[Insight-developers] gcc 4.3.2 on dashboard : KWStyle versioning.

Julien Jomier julien.jomier at kitware.com
Mon Dec 8 15:47:54 EST 2008


Luis,

I have added a check on the KWStyle version in the 
Utilities/KWStyle/KWStyle.cmake. Style checking will be enabled only if 
the correct KWStyle version is found.

Julien

Luis Ibanez wrote:
> 
> Julien,
> 
> Would this generate a CMake error message if the user attempts
> to invoke an older (than required) KWStyle executable ?
> 
> If so, yes, by all means, that will do the trick.
> 
> 
>    Thanks
> 
> 
>       Luis
> 
> 
> -------------------
> Julien Jomier wrote:
>> Luis,
>>
>> I can put a KWStyleConfig.cmake at the top level of the KWStyle source 
>> tree, would that work for you?
>>
>> Julien
>>
>> Luis Ibanez wrote:
>>
>>>
>>> Arnaud,
>>>
>>>
>>> Good point,
>>>
>>> We have actually fall for this one often enough that we probably
>>> should do something about it.
>>>
>>> For example:
>>> have a KWStyle required version in our top CMakeLists.txt file.
>>>
>>> Julien: is there a way of getting a version number from KWStyle and
>>>         checking it with CMake  ?
>>>
>>>
>>>   Thanks for any suggestions,
>>>
>>>
>>>     Luis
>>>
>>>
>>> ---------------------
>>> Arnaud Gelas wrote:
>>>
>>>> ooops, the path to KWStyle was set to an older version...
>>>>
>>>> It now works properly:
>>>> http://www.cdash.org/CDash/viewTest.php?onlypassed&buildid=223083
>>>> http://www.cdash.org/CDash/viewTest.php?onlypassed&buildid=223082
>>>>
>>>> Arnaud
>>>>
>>>> On Dec 1, 2008, at 3:56 PM, Gelas, Arnaud Joel Florent wrote:
>>>>
>>>>> Ok, I'll turn it off.
>>>>>
>>>>> Arnaud
>>>>>
>>>>>
>>>>> Bill Lorensen wrote:
>>>>>
>>>>>> Arnaud,
>>>>>>
>>>>>> I know, I've seen that before. Strange... Maybe you should turn off
>>>>>> KWStyle testing for that platform. We have several others that run
>>>>>> KWStyle.
>>>>>>
>>>>>> Bill
>>>>>>
>>>>>> On Mon, Dec 1, 2008 at 3:46 PM, Arnaud Gelas
>>>>>> <arnaud_gelas at hms.harvard.edu> wrote:
>>>>>>
>>>>>>> Bill,
>>>>>>>
>>>>>>> When I run ctest on KWStyle, there is no error.
>>>>>>> With my machine (using gcc 4.3.0), there is no style error for  
>>>>>>> itk...
>>>>>>>
>>>>>>> It seems there is something wrong with gcc-4.3.2...
>>>>>>>
>>>>>>> Note that most of these errors occur after:
>>>>>>> public:
>>>>>>> ...
>>>>>>> #ifdef ITK_USE_CONCEPT_CHECKING
>>>>>>> ...
>>>>>>> #endif
>>>>>>>
>>>>>>> protected:
>>>>>>> constructor
>>>>>>> destructor
>>>>>>>
>>>>>>> Arnaud
>>>>>>>
>>>>>>> Bill Lorensen wrote:
>>>>>>>
>>>>>>>> Arnaud,
>>>>>>>>
>>>>>>>> Some of these style errors look suspicious:
>>>>>>>> http://www.cdash.org/CDash/testDetails.php?test=13357110&build=223052 
>>>>>>>>
>>>>>>>>
>>>>>>>> I don't see these errors on my builds.
>>>>>>>>
>>>>>>>> Bill
>>>>>>>>
>>>>>>>> On Mon, Dec 1, 2008 at 3:21 PM, Arnaud Gelas
>>>>>>>> <arnaud_gelas at hms.harvard.edu> wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>>> Hi,
>>>>>>>>>
>>>>>>>>> I have just upgraded sanchome, and from tonight (normally) itk  
>>>>>>>>> will be
>>>>>>>>> compiled and tested with gcc-4.3.2.
>>>>>>>>> I have just ran an experimental ctest this afternoon in debug,  
>>>>>>>>> and in
>>>>>>>>> release (after updating KWStyle), and there are only few style  
>>>>>>>>> errors...
>>>>>>>>>
>>>>>>>>> Regards,
>>>>>>>>> Arnaud
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> Insight-developers mailing list
>>>>>>>>> Insight-developers at itk.org
>>>>>>>>> http://www.itk.org/mailman/listinfo/insight-developers
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Insight-developers mailing list
>>>>> Insight-developers at itk.org
>>>>> http://www.itk.org/mailman/listinfo/insight-developers
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Insight-developers mailing list
>>>> Insight-developers at itk.org
>>>> http://www.itk.org/mailman/listinfo/insight-developers
>>>>
>>>
>>
> _______________________________________________
> Insight-developers mailing list
> Insight-developers at itk.org
> http://www.itk.org/mailman/listinfo/insight-developers
> 


More information about the Insight-developers mailing list