Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread robert therriault
Thank you Henry, I reported it here. http://code.jsoftware.com/wiki/System/Interpreter/Bugs/Errors#Error_in_float_representation_between_1e15_and_1e19 Let me know if you had somewhere else in mind. Cheers, bob > On Aug 9, 2017, at 9:05 AM, Henry Rich wrote: > > This is

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Skip Cave
Markus, Actually, yes, I looked at using extended precision, and here's what happened: v=:2%(3r19-%x:1 2 3 245 246 247 248) ​v _19r8 _76r13 _57r5 4655r358 9348r719 13 9424r725​ fp=. * * 1|| fp v _3r8 _11r13 _2r5 1r358 1r719 0 724r725 ip=: * * <.@| ip v _2 _5 _11 13 13 13 12 So our

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Roger Shepherd
Continued fractions come immediately to mind as an application, per http://code.jsoftware.com/wiki/Essays/Continued_Fractions Suggestions here might be suitable to remove restrictions mentioned in that essay. Continued fractions themselves are applied in at least the ways stated in

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread robert therriault
I was more surprised by the type change within a certain numeric range. That just seemed a little...odd. I guess that even though it feels strange to have a constant change type just by being entered, it's not really any different than these examples. (; datatype) 3j0 ┌─┬───┐

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread Vijay Lulla
Though not quite the same issue but is this related to the behavior listed at http://geocar.sdf1.org/numbers.html ? On Wed, Aug 9, 2017 at 2:17 PM, robert therriault wrote: > I was more surprised by the type change within a certain numeric range. > > That just seemed a

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Schmidt-Gröttrup , Markus
Skip, have you considered to use extended precision? In your example only small fraction are used, and the methods discussed would apply as well without precision problems. v=:2%(3r19-%x:1 2 3 245 246 247 248) I haven’t read the whole thread, so sorry if this contribution doesn’t fit. Markus

Re: [Jprogramming] Boxed verbs as alternate gerunds

2017-08-09 Thread Henry Rich
:: 0 Is a j8.06 feature. Henry Rich On Aug 8, 2017 23:42, "Jose Mario Quintana" wrote: > "I still like the @.] test." > > What am I doing wrong? I have tried several times and I keep getting, > >JVERSION > Engine: j805/j64/windows > Release:

Re: [Jprogramming] Boxed verbs as alternate gerunds

2017-08-09 Thread 'Pascal Jasmin' via Programming
:: noun is equivalent to :: (noun"_)? From: Henry Rich To: Programming forum Sent: Wednesday, August 9, 2017 5:25 AM Subject: Re: [Jprogramming] Boxed verbs as alternate gerunds :: 0 Is a j8.06 feature. Henry Rich On Aug 8, 2017

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Louis de Forcrand
Why so? Did I miss a non-scalar verb somewhere? Thanks, Louis > On 9 Aug 2017, at 14:47, Raul Miller wrote: > > I think you have assumed that the user will use these with "0 or only > on atomic numbers? > > Thanks, > > -- > Raul > > >> On Wed, Aug 9, 2017 at 3:39

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Raul Miller
I think you have assumed that the user will use these with "0 or only on atomic numbers? Thanks, -- Raul On Wed, Aug 9, 2017 at 3:39 PM, Louis de Forcrand wrote: > A few handy tests which are good to know: > > N=: GI *. 1 0 e.~ * NB. naturals > Z=: GI *. R NB. integers >

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread Henry Rich
Quite right. Henry Rich On Aug 9, 2017 20:46, "Raul Miller" wrote: > Well, since it's encoded as an integer (which I would have noticed if > I had read Bob Therriault's original post more closely), and not [like > I was thinking] a float, I agree that dropping the .3 is

[Jprogramming] J for iOS

2017-08-09 Thread Eric Iverson
An updated J is available at the Apple App Store. This fixes a few minor problems with the release on July 4th, adds a bell and whistle, and provides a new way to move scripts and data between your device and desktop called jj (j to j). See http://code.jsoftware.com/wiki/Guides/iOS for more

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread Raul Miller
Well, since it's encoded as an integer (which I would have noticed if I had read Bob Therriault's original post more closely), and not [like I was thinking] a float, I agree that dropping the .3 is better than adding a 1. That said, I guess we also should not object too loudly if

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread Henry Rich
Surely integer 999...9 is a better value than 1000...0 . Henry Rich On Aug 9, 2017 18:33, "Raul Miller" wrote: > It's not a bug, it's an artifact of the 64 bit floating point standard. > >2 ^. > 53.1508 > >

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Louis de Forcrand
A few handy tests which are good to know: N=: GI *. 1 0 e.~ * NB. naturals Z=: GI *. R NB. integers R=: = + NB. reals C_R=: + = - NB. pure imaginaries (C-.R) GI=: = <. NB. gaussian integers These were made to accept any J number. They could be optimised if one knows that they are working only

Re: [Jprogramming] J for iOS

2017-08-09 Thread Joey K Tuttle
Eric, Bravo on the keyboard change and the repeating keys (just really nice to have them!) I think your setup for j to j is exciting news too, but am disappointed after starting the new release getting - jj 0 |value error: jj The suggested scripts in desktop JHS seem to work in setting up

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Tom Arneson
No space saving for 64 bit integers. -Original Message- From: Programming [mailto:programming-boun...@forums.jsoftware.com] On Behalf Of Skip Cave Sent: Tuesday, August 8, 2017 23:48 To: programm...@jsoftware.com Subject: Re: [Jprogramming] Fractional parts Don, You're right. Your

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread Raul Miller
It's not a bug, it's an artifact of the 64 bit floating point standard. 2 ^. 53.1508 https://en.wikipedia.org/wiki/IEEE_754#Basic_and_interchange_formats The binary64 format has 53 binary digits or 15.95 decimal digits. This means ".16#'9' cannot be represented exactly using

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Skip Cave
Martin, The original problem I was working on was a post on Quora ( https://goo.gl/NrZde2). I use these Quora math questions to help sharpen my J skills. I try to see if I can "brute force" the solutions using J, while most other posters try to solve these things by algebraic manipulation. My

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread Henry Rich
This is a bug, since 999...9.3 should become 999...9 rather than 100...0. I'm away from home now, but I think what's happening is this: 999...9 is converted to integer . is encountered and turns it to float It's rounded to the nearest float which is 100...0 As a final step the JE checks to

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Skip Cave
Bo said: v=(<.v)+(1|v) NB. number = integer part + fractional part? Yes! True. However, if ]v=:2%(3r19-%1 2 3 245 246 247 248) _2.375 _5.84615 _11.4 13.0028 13.0014 13 12.9986 And we define the fractional part of v your way: 1|v 0.625 0.153846 0.6 0.0027933 0.00139082 0 0.998621 I

Re: [Jprogramming] J for iOS

2017-08-09 Thread Eric Iverson
Strange. As we always say, it works for me. What is the value of VERSION_j_ ? It should be '701.1 2'. But that has to be right as you indicate the backspace key repeats if you hold it down. Is there any chance you are trying jj in the previous j701 version? I also just realized that I probably

Re: [Jprogramming] J for iOS

2017-08-09 Thread Joey K Tuttle
VERSION_j_ 701.1 2 jj |value error: jj explicitly doing load '~system/util/jj.ijs' NB. executes with no complaint (and, of course, no result shown) NB. but then when (as suggested in http://code.jsoftware.com/wiki/Guides/iOS ) I do - jjset '10.1.1.23:65001 pswd' NB. for my

Re: [Jprogramming] J for iOS

2017-08-09 Thread Joey K Tuttle
Actually not too tedious, the copy paste across devices combined with your jsetcb '... avoided having to retype everything. Some added thoughts about this - I started from a clean start (if not install) by killing the J task then restarting it. Then, I did - untable >nl_jj_ &.> 0 3 TAR

Re: [Jprogramming] J for iOS

2017-08-09 Thread J. Patrick Harrington
Eric, I get the same results as Joey. In addition, names_jj_ 0 PASS SERVER TAR URL posttemplate and PASS has the password I entered and SERVER has 192.168.0.19:65001 Patrick On Wed, 9 Aug 2017, Joey K Tuttle wrote: VERSION_j_ 701.1 2 jj |value error: jj

Re: [Jprogramming] J for iOS

2017-08-09 Thread Eric Iverson
Patrick, You give 192.168.0.19:65001 which is exactly what the wiki page gives by way of an example. This could be true on your system, but is unlikely, The jjset arg should be the one reported in the init_jjserver_ on the JHS server. However. the other problems you and Joey report remains.

Re: [Jprogramming] J for iOS

2017-08-09 Thread Eric Iverson
Joey, Patrick, It looks as if you have the new j701 binary, but don't have the new data bundle (scripts). You are seeing the scripts from the earlier j701 release before the jj stuff was cleaned up. I think the bundle at the app store is good based on the fact that when I do a clean install I

Re: [Jprogramming] J for iOS

2017-08-09 Thread Eric Iverson
Joey, I agree with you re / on keyboard. In next version I will squeeze in a few more 'important' keys. My current set is / and ~ On Wed, Aug 9, 2017 at 7:55 PM, Eric Iverson wrote: > Joey, Patrick, > > It looks as if you have the new j701 binary, but don't have the

