QuadFunction.cc:1425
⎕es 'DOMAIN ERROR'
error.error_code:'1' at QuadFunction.cc:1425
DOMAIN ERROR
⎕ES 'DOMAIN ERROR'
^
That is a language extension specific to MicroAPL. It is not part of
either ISO Extended APL or IBM APL 2, which are the references upon
which GNU APL is based.
The README-7-more-info file in the GNU APL distribution contains links
to the reference documents.
On Wed, 2014-05-28 at 20:15 -0700, A
quad-ET should always reflect the error that caused the top suspension
on the SI. When the SI is empty, quad-ET should yield 0 0.
⎕et
0 0
foo
VALUE ERROR
foo
^
⎕et
3 1
1÷0
DOMAIN ERROR
1÷0
^^
⎕et
5 4
→
⎕et
5 4
^^ NOTE: should be 3
No, sqrt(-1) is equal to i. 0J1 is how i is represented in APL.
Regards,
Elias
On 29 May 2014 11:12, "Akiva Avraham" wrote:
> ¯1*0.5
>
> 0J1 ⍝ Shouldn't this be a Domain Error?
> http://www.microapl.co.uk/apl_help/ch_020_020_200.htm
>
> --
>
I am not experienced enough in gnu-APL to determine whether you intend
for this type of expression to be executable, so I just bring it to your
attention.
http://www.microapl.co.uk/apl_help/ch_020_020_310.htm
If the arguments contain object (or class) references, the elements are
considered equal
¯1*0.5
0J1 ? Shouldn't this be a Domain Error?
http://www.microapl.co.uk/apl_help/ch_020_020_200.htm
--
⎕fx 'bar' '1'
bar
⎕ec 'bar'
1
unexpected result tag 0x56020011 in eoc
==
Assertion failed: 0
in Function: eoc
in file: QuadFunction.cc:1305
Call stack:
On Wed, 2014-05-28 at 18:11 -0500, Blake McBride wrote:
> Dear David,
>
>
> Thanks for the idiom reference!
>
You're welcome. :)
>
> With a lot of respect, I think you missed my point. The APL 1 idiom
> works as you want and expect in APL 1, APL 2, and GNU APL. Look
> closely at my example.
Dear David,
Thanks for the idiom reference!
With a lot of respect, I think you missed my point. The APL 1 idiom works
as you want and expect in APL 1, APL 2, and GNU APL. Look closely at my
example.
On the other hand, if it is I who is missing the point (quite possible),
please explain.
Thank
Yes, it is interesting.
I now understand that ⍎BS/'expr' is an APL 1 idiom that became obsolete
with APL 2.
The APL 2 IDIOMS Library[*] offers this instead:
fn¨BS/⊂A ⍝ Conditional execution of monadic function.
As-is, this is not as general as APL 1's ⍎BS/'expr', but it's still
useful.
With a h
http://i.imgur.com/7vMrDan.jpg
?(3⍴5) (2 3⍴10)
DOMAIN ERROR
?(3⍴5)(2 3⍴10)
^^
(3⍴5) (2 3⍴10)
5 5 5 10 10 10
10 10 10
?(3⍴5)
5 3 4
?(2 3⍴10)
9 6 5
5 10 7
?(3⍴5) 30
DOMAIN ERROR
Hello David,
FYI, Figure 8, on page 49 of IBM's "APL2 Programming: Language
Reference" (2nd ed, Feb 1994) says an empty character vector displays as
a blank line.
Regards,
Fred
On Tue, 2014-05-27 at 12:32 -0700, David B. Lamkins wrote:
> I find this confusing and counterintuitive.
>
>
Additional information - ∆RD calls no user functions and is not called by
any other function. So, ∆RD is not being called in because it is used by
another function.
Thanks.
Blake
On Wed, May 28, 2014 at 1:47 PM, Blake McBride wrote:
> )CLEAR
> CLEAR WS
> )COPY Devices ∆SS
> SAVE
)CLEAR
CLEAR WS
)COPY Devices ∆SS
SAVED 2014-5-27 2:53:43 (GMT-5)
)FNS
∆SS
)COPY Devices ∆LP
SAVED 2014-5-27 2:53:43 (GMT-5)
)FNS
∆LP ∆LPH∆RD ∆SS
Where did ∆LPH and ∆RD come from? (Devices is a workspace and not a dump
file.)
Thanks.
Blake
Dear David,
Interesting observation. More interesting is how GNU APL correctly handles
it. i.e.:
∇test[⎕]∇
∇
[0] test
[1] '1'
[2] ⍎0/'test2'
[3] '2'
[4] ''
[5] '3'
[6] ⍎1/'test2'
[7] '4'
∇
∇test2[⎕]∇
∇
[0] test2
[1] 'test2'
∇
test
1
2
3
tes
Hi Blake,
I have fixed the WSID, see SVN 297.
I also aligned some messages.
If both an .xml and a .apl file with the same name, a warning is printed and
the .xml file is loaded (for backward compatbility).
I found the )SAVE message a bit sparse so I took the freedom to be more
verbose on )DUMP
Now that I can see how printing '' as CR affects a larger program, I've
gotta say that I'm stumped as to how this behavior can be considered
usable.
