Re: [Gretl-devel] pre-release testing for gretl 1.8.0
On Thu, 15 Jan 2009, Riccardo (Jack) Lucchetti wrote:
there seems to be a problem with nested loops. Try this with CVS:
<script>
nulldata 10
loop 2
x = normal()
y = normal()
loop 2
k = 0
end loop
ols y 0 x
end loop
</script>
You get 4 ols-es instead of 2.
It seems to be a problem with the model-printing code in loops.
Only two models are estimated, but they're each printed twice.
You can see that the basic flow control is OK with this example:
loop 2
print "Top of outer"
loop 2
print "Doing inner"
end loop
print "Bottom of outer"
end loop
I have a fix in hand but I'll do some more testing before I commit
it.
Allin.