Am 12.02.2026 um 17:25 schrieb Sven Schreiber:
<hansl>
function matrices Mhey(void)
return empty # gives generic "data error", works with defarray()
end function
Mhey()
</hansl>
A somewhat related follow-up report, but not the same thing as before.
1. (Re-) start gretl to make sure the session is "clean".
2. Load the attached script marrtest.inp (in the GUI program)
3. Run it the 1st time; for me this prints out "yay" without error.
4. Run it another time, then I get a "data not conformable" error
It's weird because the error only appears in the second run -- but it's
reproducible here on Windows with a recent post-2026a snapshot.
Note when I replace the line "return defarray()" with a more explicit
two-liner:
matrices M = empty
return M
then things seem to be more robust, so I suspect it's related to the
fact that defarray() alone doesn't convey sub-type information. OTOH,
using array(0) instead also works.
thanks
sven