t;-- MAY_USE(0x03e90001, 0)
<-- MAY_USE(0x03e90001, 1)
<-- MAY_USE(0x03e90001, 2)
<-- MAY_USE(0x03e90001, 3)
```
I have removed the rest of the output, it repeats "MAY_USE(0x03e90001, N)"
where N is an incrementing integer in what seems like an endless loop.
I was wondering if somebody had familiarity with this workspace, or what
might be causing this issue.
Thanks,
- Rowan Cannaday
What about excluding certain "error on warning" flags, except when
compiling in developer mode?
For example, make `CXXFLAGS=-Werror=maybe-uninitialized ./configure` the
default and force an error when configuring with `DEVELOP_WANTED=yes`.
- Rowan
On Fri, May 1, 2020 at 6:41 PM Dr. Jürgen Sauerm
Sorry for the late reply. I was out venturing for groceries, which has
turned into a significantly more stressful task than anticipated it would.
Thanks to responders who have provided simpler demonstrations of the effect.
- Rowan
On Thu, Apr 30, 2020, 3:30 PM Kacper Gutowski wrote:
> On Thu,
{{0~⍨⍵×⍳≢⍵} 0=47|⍵ ⍴ ⍵?⍵} 3600
10 45 179 183 185 203 234 299 326 353 516 609 764 766 807 916 946 976 1069
1135 1199 1217 1239 1249 1284 1443 1466 1532 1585 1605 1747 1841 1857 1926
1932 1957 2027 2048 2145 2147 2178 2179 2242 2372 2402 2474 2513 2563 2601
2644 2658 2698 2777 2807 2869 3000 30
It looks like the Cholesky decomposition article uses block matrix
notation, so it looks like it will be useful, just not for the reason I was
expecting. Thank you Ala'a.
- Rowan
On Wed, Apr 29, 2020 at 2:24 AM Rowan Cannaday wrote:
> I believe I've figured out what was conf
28, 2020 at 9:04 PM Ala'a Mohammad wrote:
> Hi,
>
> I assumed that same technique can be used for QR algorithm. check the
> algorithm section.
>
> Regards,
>
> Ala'a
>
> On Wed, Apr 29, 2020 at 12:54 AM Rowan Cannaday
> wrote:
>
>> Thank you. The C
ikipedia.org/wiki/QR_algorithm
<https://en.wikipedia.org/wiki/QR_algorithm>*
- Rowan
On Tue, Apr 28, 2020, 4:37 PM Ala'a Mohammad wrote:
> Hi,
>
> This may be related:
> https://www.dyalog.com/blog/2015/01/cholesky-decomposition/
>
> HiH.
>
> Regards,
>
> Ala&#
Is anybody familiar with the QR algorithm, or the notation used in Golub &
Van Loan's book "Matrix Computations"?
I am working my way towards building an APL linear algebra toolkit, and am
having trouble deciphering what some of the steps signify.
Any other communities that might be able to assis
nstrate what is happening, but I thought I'd reach out in
meantime to see if there is a workaround.
Thank you,
- Rowan
On Mon, Apr 13, 2020 at 6:32 PM Rowan Cannaday wrote:
> Another deficiency I've noticed in my above solution is that it does not
> converge for complex eigenval
Another deficiency I've noticed in my above solution is that it does not
converge for complex eigenvalues.
- Rowan
On Mon, Apr 13, 2020 at 6:07 PM Rowan Cannaday wrote:
> Here is a quick solution to the eigenvalue problem. Should be refined and
> extended to calculate eigenvect
Here is a quick solution to the eigenvalue problem. Should be refined and
extended to calculate eigenvectors.
{0~⍨∈⍵×∘.=⍨⍳≢⍵}({r+.×⍉qt⊣(qt r)←⌹[⎕CT]⍵ ;qt;r}⍣{(|⍺)≡|⍵}) 3 3 ⍴ ⍳9
16.11684397 ¯1.11684397 ¯3.625973215E¯16
As a quick note, my ⎕CT was 1E¯13, however it doesn't seem to zero out the
>
> On 4/13/20 4:57 PM, Rowan Cannaday wrote:
>
> "I added a
> new primitive in GNU APL which computes the QR factorization of a real
> or complex matrix"
>
> Looking through the source code I see the following in `src/LApack.hh`:
> ```
>// Compute QR
than
> a direct implementation in, say, C/C++. But that is the only drawback that
> I can currently see. Given the many advantages that a translation of LApack
> to APL could have, that would be my preferred option (= the last option in
> your "way forward" list below).
