Greetings,
I'm currently attempting a patch for
http://www.cmake.org/Bug/view.php?id=15578 and trying to do it how Brad
suggests. So far, I've propagated a string down into the comparision
functor. This string comes from
cmGlobalVisiaulStudioGenerator::GetStartupProjectName(). However, I
Hi,
I noticed that sometimes when a target is reloaded, that the virtual folders
and targets for the complete visual studio solution get expanded.
For a small project this isn't a problem, but in a large project (~ 100
targets) this is a really big problem to close all targets every time.
Did a
Using a callback I now feed the error information back.
Here's the latest change:
https://github.com/ulrichard/CMake/commit/adf1cf5184d5a13702a03a090c501f1a6d0e39f4
The following line is still commented out:
this->Makefile->StoreMatches(re);
Is it important enough to add another callback?
What
We are pleased to announce that CMake 3.2.3 is now available for download.
Please use the latest release from our download page:
http://www.cmake.org/download/
Thanks for your support!
-
Changes in 3.2.3 since 3.2.2:
Brad
On 06/01/2015 03:41 PM, Tamar Kranenburg wrote:
> Thanks for splitting the files, much cleaner.
No problem. It makes it easier to review when the changes
are isolated with their own explanation. It also helps with
"git bisect" to find problems in the future.
> Just ran the tests on Mac, Windows
Hi Brad,
Thanks for splitting the files, much cleaner. Just ran the tests on Mac,
Windows and Linux and I confirm that it still works.
Regards,
Tamar
> On 1 jun. 2015, at 17:04, Brad King wrote:
>
> On 05/30/2015 09:27 AM, Tamar Kranenburg wrote:
>> Attached is my second attempt to add suppor
Removing the line
set_source_files_properties(${OUTPUT_FILE} PROPERTIES GENERATED 1)
...fixed it.
Indeed the documentation says "Each output file will be marked with the
GENERATED source file property automatically". However it does not mention that
setting that property ourselves prevents the
On 06/01/2015 02:03 PM, Robert Goulet wrote:
> Forgot to mention, we also do this after:
> set_source_files_properties(${OUTPUT_FILE} PROPERTIES GENERATED 1)
That shouldn't be necessary. add_custom_command does that for you.
> list(APPEND SOURCE_FILES ${OUTPUT_FILE})
> [...]
> add_library(${PROJ
> Are these absolute paths?
Yes. All absolute file paths.
> Where is the output file referenced? It needs to be in a target somewhere.
Forgot to mention, we also do this after:
set_source_files_properties(${OUTPUT_FILE} PROPERTIES GENERATED 1)
list(APPEND SOURCE_FILES ${OUTPUT_FILE})
[...]
add_
On 06/01/2015 01:19 PM, Robert Goulet wrote:
> Does CMake custom_command rebuild the output file if the
> dependency files change?
Yes, it should. This is well tested.
> DEPENDS ${INPUT_FILES}
Are these absolute paths?
> OUTPUT ${OUTPUT_FILE}
Where is the output file referenced? It needs to
Hi,
Does CMake custom_command rebuild the output file if the dependency files
change? Seems like it should but I can't get it to work.
The CMake command is written like this:
add_custom_command(
OUTPUT ${ OUTPUT_FILE}
On 06/01/2015 11:55 AM, Paul Anton Letnes wrote:
> CMake combined with GNU Make does indeed handle dependencies
> for Fortran, so it can't be impossible - but maybe it is hard
> enough that noone bothered implementing it for ninja.
Design for that was discussed in a thread on cmake-developers here
Sorry David, my bad.
CMake combined with GNU Make does indeed handle dependencies for Fortran, so it
can’t be impossible - but maybe it is hard enough that noone bothered
implementing it for ninja.
Paul
> I have no experience with Fortran, and from what I understand,
> parallel builds of Fort
On 05/30/2015 09:27 AM, Tamar Kranenburg wrote:
> Attached is my second attempt to add support for PostgreSQL 9.3 and 9.4
> on Debian/Ubuntu.
Thanks. Although we currently have a freeze in place in preparation
for 3.3, this can be considered an environmental regression due to
the availability of
On 05/30/2015 09:50 AM, Sylvain Joubert wrote:
> The recent development to support different LIBRARY_DIR for Boost have
> made the new cache entries appear in the default cache.
For reference, this was here:
FindBoost: Search for debug and release libraries separately
http://cmake.org/gitweb?p
On 06/01/2015 05:10 AM, Nagy-Egri MĂĄtĂŠ Ferenc wrote:
> Obtaining a list of files with identical compiler switches might be tricky.
[snip]
> Deferred generation of the makefile is not nice, but might be the only way.
Yes. However, I'd prefer not to introduce major refactoring to the
Makefile gen
CMake + Ninja is definitely the best move you can make for giving your
developers the minimum possible build time on Windows.
NMake is basically useless, its underlying code hasn't been updated in
years, and it does absolutely nothing to try to take advantage of
multiple cores.
Fastest Windows bu
At my company, we didn’t bother trying to use NMake.
We used the Visual Studio solution generator and the proper compilation flags
and msbuild flags to build in parallel. This proved to speedup our builds quite
a lot and was very easy to integrate.
The other solution you mentioned is “ninja” and
The following issue has been SUBMITTED.
==
http://www.cmake.org/Bug/view.php?id=15599
==
Reported By:trsystran
Assigned To:
==
Thanks Brad for the detailed description. From what I’ve read about the 3 level
makefile system, it should be pretty straightforward to implement batch-mode
support.
Actual compilation takes place only in the last level, and throughout
generation, this is the point where all the compilation op
20 matches
Mail list logo
