Peter Willendrup

Ended up with an actual exit(-1), which I think is for the better (here run with the original flipfield.dat): ``` (mcstas-3.x-dev/miniconda3) Test_Pol_Tabled-hack $ mcrun Test_Pol_Tabled.instr MF=flipfield.dat interpol_method=regular -n1e4 [Test_Pol_Tabled] Initialize...

BTW I also spotted yet another annoyance in the regular -> kdtree switch case logic: https://github.com/McStasMcXtrace/McCode/blob/195d18aa0bed5b06958c5923a58149aafda44ee7/common/lib/share/interpolation-lib.c#L434-L462 An `if (this_step` was actually missing - so even on actual regular grids, the...

On the McStas side I have seen that there are sometimes issues along these lines if a component instance name starts with a numeric (e.g. `1stmonitor`) is this potentially the...

@farhi I found what the issue is: The "missing" newline in the `INITIALIZE fprintf` makes the whole difference, the attached .instr.txt uses ` fprintf(stdout,"%s: Energy %g [keV] Mono Angle %g...

Edited the instr on main, but maybe we should think of a better / more robust way of fixing this...

Would it be too ugly if the cogen inserted a `printf("\n");` at the end of INTITALIZE? ( I am guessing this would fix almost anything we can imagine people doing...

... if documented properly :-D (Otherwise it will be the good old "WTF did anyone put a printf here... ;-) " )

Implemented for release of McStas 3.2 and 2.7.2

Hi @KBGrammer, I am now finally in the process of merging in your patches for release with the forthcoming McStas 2.7.2 and 3.2 releases (and by symmetry McXtrace 1.7.1 and...

'heck I am pushing the edits now under the assumption that ~ (0,0,0) works - and we will see in tomorrow's test results what / if there is something to...