I've got a problem that has been vexing me for several months: It seems that
when I create a custom rule using add_custom_command, the DEPENDS clause
sometimes works and sometimes doesn't, and I can't figure out why.
Here's what my custom rule looks like:
# Link with stdlib
string(REGEX REPLA
What's the recommended way in CMake to do automated dependency generation
for languages that aren't directly supported by CMake? Suppose my compiler
generates an output file which contains a list of all the source files that
were used to make it, and I have some tool that can extract this
informati
Thanks for those suggestions, I was on that path, but I can not get it to work
properly:
foo$ cmake .
-- Configuring done
-- Generating done
-- Build files have been written to: /home/bart/foo
foo$ make
Scanning dependencies of target foo
make[2]: *** No rule to make target ` /home/bart/foo/foo_i
according to
Notes to authors of FindXXX.cmake files
We would like all FindXXX.cmake files to produce consistent variable
names.
Please use the following consistent variable names for general use.
XXX_FOUND :
Set to false, or undefined, if we haven't found, or don't want to use
XXX.
XXX_
On Tue, Dec 29, 2009 at 6:01 PM, Hicham Mouline wrote:
> Hello,
>
> I am writing software that uses boost and Qt and other 3rd party libraries
> for win/mingw, win/msvc2008 and linux/gcc.
> The location of the 3rd party header files is different on each of the boxes
> I compile on.
>
> Is there
Hello,
I am writing software that uses boost and Qt and other 3rd party libraries for
win/mingw, win/msvc2008 and linux/gcc.
The location of the 3rd party header files is different on each of the boxes I
compile on.
Is there a file to put in the build directory? Is such a file automatically
re
I have a question about policy CMP0011.
I have:
$ cat Call_process_dashboard.cmake
# Call process_dashboard
# These files are maintained by Richard Wackerbarth
set(maintainer_email_account "Richard")
set(maintainer_email_domain "NFSNet.org")
IF(COMMAND CMAKE_POLICY)
CMAKE_POLICY(SET CMP0011
Hi,
I'm trying to find a way to build a library with two different instances of the
same source file.
for example I would like to compile foo.c with -DDINT into foo_int.o and with
-DDLONG into foo_long.o
and have both objects added to the same library libfoo.a
I have experimented with custom co
On 29. Dec, 2009, at 21:44 , Eric Noulard wrote:
> 2009/12/29 Richard Wackerbarth :
>> Not as a "real" solution, but more as a "workaround", you could create a
>> "derived source file" which is just a copy of the source file and then you
>> would have two separate files to be compiled with the
2009/12/29 Richard Wackerbarth :
> Not as a "real" solution, but more as a "workaround", you could create a
> "derived source file" which is just a copy of the source file and then you
> would have two separate files to be compiled with the appropriate flags.
That's one solution, see below for m
Not as a "real" solution, but more as a "workaround", you could create a
"derived source file" which is just a copy of the source file and then you
would have two separate files to be compiled with the appropriate flags.
On Dec 29, 2009, at 2:22 PM, Bart wrote:
> Hi,
>
> I'm trying to find a
Hi,
I'm trying to find a way to build a library with two different instances of the
same source file.
for example I would like to compile foo.c with -DDINT into foo_int.o and with
-DDLONG into foo_long.o
and have both objects added to the same library libfoo.a
I have experimented with custom co
Just put the implementation of the template class in the declaration. Get
that to work, then try to get fancy if you want to.
-
Aaron Wright
"e...@cs.bgu.ac.il"
Sent by: cmake-boun...@cmake.org
12/29/2009 08:54 AM
To
Eric Noulard
cc
cm
how much can this bloat my executable?
On Tue 29 Dec 19:16 2009 aaron_wri...@selinc.com wrote:
> With templates it's easiest to declare
> and implement in one file. The compiler likes this. Anything else you try
> requires extra work. Unless you're working in an embedded environment I
> don't see h
With templates it's easiest to declare and implement in one file. The
compiler likes this. Anything else you try requires extra work. Unless
you're working in an embedded environment I don't see how executable size
should even be a concern.
---
I'm trying to refrain from using that because it can bloat up the executable
size... unless you can tell me otherwise. in general the program will be
compiled using g++ for now, maybe I'll add support for windows compiler later.
also, I'm trying to keep the definitions to the header files and the
--On Tuesday, December 29, 2009 5:55 PM +0100 Michael Wild
wrote:
Have a look at the -C option which allows you to set cache variables
before anything else runs. You can use that in a batch script to set up
things automatically.
My initial attempt at that used John's example of if(MSVC) but
On 29. Dec, 2009, at 17:26 , Kenneth Porter wrote:
> Where can I put build instructions that are build-host-wide, not project-wide?
>
> I'm relatively new to cmake. I want to "inject" some code before the
> project's own script, and I'd prefer not to modify the project (since then
> I'd have t
ok, I've read the links, let say I'm not use the export option, when I look at
the how to avoid link errors here:
http://www.parashift.com/c++-faq-lite/templates.html#faq-35.13 I see that what
I've did is the same, isn't it?
I'm not sure where to follow from here... if I've did what they say an
> Where can I put build instructions that are build-host-wide, not
> project-wide?
>
> I'm relatively new to cmake. I want to "inject" some code before the
> project's own script, and I'd prefer not to modify the project (since then
> I'd have to maintain the patch).
>
> I want to set the CMAKE__PR
Where can I put build instructions that are build-host-wide, not
project-wide?
I'm relatively new to cmake. I want to "inject" some code before the
project's own script, and I'd prefer not to modify the project (since then
I'd have to maintain the patch).
I want to set the CMAKE__PREFIX vari
On Tue, Dec 29, 2009 at 10:39 AM, Kenneth Porter wrote:
> --On Monday, December 28, 2009 10:45 AM -0500 John Drescher
> wrote:
>
>> Here is what I do.
>
>> [script]
>
> Thanks. Must I put that in the project's CMakeLists.txt or can I capture
> that to a file I can inject each time I run cmake? I'
On Tue 29 Dec 17:16 2009 Eric Noulard wrote:
> 2009/12/29 e...@cs.bgu.ac.il :
> > hello Eric, thanks for the response.
> > I'd like to solve the Array issue first, then the boost because I have a
> > feeling it is a different one because when I disable the relevant code it
> > doesn't generates
2009/12/29 e...@cs.bgu.ac.il :
> hello Eric, thanks for the response.
> I'd like to solve the Array issue first, then the boost because I have a
> feeling it is a different one because when I disable the relevant code it
> doesn't generates the error even when other parts of the program (such as
hello Eric, thanks for the response.
I'd like to solve the Array issue first, then the boost because I have a
feeling it is a different one because when I disable the relevant code it
doesn't generates the error even when other parts of the program (such as the
threadpool).
here are the array fi
2009/12/29 e...@cs.bgu.ac.il :
> hello.
> I'm writing a c++ program that uses three shared libs that I wrote, for some
> reason, when it gets to the main cpp file, it fails with undefined error.
> here is the error report:
Concerning the link error [apparently] related to boost usage I let
Boost
hello.
I'm writing a c++ program that uses three shared libs that I wrote, for some
reason, when it gets to the main cpp file, it fails with undefined error. here
is the error report:
Linking CXX executable ../build/ossm
cd /home/da
27 matches
Mail list logo
