Re: [Jprogramming] Calculus primitives are removed

NuVoc. Henry Rich On 8/27/2019 8:18 PM, 'Jon Hough' via Programming wrote: Thanks for the explanation. I will give it a try. By the way, there is some mention of a new adverb primitive u. Where is this documented? On Tuesday, August 27, 2019, 10:20:22 AM GMT+9, Henry Rich wrote:

Re: [Jprogramming] Calculus primitives are removed

Thanks for the explanation. I will give it a try. By the way, there is some mention of a new adverb primitive u. Where is this documented? On Tuesday, August 27, 2019, 10:20:22 AM GMT+9, Henry Rich wrote: Backwards compatibility is important, but not the only consideration. 1. Calc

Re: [Jprogramming] j901 beta-h available

after base library is updated, it needs to restart J session or load pacman again to let the new je_update to take effect. The base library version is 9.0.7 now. load'pacman' 'update'jpkg'' 'upgrade'jpkg'' load'pacman' je_update_jpacman_'' at any rate, the latest JE binaries can be downloaded fro

Re: [Jprogramming] j901 beta-h available

Eric, That worked - I had, in fact, neglected to update the base library. Now, after the second run of je_update_jpacman_'' I get - 9!:14'' j901/j64avx/darwin/beta-h/commercial/www.jsoftware.com/2019-08-26T15:19:32 and the impressive timing improvements came back. Maybe including a specifi

Re: [Jprogramming] J 9.01 Beta-h (9!:41) side effect

I haven't written any Fold code for real use yet - just testcases. Henry Rich On 8/27/2019 2:53 PM, John Baker wrote: Hi Henry, The jod startup code always sets (9!:41) 0. The task addon loads and works as expected if I do it first. I suspect something is different in locale searching. B

Re: [Jprogramming] j901 beta-h available

You guys are just showing off! JVERSION |value error: JVERSION 2!:7'' |domain error | 2!:7'' (J701 - IOS) > On Aug 27, 2019, at 2:40 PM, 'Pascal Jasmin' via Programming > wrote: > > 2!:7 '' > > avx avx2 -- For in

Re: [Jprogramming] Document Object Model

Free is important! > On Aug 27, 2019, at 2:24 PM, Raul Miller wrote: > > I just meant to refer to the PE File Browser, which they offer for free. > > I have not tried their other products. > > FYI, > > -- > Raul > > On Tue, Aug 27, 2019 at 1:46 PM 'Jim Russell' via Programming > wrote: >>

Re: [Jprogramming] J 9.01 Beta-h (9!:41) side effect

Hi Henry, The jod startup code always sets (9!:41) 0. The task addon loads and works as expected if I do it first. I suspect something is different in locale searching. By making a slight change I can make JOD work in 9,01 h but I don't understand why flipping the two lines matters. I will ke

Re: [Jprogramming] j901 beta-h available

D'oh Thanks Pascal and Raul, It was the avx on the first line that I was missing. Still I am not seeing a marked increase in performance for avx versus non avx JVERSION Engine: j807/j64/darwin Release-d: commercial/2019-03-18T16:07:14 Library: 8.07.26 Qt IDE: 1.7.9/5.9.6 Platform: Darwin 64

Re: [Jprogramming] j901 beta-h available

I think 2!:7 tells what your machine is capable of, not what your J version will do. Henry Rich On 8/27/2019 2:40 PM, Raul Miller wrote: On Tue, Aug 27, 2019 at 2:37 PM 'robert therriault' via Programming wrote: What should I be looking for to know whether avx is in operation? Is it 2!:7 ''?

Re: [Jprogramming] j901 beta-h available

On Tue, Aug 27, 2019 at 2:37 PM 'robert therriault' via Programming wrote: > What should I be looking for to know whether avx is in operation? Is it 2!:7 > ''? That would work. Or you could look at the first line from JVERSION. FYI, -- Raul

Re: [Jprogramming] j901 beta-h available

2!:7 '' avx avx2 works, but the very first line of JVERSION will have avx reference on "Engine:" On Tuesday, August 27, 2019, 02:37:12 p.m. EDT, 'robert therriault' via Programming wrote: Hi Pascal, What should I be looking for to know whether avx is in operation? Is it 2!:7 ''?

Re: [Jprogramming] j901 beta-h available

Hi Pascal, What should I be looking for to know whether avx is in operation? Is it 2!:7 ''? Cheers, bob > On Aug 27, 2019, at 11:34 AM, 'Pascal Jasmin' via Programming > wrote: > > Yes. Seems like a second je_update (does not require console command) brings > avx. >On Tuesday, August 27

Re: [Jprogramming] j901 beta-h available

Yes.  Seems like a second je_update (does not require console command) brings avx. On Tuesday, August 27, 2019, 01:49:49 p.m. EDT, Eric Iverson wrote: Joey, Pascal, Sounds like you are not getting the avx version. Check the result of 9!:14'' or JVERSION. It should indicate avx, if not

Re: [Jprogramming] J 9.01 Beta-h (9!:41) side effect

I don't think I touched 9!:41.  I can load task in my Windows system, and I always run with 9!:41 ] 0. Henry Rich On 8/27/2019 2:04 PM, John Baker wrote: Greetings all, This morning I updated to J 9.01 beta-h and started running through my jod addon tests. 9.01 beta-h immediately broke JOD.

Re: [Jprogramming] Document Object Model

I just meant to refer to the PE File Browser, which they offer for free. I have not tried their other products. FYI, -- Raul On Tue, Aug 27, 2019 at 1:46 PM 'Jim Russell' via Programming wrote: > > Thanks. Looks pricey, now that I avoid Windows environments. But I’m sure it > would have muc

Re: [Jprogramming] j901 beta-h available

Joeh, Just noticed that the je_update_pacman_ result you report is from the old version. First you need to do an update to the base library. This will get you the new version of je_update and then you might get the new avx version installed. 1. update base library 2. je_update_jpacman_'' NB. this

Re: [Jprogramming] Calculus primitives are removed

For Mike Powell - checking those APL differentials and integrals: I sometimes use pari GP when I can't get extended precision to give the "correct" results, and hardly at all for its very powerful higher mathematical library,  but was able to check just now: it has derivative and integration pr

[Jprogramming] J 9.01 Beta-h (9!:41) side effect

Greetings all, This morning I updated to J 9.01 beta-h and started running through my jod addon tests. 9.01 beta-h immediately broke JOD. The problem derives from loading the task addon. By changing the JOD startup script I was able to patch JOD and maintain backwards compatibility but I came

Re: [Jprogramming] j901 beta-h available

Here is the result of running things again - MBpro-3:Downloads jkt$ jb 2!:7'' avx avx2 JVERSION Engine: j901/j64/darwin Beta-h: commercial/2019-08-26T15:18:27 Library: 9.01.06 Platform: Darwin 64 Installer: J901 install InstallPath: /applications/j901 Contact: www.jsoftware.com 9!:14''

Re: [Jprogramming] j901 beta-h available

Joey, Pascal, Sounds like you are not getting the avx version. Check the result of 9!:14'' or JVERSION. It should indicate avx, if not then you are not running the avx version. The 2!:7'' result should indicated the versions your hardware can run. You may just need to do the install again so tha

Re: [Jprogramming] Document Object Model

Thanks. Looks pricey, now that I avoid Windows environments. But I’m sure it would have much to teach me. > On Aug 27, 2019, at 12:09 PM, Raul Miller wrote: > > In that case, you might also want to pick up > https://www.softwareverify.com/cpp-pe-file-browser.php -

Re: [Jprogramming] Calculus primitives are removed

Ok, I was too hasty looking at the possible repos. Thanks. One more question, though: What's the procedure for suggesting changes to an addon? For example, I just now submitted a pull request against one of the 100+ repositories for jsoftware on github -- https://github.com/jsoftware/general_mis

Re: [Jprogramming] j901 beta-h available

Also, should JVERSION mention avx if it has found/upgraded to that engine?  (windows 10-64) On Tuesday, August 27, 2019, 01:08:32 p.m. EDT, Joey K Tuttle wrote: Eric, I'm befuddled by the NOTE: in your announcement. The install j901 beta-h went fine, I assumed you intended '' instead

Re: [Jprogramming] j901 beta-h available

Eric, I'm befuddled by the NOTE: in your announcement. The install j901 beta-h went fine, I assumed you intended '' instead of a single ' ... But, when I did my usual benchmark, it took twice as long as the version (beta-g) that was replaced. Then I reread your note and ran it again (with no c