>
t;
> ⍨ operator: you are using it to avoid parentheses?
>
> Beside of restructuring code, avoiding duplicate code-fragment and
> parentheses, ⍨ has no "functional" impact?
>
> br & many thanks - Otto
>
> Rowan Cannaday writes:
>
> > addendum, f
I've been mulling over methods of bringing linear algebra methods into
gnu-apl, specifically eigenvalues & eigenvectors (leaving open the
possibility for more).
The canonical methods for this are from LAPACK (which was formerly a
compilation dependency).
Specifically:
dgeev
dsyev
zheev
addendum, forget an extra ⍳≢
{(⍵⍴⍨≢⍵)⊂⍨(≢⍵)↑⍺/⍳≢⍵}
Cheers.
On Sat, Apr 11, 2020 at 9:19 PM Rowan Cannaday wrote:
> quick attempt:
>
> 3 {(⍵⍴⍨≢⍵)⊂⍨(≢⍵)↑⍺/⍵} 1 2 3 4 5 6 7
> ┏→━━┓
> ┃┏→┓ ┏→┓ ┏→┓┃
> ┃
quick attempt:
3 {(⍵⍴⍨≢⍵)⊂⍨(≢⍵)↑⍺/⍵} 1 2 3 4 5 6 7
┏→━━┓
┃┏→┓ ┏→┓ ┏→┓┃
┃┃1 2 3┃ ┃4 5 6┃ ┃7┃┃
┃┗━┛ ┗━┛ ┗━┛┃
┗∊━━┛
On Sat, Apr 11, 2020 at 9:02 PM Otto Diesenbacher-Reinmüller <
ok...@diesenbacher.net> wrote:
> Hi APLers!
>
> as I had a look i
Heres the contents of /proc/meminfo on linux 5.4.0-2-amd64:
MemTotal:7833580 kB
MemFree: 949240 kB
MemAvailable: 997536 kB
Buffers: 108 kB
Cached: 1062792 kB
SwapCached: 532548 kB
Active: 3367828 kB
Inactive:2045752 kB
Active(anon):
just throw a syntax error as in:
>>
>> (+/10 10⍴⍳100)[2]←4
>> SYNTAX ERROR
>> (+/10 10⍴⍳100)[2]←4
>> ^^
>>
>> (Surely a syntax error is more appropriate th
hello y'all, hope everyone is staying safe.
x ← 1 (2 3 (4 5))
(⊃⊃x)[2;3;] ← 6 7
===
SEGMENTATION FAULT
-- Stack trace at main.cc:88
0x7F032D7BABBB __libc
Hello, receiving the following segfault when attempting to use dfn's with
all possible arguments: ⍵ ⍺ ⍹ ⍶ χ.
{⍵+1}{⍹ ⍶ χ+⍵}[1]{⍵+1} 1
===
SEGMENTATION FAULT
-- Stack trace at main.cc:88
-
1↑⍺} ltrav) list ⍳10
Cheers,
Rowan Cannaday
cally generated hast table that does lookups essentially in time
> O(1) per prefix.)
>
> This lookup table replaces the AST that you would have in a compiled
> language,
>
> Best regards,
> Jürgen Sauermann
>
>
> On 10/16/19 6:55 PM, Rowan Cannaday wrote:
>
> Thanks aga
stem functions and variables
> are not in )help,
> but I suppose extending file *src/Help.def* could easily add them.
>
>
> Compiling APL is IMHO a wrong path. Too many problems, too little gain.
>
> Best Regards,
> Jürgen Sauermann
>
>
> On 10/16/19 5:01 PM, Rowan Cannada
,
>>
>> fixed in *SVN 1191*.
>>
>> You should not be too enthusiastic, though, because the speed-ups that
>> can be achieved are somewhat disappointing. And due to that, I
>> haven't put too much effort into fixing faults (sometimes apl hangs
>> on a semap
Hello,
intrigued by the ability to parallelize APL, thought I'd try to test it:
`apl --cfg` followed by a line of '=' signs followed by `apl -q`:
configurable options:
-
ASSERT_LEVEL_WANTED=2
SECURITY_LEVEL_WANTED=0 (default)
APSERVER_PATH=/tmp/GNU-APL/APserver (
Working version!
dfs ← {⊃(⍶ {⍺(⍶ dfs ⍹)⍵} ⍹)⍨/⌽(⊂⍺ ⍶ ⍵),⍺ ⍹ ⍵}
0 ({⍺+0=≡⍵} dfs {⍺⊢(0=≡⍵)↓,⍵}) ((1 2)(3 4 (5 6 7)))
⍝ isomorphic to 'enlist'
⍬ {⍺,(0=≡⍵)↑⊂⍵} dfs {⍺⊢(0=≡⍵)↓,⍵} (1 2)(3 4(5 6 7))
On Thu, Oct 10, 2019 at 12:53 PM Rowan Cannaday wrote:
> btw not correct, but at least
btw not correct, but at least its valid apl, which is a start.
