Automatic recording of required packages...

[Allen Winter]

On Sunday 31 October 2010 8:40:39 am Alexander Neundorf wrote:
> Hi,
> 
> the upcoming 2.8.3 release of cmake (_please_ test the rc4) will have an 
> improved logging for find_package() calls.
> Together with the enhanced FeatureSummary.cmake module of cmake 2.8.3 it can 
> produce a text file which contains which packages where found and which were 
> not found for each find_package() call in a project, no further work required 
> (except one call to feature_summary() at the end of the toplevel 
> CMakeLists.txt
> 
> This is already supported in MacroLogFeature.cmake in kdelibs.
> The result of this is a file ${CMAKE_BINARY_DIR}/FindPackageLog.txt, I 
> attached such a file (FindPackageLog.txt) from one of my builds.
> 
> If you look at this file, you may notice that in the list of packages which 
> have been found it is not recorded whether the package was optional or 
> required.
> 
> So I enhanced this a bit more, which resulted in FindPackageLog.txt-required.
> Maybe you see the problem, basically all packages are listed as "OPTIONAL", 
> while I guess not all of them are optional.
> The problem is our macro macro_log_feature(), or more to the point 
> the "REQUIRED" argument for macro_log_feature().
> This is how we use that:
> 
> find_package(ZLIB)
> macro_log_feature(ZLIB_FOUND
>                   "ZLib" 
>                   "The Zlib compression library"
>                   "http://www.zlib.net" 
>                   TRUE 
>                   "" 
>            "Required by the core KDE libraries and some critical kioslaves.")
> 
> Here this argument is set to TRUE.
> So also if ZLIB has not been found, cmake processing continues, and if at the 
> end some packages which were marked this way as "REQUIRED" have not been 
> found, an according error message is printed and cmake exits with an error.
> 
> This has two problem: cmake processing continues, while some variables which 
> are expected to be set (e.g. ZLIB_INCLUDE_DIR) are used in CMakeLists.txt 
> expecting that they are set to a valid location since ZLIB is REQUIRED, but 
> they are not. Same for library or other variables. This can cause other cmake 
> errors.
> 
> The second problem is what you see in FindPackageLog.txt-required . Basically 
> all our packages are listed as OPTIONAL there.
> 
> I never liked this "REQUIRED" argument for macro_log_package(), and now I like 
> it even less.
> We tried to outsmart/work around cmake behaviour instead of going for a proper 
> solution.
> 
> So, where am I going with this...
> 
> I think I'd like to propose that macro_log_feature() ignores the REQUIRED 
> keyword (we can't simply remove it due to compatibility reasons) and that we 
> use REQUIRED in the find_package() calls instead.
> 
> Pros: 
> - correct listing of all found and not-found packages by FeatureSummary.cmake
> - cmake stops processing when something is REQUIRED has not been found
> 
> Cons: no summary of multiple missing required packages at the end of the cmake 
> run, only the last missing REQUIRED dependency will be printed.
> 
There is another con, the one that always irked me:

if  we have find_package(foo REQUIRED) and foo isn't found, then we do not see
the nice message telling us what foo is and how to get it and why it is needed.

So that's why i like the ZLIB feature discovery done like it is.
But if your proposal solves this, then I have no objections.

Bottom line: if a required package is not found, then we still should see nice user messages.





-- 
Allen Winter | allen at kdab.net | Software Engineer
KDAB (USA), LLC, a KDAB Group company
Tel. USA +1-866-777-KDAB(5322), Sweden (HQ) +46-563-540090
KDAB - Qt Experts - Platform-independent software solutions