Hi, Yes sorry it should have been float(tmp).
Yes, could be added to the java code, and if it's complex its probably not an issue in many cases. Regards /Petrus On 14 Dec 2013, at 20:59 , Andi Vajda <va...@apache.org> wrote: > >> On Dec 14, 2013, at 16:46, Petrus Hyvönen <petrus.hyvo...@gmail.com> wrote: >> >> Hi! >> >> I am converting some java code to python scripts, using a JCC wrapped >> library. >> >> One thing that strikes me is that java seems to have a automatic typecast of >> int to float, which is not present in the JCC wrapped library. >> >> i.e. if a tmp=100 and obj.shift(tmp) is executed, and the method expects a >> float, it seems to fail in the wrapped code while working in java. Of course >> it can be explicitly typed obj.shift((float) tmp) but it does look as >> readable. > > In python, no, that syntax is incorrect. But you could use 100.0 instead of > 100. Almost as readable. Or add an intermediate method that casts to float. > > Could that feature be added to jcc ? sure, but it gets complicated fast. If > you control both ends of the app (the python and the java code), it's easiest > to add an int overload to the shift method and call the float version. > > Andi.. > >> >> This is a minor thing, but could it be enabled so that this automatic >> conversion from int to floats are done if no other matching signatures are >> found? (I assume this is the way java works, I'm no java expert..) >> >> Many thanks >> /Petrus >>
