Eike Beer]

Clinton Stimpson wrote:
> On Wednesday, June 08, 2011 12:59:36 pm Alexander Neundorf wrote:

> > Beside that, I didn't ask explicitely, but I'm sure without automoc
> > cmake
> > is out of the game for becoming the buildsystem for Qt5 (because in Qt4
> > they do automoc in qmake).
> 
> What if this was a generic tool that could scan code to handle custom
> dependencies, with the ability to do two things: (re)generated dependent
> files or (re)generate the file with dependencies (sounds like a make tool
> ;) ). Don't some Visual Studio users do this type of custom build
> phases/steps/thingy by creating an Add-In?
> 
> Would that be an automoc equivalent if there was a macro that gave it the
> rules for generating the dependencies?
> 
> Also, (a long outstanding bug), a Qt .qrc file has dependencies specified in
> xml, which should be handled at build time.

Isn't there already a build-time scanner to add dependencies? Those C/C++ files 
need to be scanned to identify header dependencies which may change between 
build runs also, no? So isn't this all about the same stuff? And for automoc 
it's basically only a special case of this since this is also an include, just 
one with a special pattern that adds additional dependencies.

I'm all in favor of a generic solution. So maybe we could step back and look 
from a slightly greater distance at this. What would such a scanner need to 
know? And what does it need to tell CMake to run properly?

Tell CMake:
-which source types it wants to pass (C/C++/Java/ASM/other/*)
-if it cares for header files
-to avoid tokenizing all files in all scanners a pattern to look for. If that 
pattern is empty every file is passed to the scanner, otherwise only those 
where at least one matches that pattern
-a flag if only matching lines should be send to the scanner (greatly reducing 
computation time)

What it would return:
-for every file passed a list of files this one would depend on (testcases to 
reject: file depending on itself, introducing simple loops)

That would work out for the qrc and header scanner as the files they depend on 
are usually already there or defined by some target. This would not be enough 
for the automoc case since someone needs to create the targets that generates 
the moc_foo.cxx or foo.moc. So the scanner needs to be able to return 
additional CMake code that would be parsed and used as if it was defined right 
before the library/executable we try to create.

The next step would be to think about how people could come up with additions. 
Say I want to write a scanner for some language. Must it be C++ code? Or can 
this somehow be hooked into CMake just as CMake script? How would that go? 
This way we could implement the whole automoc stuff just this way.

Alex, would this be a sufficient description for automoc?

language: C++
header files: only
pattern: Q_OBJECT

result files: none (will be handled by the usual include scanner of the C++ 
file)
result code: ADD_CUSTOM_TARGET(... ${QT_MOC_EXECUTABLE} ...)

Eike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://public.kitware.com/pipermail/cmake-developers/attachments/20110609/754cb38f/attachment.sig>