1 p.m.
Hi folks,
using latest git under Ubuntu yields a weird error with a loop. The
following definitely used to work in the past but fails now with an
"Data error".
The function get_dynamical_features() simply constructs a list of lags
of some series. I call the function within a loop three times. At the
second iteration it fails (see error message below).
<hansl>
clear
set verbose off
function list get_dynamical_features (const list Dataset,
const bundle Config)
series dx = Dataset.dx
list L = dx(0 to -Config.LAG_LENGTH)
return L
end function
nulldata 10
loop i=1..3
print i
series dx = normal()
bundle B = _(LAG_LENGTH = 2)
list R = get_dynamical_features(dataset, B) # breaks at 2nd iteration
R
endloop
</hansl>
with the error:
<error>
gretl version 2023c-git
Current session: 2023-07-26 14:56
? clear
i = 1.0000000
dx dx_1 dx_2
i = 2.0000000
Data error
*** error in function get_dynamical_features, line 1
series dx = Dataset.dx
> list R = get_dynamical_features(dataset, B)
</error>
Any idea what's going on here?
Best
Artur
1:52 p.m.
Am 26.07.2023 um 15:00 schrieb Artur T.:
Hi folks, > > using latest git under Ubuntu yields a weird
error with a loop.
The > following definitely used to work in the past but fails
now with
an > "Data error". > > The function get_dynamical_features() simply
constructs a list of > lags of some series. I call the function within a
loop three times. > At the second iteration it fails (see error message
below).
Hi, yes it fails here, too, and it's weird. (gretl 2023b) Depending on
what the precise cause is, this could be a nasty bug indeed...
cheers
sven
4:37 p.m.
On Wed, Jul 26, 2023 at 9:00 AM Artur T. <atecon(a)posteo.de> wrote:
Hi folks,
using latest git under Ubuntu yields a weird error with a loop. The
following definitely used to work in the past but fails now with an
"Data error".
Thanks for the report, Artur. Here's a minimal script to expose the
bug you found:
<hansl>
function void list_access (const list L)
series dx = L.dx
end function
nulldata 10
series dx = index
loop i=1..3
# fails after first iteration
list_access(dataset)
endloop
</hansl>
This sort of thing last worked with gretl 2022b; unfortunately it has
been broken since 2022c. That is because (a) 2022c introduced
compilation of "genrs" within functions that are called from a loop
(as in this example), and (b) the internal code to handle expressions
of the form listname.seriesname was not robust with respect to such
compilation. In fact it was funky old ad-hocery.
The problem is now fixed in git, with a proper, modern backend for
listname.seriesname.
To be clear, here's the exact scope of the bug that's now fixed:
* It bites when function F uses listname.seriesname and F is called
within a loop.
* It does not bite if F is called repeatedly, but not in a loop.
* It does not bite if listname.seriesname is called directly within a
loop (not via a user function).
Allin
4:46 p.m.
Hi Allin,
Thanks for the quick fix which works fine here on my machine.
Also, thanks, Sven, for confirming the bug.
Best
Artur
Am 26.07.23 um 18:37 schrieb Cottrell, Allin:
> On Wed, Jul 26, 2023 at 9:00 AM Artur T. <atecon(a)posteo.de> wrote:
>>
>> Hi folks,
>>
>> using latest git under Ubuntu yields a weird error with a loop. The
>> following definitely used to work in the past but fails now with an
>> "Data error".
>
> Thanks for the report, Artur. Here's a minimal script to expose the
> bug you found:
>
> <hansl>
> function void list_access (const list L)
> series dx = L.dx
> end function
>
> nulldata 10
> series dx = index
> loop i=1..3
> # fails after first iteration
> list_access(dataset)
> endloop
> </hansl>
>
> This sort of thing last worked with gretl 2022b; unfortunately it has
> been broken since 2022c. That is because (a) 2022c introduced
> compilation of "genrs" within functions that are called from a loop
> (as in this example), and (b) the internal code to handle expressions
> of the form listname.seriesname was not robust with respect to such
> compilation. In fact it was funky old ad-hocery.
>
> The problem is now fixed in git, with a proper, modern backend for
> listname.seriesname.
>
> To be clear, here's the exact scope of the bug that's now fixed:
>
> * It bites when function F uses listname.seriesname and F is called
> within a loop.
> * It does not bite if F is called repeatedly, but not in a loop.
> * It does not bite if listname.seriesname is called directly within a
> loop (not via a user function).
>
> Allin
> _______________________________________________
> Gretl-devel mailing list -- gretl-devel(a)gretlml.univpm.it
> To unsubscribe send an email to gretl-devel-leave(a)gretlml.univpm.it
> Website: https://gretlml.univpm.it/postorius/lists/gretl-devel.gretlml.univpm.it/
10:10 a.m.
Am 26.07.2023 um 18:46 schrieb Artur T.:
Hi Allin,
Thanks for the quick fix which works fine here on my machine.
Also, thanks, Sven, for confirming the bug.
I think the good news is that it's not a brand-new bug/regression
introduced with 2023b. If we survived at least two releases with this,
then it cannot be that bad in practice.
thanks
sven
512
days inactive
514
days old
4 comments
3 participants
participants (3)
-
Artur T. -
Cottrell, Allin -
Sven Schreiber