On Thu, Oct 10, 2019 at 12:34 PM Rowan Cannaday wrote:
> Thank you Jürgen.
>
> I found translating the late John Scholes APL videos on youtube from
> dyalog specific syntax to gnu-apl helpful.
>
> For example,
, you need to properly use parentheses like
> this:
>
> Z←(f⍤1) 2 3 4 5
>
> or, as a cleaner though non-standard alternative syntax, make *j* an axis
> argument:
>
> Z←f⍤[1] 2 3 4 5
>
> It is very difficult to get all that right, therefore I never use ⍤.
>
&g
somebody pointed out that the previous fn was failing because i wasn't
calling 'foo' w/ dyadic arguments in the inner dfn
On Wed, Oct 9, 2019 at 2:16 PM Rowan Cannaday wrote:
> interestingly enough this is ok:
> foo ← {⍬{⍺,(1+1↑⍵),(foo 1↓⍵)}⍣(~0=⍴⍵)⍵}
> f
this seems to be behaving much better:
foo ← {⍺,{((1+1↑⍵) foo 1↓⍵)}⍣(~0=⍴⍵)⊢⍵}
⍬ foo 1 2 3
2 3 4
5 4 3 foo 1 2 3
5 4 3 2 3 4
On Wed, Oct 9, 2019 at 3:40 PM Rowan Cannaday wrote:
> somebody pointed out that the previous fn was failing because i wasn't
> calling
making progress...
foo ← {{(1+1↑⍵),(foo 1↓⍵)}⍣(~0=⍴⍵)⍵}
foo 1 2 3
2 3 4
On Wed, Oct 9, 2019 at 12:47 PM Rowan Cannaday wrote:
> this is a slightly better way of writing my (still broken) accumulator:
> acc←{⍺{⍺ acc 1↑⍵}⍣(0=⍴⍵)⊢⍵}
>
> On Wed, Oct 9, 2019 at 12:40 PM Ro
)⍵
On Wed, Oct 9, 2019 at 1:29 PM Rowan Cannaday wrote:
> making progress...
>
> foo ← {{(1+1↑⍵),(foo 1↓⍵)}⍣(~0=⍴⍵)⍵}
> foo 1 2 3
> 2 3 4
>
> On Wed, Oct 9, 2019 at 12:47 PM Rowan Cannaday
> wrote:
>
>> this is a slightly better way of writing my (s
returns the input
vector ⍵.
Thoughts?
On Fri, Sep 27, 2019 at 3:44 PM Rowan Cannaday wrote:
> Hello y'all.
>
> I have been attempting to learn function composition & higher-order
> functions in gnu-apl, and how to use it to perform tree traversal.
>
> https:
this is a slightly better way of writing my (still broken) accumulator:
acc←{⍺{⍺ acc 1↑⍵}⍣(0=⍴⍵)⊢⍵}
On Wed, Oct 9, 2019 at 12:40 PM Rowan Cannaday wrote:
> Given a recursive factorial definition:
> fact←{{⍵ × fact ⍵-1}⍣(⍵>2)⊢1⌈⍵}
>
> [written by Kacper Gutowski in the 'Recu
Hello y'all.
I have been attempting to learn function composition & higher-order
functions in gnu-apl, and how to use it to perform tree traversal.
https://en.wikipedia.org/wiki/Function_composition_(computer_science)#APL
https://en.wikipedia.org/wiki/Higher-order_function#APL
https://rosettacode
n,
>
> my best guess is that your terminal setting is incorrect.
> 0xA0 is the unicode for non-braking space which has no APL significance
> and might come from a regular space (0x20) with the upper bit set.
>
> Best Regards,
> Jürgen
>
>
> On 9/27/19 9:11 PM, Rowan Can
Greetings y'all,
I wanted to ask a far more interesting question, but unfortunately I'm
having APL keyboard issues.
