You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I noticed that in blocktest the scriptbuilder tool is silently not build if Qt5 is not found, while it is always enabled if Qt5 is found.
While this behavior make sense as a default behavior, from the point of view of packaging it may not be ideal. For example in the robotology-superbuild, it we want the blocktest to built scriptbuilder, we want to get a clear error if Qt5 is not found, instead of silently disabling the compilation of scriptbuilder. See https://www.youtube.com/watch?t=654 for more details.
A good tradeoff to maintain the current behaviour while giving to packagers the possibility of explicitly specifying if Qt5 should used or not is the following:
I noticed that in blocktest the scriptbuilder tool is silently not build if Qt5 is not found, while it is always enabled if Qt5 is found.
While this behavior make sense as a default behavior, from the point of view of packaging it may not be ideal. For example in the robotology-superbuild, it we want the blocktest to built scriptbuilder, we want to get a clear error if Qt5 is not found, instead of silently disabling the compilation of
scriptbuilder
. See https://www.youtube.com/watch?t=654 for more details.A good tradeoff to maintain the current behaviour while giving to packagers the possibility of explicitly specifying if Qt5 should used or not is the following:
In this way, the default behavior remains the same, but if I explicitly set the BLOCKTEST_USES_Qt5 option to ON, the build fails if Qt5 is not found. See https://github.com/robotology/idyntree/blob/v2.0.2/cmake/iDynTreeDependencies.cmake#L46 for an example of such usage.
The text was updated successfully, but these errors were encountered: