For me manuals have always been the first resort. When you read it you
know more than anyone else about the language, because noone else read
it :) The only exception is the J manual, which I find totally
unreadable. /Erling
Den 2017-12-13 kl. 23:46, skrev Jose Mario Quintana:
I am surprised
Hi all !
More generally stated, for non-scalar verbs:
u(terminal) <=> u(manual page)"u
Rank is responsible for adding new dimensions you usually don't expect
or want and adding fill you usually don't expect or want. It has these
very strange features when one argument or both are empty which
Hi all !
My main complaint about these four helper programs is that they are not
defined as concepts to which you can associate the functionality of
them. The functionality might be described somewhere, but how do you
understand this functionality if there is no concept to associate it with?
Hi,
I am writing a mini-J interpreter.
Lexer appears to work fine. Dyad execution appears to be working (with
rank, frame, cell, padding, implemented.)
Is there any order to
https://github.com/jsoftware/jsource/tree/master/test ?
I'm looking for a resource where it is an ever increasin
Typically the tests are run in directory listing order. But they each
can run independently.
Since most primitives are independent of each other, I don't think a
case could be made for a different system of ordering.
Thanks,
--
Raul
On Thu, Dec 14, 2017 at 6:27 AM, TongKe Xue wrote:
> Hi,
>
I used xUnit in my development environment. Most tests are done directly
against the developed pieces of code, very few from the user interface
in the form of JWithATwist code. My tests have some degree of generality
allowing refactoring, but are partly white box, tests of the actual code
writt
Because tables are a common data structure, is there interest in
shorthand indexing for them?
(<+.2j5) { i.8 8
21
Note 'concept'
let the index be
(<2j5) { rank 2 or higher
or even
2j5 { rank 2 or higher
(<+.2j5){ i.3#9 NB. -: 2j5 { i.3#9
207 208 209 210 211 212 213 214 21
Have you looked at (<"1@[ { ]) which is backed by special code?
v0`(<"1@[)`v2} has no special code, but that might be worth doing.
Henry Rich
On 12/14/2017 12:37 PM, David Lambert wrote:
Because tables are a common data structure, is there interest in
shorthand indexing for them?
(<+.2j5
Are there any plans for updating the Dictionary?
On Wed, Dec 13, 2017 at 1:16 PM, Henry Rich wrote:
> More details at
>
> http://code.jsoftware.com/wiki/Vocabulary/EmptyArguments
>
> I consider the treatment of empties to be as fixed as anything in J. Its
> omission from the Dictionary is just
We will get to it when there is enough critical material. We aren't there
yet.
For now, the community gets far more bang for our buck with effort on NuVoc
and the wiki.
On Thu, Dec 14, 2017 at 1:03 PM, Jose Mario Quintana <
[email protected]> wrote:
> Are there any plans for updating
Complex number indexing for rank 2 arrays sounds cute, but it has a
couple issues:
(0) real numbers are a subset of complex, but the shape of real number
indexing would become different from the shape of complex number
indexing
(1j1) computed indexes (e.g. i.) do not get simpler computations this
And the forum archives are also a treasure trove.
On Thu, Dec 14, 2017 at 1:43 PM, Eric Iverson
wrote:
> We will get to it when there is enough critical material. We aren't there
> yet.
>
> For now, the community gets far more bang for our buck with effort on NuVoc
> and the wiki.
>
> On Thu, De
!! That's a GREAT idea! Having to box the damned indexes has been a
recurring trouble-spot for me, and when it matters for performance
workarounds are hard. I have never had a case where I was tempted to
have a multidimensional array of atomic indexes for amend, and of course
if one came up
I'm not following your thought process on the special case for
complementary indexing. Can you show me where this helps?
Thanks,
--
Raul
On Thu, Dec 14, 2017 at 2:29 PM, Henry Rich wrote:
> !! That's a GREAT idea! Having to box the damned indexes has been a
> recurring trouble-spot for me,
That is fair enough. However, when the time comes I would suggest
considering to mention in III. Definitions ( http://www.jsoftware.com/
help/dictionary/dict3.htm ) the fact that the definitions of all the verbs
in the body of the Dictionary specify their behavior only on cells of their
correspond
I want to disallow complementary indexing. But I want to allow
subarrays, to wit:
0 1} NB. selectors are 0 and 1, two items
(2 2 $ 1 2 3 4)} NB. selectors are (<1 2) and (<3 4), two _2-cells
(,: 1 2;3 4)} NB. selector is < (<1 2) , <3 4) i.e. a subarray
Henry Rich
On 12/14/2017 2:37 P
Ok, but why do you want to disallow complementary indexing? (What
problem does that solve?)
Thanks,
--
Raul
On Thu, Dec 14, 2017 at 3:31 PM, Henry Rich wrote:
> I want to disallow complementary indexing. But I want to allow subarrays,
> to wit:
>
> 0 1} NB. selectors are 0 and 1, two items
I thought that it wasn't supported, but I see that I was wrong; so no
exception need be made for it.
Henry Rich
On 12/14/2017 3:39 PM, Raul Miller wrote:
Ok, but why do you want to disallow complementary indexing? (What
problem does that solve?)
Thanks,
---
This email has been checked for
While we're at it, we could define m of rank 3 (shape a,2,b) to specify
a (set of) subarrays, with the first list giving the top-left corner and
the second list giving the shape of the subarray.
Henry Rich
On 12/14/2017 3:39 PM, Raul Miller wrote:
Ok, but why do you want to disallow complemen
Here be dragons, which even if not dangerous are opportunities which can
easily be missed (and have been missed). I'll send a more detailed
response when I get a break in the action.
--
For information about J forums see http://ww
Varying sized shapes gets into fill issues.
Perhaps with a good solution, if there were frequently used
applications for such things.
Thanks,
--
Raul
On Thu, Dec 14, 2017 at 5:37 PM, Henry Rich wrote:
> While we're at it, we could define m of rank 3 (shape a,2,b) to specify a
> (set of) suba
Fill caused by selectors of amend is a domain error.
Henry Rich
On 12/14/2017 7:45 PM, Raul Miller wrote:
Varying sized shapes gets into fill issues.
Perhaps with a good solution, if there were frequently used
applications for such things.
Thanks,
---
This email has been checked for virus
22 matches
Mail list logo