Specifically, it's now (unless there's a trick I haven't yet discovered)
difficult (see below) to silently (i.e. without affecting the display)
execu
∇unbox;⎕'∇
Workspace::more_error(): 'UserFunction.cc:179' at Nabla.cc:554
DEFN ERROR+
∇unbox;⎕'∇
^
Could probably do without the Workspace:: line.
Thanks.
Blake
The "significantly worse" problem is now fixed, but the original problem (May
27, 2014 at 9:23 PM) remains.
Thanks.
Blake
On Tue, May 27, 2014 at 10:16 PM, Blake McBride wrote:
> I just updated to the latest GNU APL. The problem is significantly worse.
> Now )LOAD doesn't work either. The
)CLEAR
CLEAR WS
1 2 3[2]
RANK ERROR
1 2 3[2]
^^
1 2 3[2 2]
RANK ERROR
1 2 3[2 2]
^^
x←1 2 3
x[2 2]
2 2
Of course, they should all work as if they were assigned to a variable
first.
Blake
This issue is working very strange now. When I first tried it it still
didn't execute ⎕LX. Then I found this while trying to create a test for
you:
)CLEAR
CLEAR WS
∇test
[1] 'Hello World'
[2] ∇
⎕LX←'test'
)dump test
DUMPED WORKSPACE 'test'
TO FILE '/home/blake/workspaces
Looks good. Thanks!
On Tue, May 27, 2014 at 11:43 AM, Juergen Sauermann <
juergen.sauerm...@t-online.de> wrote:
> Hi Blake,
>
> thanks, fixed in SVN 294.
>
> /// Jürgen
>
>
>
> On 05/27/2014 04:59 AM, Blake McBride wrote:
>
>)CLEAR
> CLEAR WS
> ⎕IO
> 1
> 'abc'⍳'a'
> 1
>
Looks good. Thanks!
On Tue, May 27, 2014 at 1:07 PM, Juergen Sauermann <
juergen.sauerm...@t-online.de> wrote:
> Hi Blake,
>
> thanks, fixed in SVN 295.
>
> /// Jürgen
>
>
>
> On 05/26/2014 11:01 PM, Blake McBride wrote:
>
>)CLEAR
> CLEAR WS
> ⎕IO
> 1
> a←b←5 5⍴⍳25
>
Looks good. Thanks!
On Wed, May 28, 2014 at 10:44 AM, Juergen Sauermann <
juergen.sauerm...@t-online.de> wrote:
> Hi Blake,
>
> thanks, fixed in SVN 296.
>
> /// Jürgen
>
>
>
> On 05/27/2014 12:53 AM, Blake McBride wrote:
>
>)CLEAR
> CLEAR WS
> ⎕IO
> 1
> (4 5)[1]≥12
> 0
>
Hi David,
sorry, no idea where it went. Will be in SVN 297.
/// Jürgen
On 05/28/2014 06:59 AM, David B. Lamkins wrote:
Jürgen,
I submitted this patch in a slightly different form once before; it
didn't make it into GNU APL.
After having lived with the patch for a while on my own system, I'v
Hi David,
of course. It came back in SVN 296.
/// Jürgen
On 05/27/2014 08:18 PM, David B. Lamkins wrote:
Thanks. One more thing: quad-IO should be on that list.
See UserFunction.cc, around line 252.
On Tue, 2014-05-27 at 18:06 +0200, Juergen Sauermann wrote:
Hi,
thanks, should be fixed in
Hi Blake,
thanks, fixed in SVN 296.
/// Jürgen
On 05/27/2014 12:53 AM, Blake McBride wrote:
)CLEAR
CLEAR WS
⎕IO
1
(4 5)[1]≥12
0
(⎕TS[4 5])[1]≥12
VALENCE ERROR
(⎕TS[4 5])[1]≥12
^
Additional facts:
x←4 5
x[1]≥12
0
(x)[1]≥12
0
(x[⍳⍴x])[1]≥12
VALENCE ERROR
(x[⍳⍴x])[1]≥12
^
⍴x
2
x[1 2]
4 5
x[1 2][1]≥12
VALENCE ERROR
x[1 2][1]≥12
^
x[1 2][1]
4
⍴x[1 2][1]≥12
VALENCE ERROR
⍴x[1
I am not sure I disagree with David, I am just arguing for consistency.
Thanks!
Blake
On Wed, May 28, 2014 at 5:02 AM, Juergen Sauermann <
juergen.sauerm...@t-online.de> wrote:
> Hi,
>
> I agree with David that this is counter-intuitive, but I will
> make it IBM APL2 compatible.
>
> /// Jürge
Hi,
I agree with David that this is counter-intuitive, but I will
make it IBM APL2 compatible.
/// Jürgen
On 05/27/2014 11:37 PM, David B. Lamkins wrote:
On Tue, 2014-05-27 at 15:19 -0500, Blake McBride wrote:
Dear David,
First, I assure you, this is how it works.
I'm certainly not arguin
30 matches
Mail list logo