Re: [Jprogramming] Creating adverb

That's good advice, but not for raw beginners.  You know your audience. Henry Rich On 8/27/2019 12:46 PM, Arnab Chakraborty wrote: Dear all, Thanks for the comments. I shall update the tutorial accordingly. My love for =. stems from an advice that I got from a C instructor: Always declare v

Re: [Jprogramming] Creating adverb

Dear all, Thanks for the comments. I shall update the tutorial accordingly. My love for =. stems from an advice that I got from a C instructor: Always declare variables in the smallest scope possible. Of course, I am not supposed to mix C philosophy with J philosophy... Thanks and regards, Ar

Re: [Jprogramming] Document Object Model

In that case, you might also want to pick up https://www.softwareverify.com/cpp-pe-file-browser.php Mind you, it exposes a *lot* of detail, and you won't have time to study it all. But an occasional browser through it (maybe 15-20 minutes, several times a week, and an occasional deep dive when the

Re: [Jprogramming] 901 and calculus primitives (D. et. al.)

Mike, And if it doesn't meet your requirements in some way, it would be great if you could be part of the team that fixed the shortcoming. On Tue, Aug 27, 2019 at 12:04 PM Mike Powell wrote: > Having had a chance to look at math/calculus, I’m pretty sure it’ll meet > all of my needs. > > I’ll k

Re: [Jprogramming] 901 and calculus primitives (D. et. al.)

Having had a chance to look at math/calculus, I’m pretty sure it’ll meet all of my needs. I’ll keep J807 for reference. Thanks, Mike > On Aug 27, 2019, at 08:56, Eric Iverson wrote: > > 901 and calculus primitives (D. et. al.): > > If you want D. , and it completely meets your requirements

[Jprogramming] 901 and calculus primitives (D. et. al.)

901 and calculus primitives (D. et. al.): If you want D. , and it completely meets your requirements, then continue to use 807. If you want D. type facilities, but were frustrated by its limitations, bugs, and opaqueness (C code), then use the new 901 facilities. The implementation is all in J an

Re: [Jprogramming] Document Object Model

Nope; dll’s are one of the lead items in my vast storehouse of ignorance. Thanks! > On Aug 27, 2019, at 11:45 AM, Raul Miller wrote: > > Oh, and http://www.dependencywalker.com/ can be useful for having an > understanding of the dlls underneath a wrapper dll. > > But you probably already knew

Re: [Jprogramming] Document Object Model

Oh, and http://www.dependencywalker.com/ can be useful for having an understanding of the dlls underneath a wrapper dll. But you probably already knew that. Thanks, -- Raul On Tue, Aug 27, 2019 at 11:43 AM Raul Miller wrote: > > System.Windows.Forms.dll is another likely possibility, since th

Re: [Jprogramming] Calculus primitives are removed

Hey Raul, Try https://github.com/jsoftware/math_calculus The other version with a - instead of a _ between math and calculus is empty. Cheers, bob > On Aug 27, 2019, at 8:39 AM, Raul Miller wrote: > > Where can I see the addon? > > https://github.com/jsoftware/math-calculus looks empty. >

Re: [Jprogramming] Document Object Model

System.Windows.Forms.dll is another likely possibility, since that's what exports the HtmlDocument class for windows. Good luck, -- Raul On Mon, Aug 26, 2019 at 7:29 PM 'Jim Russell' via Programming wrote: > > Yes, very much. As I recall, a single DOM handled all web pages; a class of > objec

Re: [Jprogramming] Calculus primitives are removed

Where can I see the addon? https://github.com/jsoftware/math-calculus looks empty. Thanks, -- Raul On Mon, Aug 26, 2019 at 7:48 PM Henry Rich wrote: > > Look at the addon. If you want to improve it, ask for permission to > modify it. People have talked about doing this from time to time, an

Re: [Jprogramming] Calculus primitives are removed

Thanks Bob & Eric, that worked. > On Aug 27, 2019, at 08:15, 'robert therriault' via Programming > wrote: > > Hi Mike, > > In order to update the J Engine to beta h on a Mac, I needed to run > > load 'pacman' > je_update_jpacman_' ' NB. note the second quote > > That instructed me to close a

