HP49G/ROM 1.16: Serious decompiler bug.

[Ivan Cibrario Bertolotti]

Hello,

with ROM 1.16, '(-2)^4' is decompiled as -2^4 for STD stack display
*AND* command-line editing; '(-A)^B' is also decompiled as '-A^B'
when both A and B are undefined. Instead, (-2.)^4 is decompiled
correctly, and EQW works correctly with all examples.

In my opinion, this is a serious bug, because order of operation
can be altered when decompiling/recompiling an alg. expression,
even when, for example, the expression is embedded into a program.

Best regards,
Ivan

--
------------------------------------------------------------------------
Ivan Cibrario Bertolotti | phone: +39-011-3919246
IRITI - National Research Council | fax: +39-011-341882
Turin (Italy) | mailto:cibr...@cstv.to.cnr.it

[Al Arduengo]

Thge interestiong thing about this is that the answers _ARE_
correct. Even though they appear the same on the stack they provide
different answers and if you do a COMP-> on them you see that they are
indeed different. Still, I think this should be addressed.

-Al

[Ivan Cibrario Bertolotti]

Hello,

I agree, but my point was to show that answers are correct *only* as
long as you *do not edit* the expression; for example, if you enter
the following program (ensure that -2 and 4 are integers and not
reals):

<< '(-2)^4' EVAL >>

the program works correctly only until you edit it; when you
edit it, even without making any change (with down-arrow ENTER,
for example), the program will work no longer.

This is the main reason I wrote 'serious bug';
I don't want to have to double check and fix all alg. expressions
in my programs whenever I edit them. Don't you? 8-)

Best regards,
Ivan

Al Arduengo wrote:
>
> Thge interestiong thing about this is that the answers _ARE_
> correct. Even though they appear the same on the stack they provide
> different answers and if you do a COMP-> on them you see that they are
> indeed different. Still, I think this should be addressed.
>
> -Al
> Ivan Cibrario Bertolotti <cibr...@cstv.to.cnr.it> writes:
>
> > Hello,
> >
> > with ROM 1.16, '(-2)^4' is decompiled as -2^4 for STD stack display
> > *AND* command-line editing; '(-A)^B' is also decompiled as '-A^B'
> > when both A and B are undefined. Instead, (-2.)^4 is decompiled
> > correctly, and EQW works correctly with all examples.
> >

--

[Joėl Bourquard]

> This is the main reason I wrote 'serious bug';
> I don't want to have to double check and fix all alg. expressions
> in my programs whenever I edit them. Don't you? 8-)

Nasty thing.. looks pretty clear to me !
One should report it to ACO.. Have you done it yet ? If not, may I tell
Cyrille ?

Best Regards,
Joėl

[Ivan Cibrario Bertolotti]

Hello Joël,

I already forwarded a copy of my message to hp...@hpaco.aus.hp.com,
but I think it could be useful if you get in touch with
ACO people directly.

While you are at it, could you please confirm that the bug
is really there, that is, the incorrect behavior I noticed
is not related to some peculiarities of my machine, such as
flag settings? I don't think so, but a double check is always
a good thing to have. Thank you very much.

Best regards,
Ivan

"Joël Bourquard" wrote:
>
> Nasty thing.. looks pretty clear to me !
> One should report it to ACO.. Have you done it yet ? If not, may I tell
> Cyrille ?
>
> Best Regards,

> Joël

[aven...@epita.fr]

Hello

This is definitely a bug.. :(

Jean-Yves

In article <38576A97...@cstv.to.cnr.it>,
Ivan Cibrario Bertolotti <cibr...@cstv.to.cnr.it> wrote:
> Hello Jo=EBl,

>
> I already forwarded a copy of my message to hp...@hpaco.aus.hp.com,
> but I think it could be useful if you get in touch with
> ACO people directly.
>
> While you are at it, could you please confirm that the bug
> is really there, that is, the incorrect behavior I noticed
> is not related to some peculiarities of my machine, such as
> flag settings? I don't think so, but a double check is always
> a good thing to have. Thank you very much.
>
> Best regards,
> Ivan

Sent via Deja.com http://www.deja.com/
Before you buy.

[Ivan Cibrario Bertolotti]

aven...@epita.fr wrote:
Hello,

thank you very much for your patient (when my bug reports
are wrong), quick and responsive (always)
support.

I would kill (well, almost... 8-) ) to have this same level of
support for other (much, much more expensive) software/hardware
products I purchased in the past.

With my best regards,
Ivan

> Hello
>
> This is definitely a bug.. :(
>
> Jean-Yves

--

[Joėl Bourquard]

Hi Ivan,

It's done. Yesterday, they were already aware about the bug.

Best Regards,
Joėl