Hello Dirk, hello everyone,

On Thu, 2015-01-29 at 20:15 +0100, Dirk Bächle wrote:
> Hello Michael,
> 
> On 29.01.2015 17:22, Michael Lahnert wrote:
> > Is there a reason for omitting the cwd-parameter from the Glob-Function
> > in the environment class?

[with respect to the FS base class which is the function that the
Glob()-Function is calling, ML]

> the reason for this is, that setting the current working directory is not 
> required. Even though the Node.FS.FS class supports this 
> parameter, since it's written in a more generic fashion, the reading phase 
> ensures that the CWD is always set to the folder of the 
> currently parsed SConscript. In fact, large portions of the parsing phase 
> rely on this...
> 
> Does this help you further?

I'm no longer sure this is a subject that belongs on the dev list. So I
will reply to the users-list as well.

What I want to do is build a shared-library-bundle in various flavours
(e. g. debug/release, use different compilers, whatever) consisting of
several modules.
As long as I don't introduce any hierarchy inside the modules I am fine,
I can do something like it's described in
http://www.intenct.info/raymond.penners/2011/scons-build-better/
Instead of hard-coding all the source - files into my SConscript file I
can simply call Glob(), set the duplicate-option of the
variantDir-function to 0 and it will work just fine.

But now what is the "scons way" of compiling something like I sketched
below?
|-module1
   |-submodule1
       |- *.cpp
       |- *.hpp
   |-submodule2
       |- *.cpp
       |- *.hpp

Best regards
Michael

-- 
Dipl.-Inf. Michael Lahnert
Simulation großer Systeme
Institut für Parallele und Verteilte Systeme, Universität Stuttgart
Universitätsstraße 38
70569 Stuttgart | Germany

Tel: +49 (0)711 685 88560

Attachment: signature.asc
Description: This is a digitally signed message part

_______________________________________________
Scons-dev mailing list
[email protected]
https://pairlist2.pair.net/mailman/listinfo/scons-dev

Reply via email to