Re: [Jprogramming] Calculus primitives are removed

Mike, Looks like you installed the initial 901 beta, but have not done the updates. Please try the following (after making sure the base library is up to date): load'pacman' je_update_jpacman_ '' NB. this will install the latest 901 beta version On Tue, Aug 27, 2019 at 10:29 AM Mike Pow

Re: [Jprogramming] j901 beta-h available

Apologies for my typo. je_update_jpacman_ '' NB. two single quote so the the verb is run On Tue, Aug 27, 2019 at 6:49 AM 'Bo Jacoby' via Programming < programm...@jsoftware.com> wrote: > > load'pacman' > > je_update_jpacman_' > > |open quote > > | je_update_jpacman_' > > | ^ > > |[-1]

Re: [Jprogramming] Calculus primitives are removed

Hi Mike, In order to update the J Engine to beta h on a Mac, I needed to run load 'pacman' je_update_jpacman_' ' NB. note the second quote That instructed me to close any instances of J that were running, open the Terminal and run (not clear on exact instructions since, now that it is updated,

Re: [Jprogramming] Calculus primitives are removed

JVERSION Engine: j901/j64/darwin Beta-g: commercial/2019-05-28T14:58:30 Library: 9.01.08 Qt IDE: 1.8.3/5.12.3 Platform: Darwin 64 Installer: J901 install InstallPath: /applications/j901 Contact: www.jsoftware.com > On Aug 27, 2019, at 07:41, bill lam wrote: > > Can you give its JVERSION ? > >

Re: [Jprogramming] Calculus primitives are removed

Can you give its JVERSION ? It seems ok on jandroid load'math/calculus' +: deriv_jcalculus_ 1 2"0 JVERSION Engine: j901/j64/android Beta-h: commercial/2019-08-27T10:39:14 Library: 9.01.07 J Android: 1.3.02/6.0.1/23 Platform: Android 64 (arm64-v8a) Installer: unknown InstallPath: /storage

Re: [Jprogramming] Calculus primitives are removed

I downloaded j901_mac64.zip from the link https://code.jsoftware.com/wiki/System/Installation/Zips . > On Aug 27, 2019, at 07:30, Henry Rich wrote: > > Somehow you have a J Engine from an earlier beta. It doesn't support the u. > prim

Re: [Jprogramming] Calculus primitives are removed

Somehow you have a J Engine from an earlier beta.  It doesn't support the u. primitive, which is new. Henry Rich On 8/27/2019 10:29 AM, Mike Powell wrote: Try load'math/calculus' +: deriv_jcalculus_ 1 I have installed J901 beta and start J with jqt.app just fine. I’ve used the Pac

Re: [Jprogramming] Calculus primitives are removed

> Try > > load'math/calculus' > +: deriv_jcalculus_ 1 > I have installed J901 beta and start J with jqt.app just fine. I’ve used the Package Manager to install math/calculus (three times): Updating server catalog... Done. Installing 1 package of size 6.4 KB Downloading math/calculus... I

Re: [Jprogramming] Calculus primitives are removed

> math/calculus points to math/cal This is clearly wrong, since math/cal has nothing to do with calculus support. I've fixed it – but "math/calculus" now shows up in red because the page [[Addons/math/calculus]] doesn't exist yet. On Tue, 27 Aug 2019 at 12:26, 'Rob B' via Programming < programm.

Re: [Jprogramming] Calculus primitives are removed

Hi, On this page.. https://code.jsoftware.com/wiki/Vocabulary/Libraries math/calculus points to math/cal Regards, Rob. > On 27 Aug 2019, at 00:04, Henry Rich wrote: > > J has supported symbolic differentiation and integration for some time. > Starting with Release 9.01, primitive support f

Re: [Jprogramming] j901 beta-h available

    load'pacman' je_update_jpacman_' |open quote | je_update_jpacman_' | ^ |[-1] je_update_jpacman_ 3 : 0 if. IFIOS+.UNAME-:'Android' do. 'update not supported for this platform' return. end. mkdir_j_ jef sh=. 'update.',;(UNAME-:'Win'){'sh';'bat' ferase jef,sh 'plat name bname'=.