I just had to recompile ITK because something in ImageStatistics relies on review code from ITK. Is that ok? I guess yes, but perhaps there are arguments against it?
Description
Description
| Status | Assigned | Task | ||
|---|---|---|---|---|
| Resolved | None | T2494 META: Create usable end-user application / demo (3M3) | ||
| Resolved | None | T2843 Informative: Mitk3M relies on ITK review classes |
Event Timeline
Comment Actions
Is it possible to #ifdef the code? How important is the function? ITK Review ON should not be the default for MITK in general I think. It's not always well tested, the API is not yet stable.
Is it a single class, maybe just a header. Then we could import it in the MITK source tree until it is merged to ITK.
Comment Actions
I'll prepare to switch the continuous client to a NO-ITK-REVIEW version of ITK 3.16 again
Comment Actions
Xin imported two filters from ITK Review to our source tree as well. But for us the option of just including them was not viable, as their behaviour was buggy.
Comment Actions
continuous clients could be switched to a "normal" ITK any time now. I'll bring this up for discussion in the MITK meeting
Comment Actions
ok, continuous clients use an ITK version without review classes now. so far everything seems fine.