Re: [Jprogramming] J for iOS

2017-08-09 Thread Eric Iverson
Joey, This will be tedious, but could you please do: jjget_jj_ and give me the the first 3 or 3 lines (at least in summary). On Wed, Aug 9, 2017 at 7:39 PM, Eric Iverson wrote: > Patrick, > > You give 192.168.0.19:65001 which is exactly what the wiki page gives

Re: [Jprogramming] J for iOS

2017-08-09 Thread J. Patrick Harrington
Oops -- I just copied from the wiki: it looked looked the same. The JHS value actually is '192.168.1.2:65001 -- But using the correct value doesn't fix the problem. On Wed, 9 Aug 2017, Eric Iverson wrote: Patrick, You give 192.168.0.19:65001 which is exactly what the wiki page gives by way

Re: [Jprogramming] J for iOS

2017-08-09 Thread Joey K Tuttle
Yes, that does seem to be the case. I suspect you are right - in the case of an app like a game, the app remembers all of my history through all updates. I have often thought that you could define a "workspace" that would keep the current session as a history, then when restarting the app

Re: [Jprogramming] Fractional parts

2017-08-09 Thread 'Bo Jacoby' via Programming
v=(<.v)+(1|v) NB. number = integer part + fractional part? Yes! 1 1 1 1 1 1 1 Den 7:00 onsdag den 9. august 2017 skrev Don Guinn : True for 32 bit integers, but for 64 bit j integers and floats are the same size. On Aug 8, 2017 10:48 PM, "Skip Cave"

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread 'Pascal Jasmin' via Programming
in j806,    .31 probably the j805 behaviour is preferred.  If only for consistency.  But there may be a good reason for change.  From: robert therriault To: Programming forum Sent: Wednesday, August 9,

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread robert therriault
Hi Pascal, I see the same behaviour in j806 as j805. Do you see something different? JVERSION Engine: j806/j64avx/darwin Beta-4: commercial/2017-06-27T12:55:06 Library: 8.06.03 Qt IDE: 1.5.3/5.6.2 Platform: Darwin 64 Installer: J806 install InstallPath: /users/bobtherriault/j64-806 Contact:

Re: [Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread bill lam
I think this is the difference between 32 and 64-bit, 9!:14'' j602/2008-03-03/16:45 3!:0[ .3 4 In J32 a.i. 2 fc .3 0 128 224 55 121 195 65 67 a.i. 2 fc 1e16 0 128 224 55 121 195 65 67 the number has the same bit pattern as 1e16 (an integer) which can

Re: [Jprogramming] Fractional parts

2017-08-09 Thread bill lam
v=. 0 1e_40 _1e_40 v=(<.v)+(1|v) 1 1 0 each version of integer part needs a matching fractional part. On Aug 9, 2017 10:06 PM, "'Bo Jacoby' via Programming" < programm...@jsoftware.com> wrote: v=(<.v)+(1|v) NB. number = integer part + fractional part? Yes! 1 1 1 1 1 1 1 Den 7:00

[Jprogramming] Integer-floating type change for large numbers in j805 and j806

2017-08-09 Thread robert therriault
I am guessing that the following has something to do with precision of large numbers in j805 and is also true for j806. (; datatype) 999.3 ┌┬┐ │1e15│floating│ └┴┘ (; datatype) .3 ┌─┬───┐ │1│integer│

Re: [Jprogramming] Boxed verbs as alternate gerunds

2017-08-09 Thread Jose Mario Quintana
I guess somehow we both miss this: System/ReleaseNotes/J806 New features: u@n (where n is a noun) is now allowed, and executes u on the value of n, ignoring any arguments (equivalent to u@(n"_)) u :: n (where n is a noun) is now allowed, and gives the result n if u fails (equivalent to u ::

Re: [Jprogramming] Boxed verbs as alternate gerunds

2017-08-09 Thread Bill
Oh I was wrong, thank you for pointing it out. Sent from my iPhone On 9 Aug, 2017, at 6:28 AM, Jose Mario Quintana wrote: > Inline comments follow... > >> I would say neither gerundYN or isgerund is correct, they should report > value error. Even J interpreter

Re: [Jprogramming] Boxed verbs as alternate gerunds

2017-08-09 Thread Jose Mario Quintana
:D It is so natural that I did not even notice it. I guess I was already ready for it; thank you. On Wed, Aug 9, 2017 at 5:25 AM, Henry Rich wrote: > :: 0 > > Is a j8.06 feature. > > Henry Rich > > On Aug 8, 2017 23:42, "Jose Mario Quintana"

Re: [Jprogramming] Fractional parts

2017-08-09 Thread Martin Kreuzer
From what I've gathered so far is, that people seem to not mind that much, when extracting the fractional part from a (negative) float, they use (1&|) _8.11 0.89 or (1