I frequently get the following error when using gnu-apl on OSX. I have
never had this problem on linux, using the APL x11 kbd.
> SYNTAX ERROR+ Tokenizer: No token for Unicode U+00A
(also a shout out to the creator if theyre on here)
I think I also saw a wasm target for gnu-apl somewhere (I believe from Dr.
Sauermann), but my brain is starting to shut off for the day...
On Thu, Jun 27, 2019 at 3:25 PM Rowan Cannaday wrote:
> @Alexey
>
> I actually floated this id
@Alexey
I actually floated this idea to the creator of "April" (an APL subset that
compiles to common lisp). I'll link the ticket below, but his response was
as follows:
"JS is not a very good language for implementing APL because it has no true
multidimensional arrays; only nested vectors. There
I think that'll work, thanks.
On Fri, Jun 21, 2019 at 11:30 AM Dr. Jürgen Sauermann <
mail@jürgen-sauermann.de> wrote:
> Hi Rowan,
>
> ⎕PW ?
>
> Not exactly sure what you are after.
>
> Jürgen
>
>
> On 6/21/19 5:27 PM, Rowan Cannaday wrote:
> >
Hello y'all,
Is there a simple way to either increase or eliminate the wrap width in the
REPL?
In vim this would be :set nowrap and using a pager would be `less -S`.
I have some nested objects I inspect w/ ` 4 ⎕CR` but the wrap causes the
lines to overflow and make it difficult to read.
- Rowan
due to the ; in the axis specification).
>
> Best Regards,
> Jürgen Sauermann
>
>
> On 6/20/19 2:05 PM, Rowan Cannaday wrote:
>
> Not a normal use-case but I thought y'all might be interested...
>
> z ← ''
> z ← z,⊂2 2 ⍴ 1
> ⊃,[;2]/z (3 3 ⍴ 2)
&g
Not a normal use-case but I thought y'all might be interested...
z ← ''
z ← z,⊂2 2 ⍴ 1
⊃,[;2]/z (3 3 ⍴ 2)
SEGMENTATION FAULT
-- Stack trace at main.cc:88
0x7F789
Works great :)
- Rowan
On 6/5/19, Rowan Cannaday wrote:
> Hah, I hadn't fixed it yet - but I tracked down the same issue in the
> last few minutes. There was no init_macros in libapl. I was wondering
> why ALL the macros had null pointers, but it would have taken me a bit
> lo
missing initialization of the Macro subsystem in libapl.cc.
>
> Macros were introduced after libapl, so I forgot to add that initialization.
>
> Now Erlang works again (of course only on my machine).
>
> Best Regards,
> Jürgen Sauermann
>
>
> On 6/4/19 11:04 PM, Row
n the near future.
- Rowan
On Tue, Jun 4, 2019, 4:32 PM Dr. Jürgen Sauermann
wrote:
>
> Hi Rowan,
>
> see below.
>
> BR, Jürgen
>
>
> On 6/4/19 5:00 PM, Rowan Cannaday wrote:
>>
>> Hello again,
>>
>> Trying out the erlang/APL interface (its building
Hello again,
Trying out the erlang/APL interface (its building now!), running into
a few issues.
BTW I'm pretty new to FOSS mailing lists so if in the future you'd
prefer I'd start these as different threads, or use a different syntax
for distinguishing shell output just let me know.
First issue
d if that works then I am
> planning to release GNU APL 1.8.
>
> Best Rgeards,
> Jürgen
>
>
>
> On 6/2/19 10:50 PM, Dr. Jürgen Sauermann wrote:
>
> Hi Rowan,
>
> thanks for reporting this. Supposedly fixed in *SVN 1162*.
>
> Best Regards,
> /// Jürgen
>
>
est[d] = utf8[d];
~~^
cc1plus: all warnings being treated as errors
make[3]: *** [Makefile:2119: libapl_la-libapl.lo] Error 1
```
On Sun, Jun 2, 2019 at 8:50 PM Dr. Jürgen Sauermann <
mail@jürgen-sauermann.de> wrote:
> Hi Rowan,
>
> than
Hello!
I am trying to compile libapl, and am running into the following error:
```
libtool: compile: g++ -DHAVE_CONFIG_H -I. -I.. -Wall -I sql
-Wold-style-cast -Werror -I/usr/include -I/usr/include/postgresql -rdynamic
-g -O2 -MT libapl_la-libapl.lo -MD -MP -MF .deps/libapl_la-libapl.Tpo -c
liba
51 matches
Mail list logo
