I have decresed the dataset so I can plot it on my TI84. The units I'm using are LYs for distances (so the ticks on the x-axis is seconds). I'm just learning J. > From: [email protected] > Subject: Programming Digest, Vol 68, Issue 45 > To: [email protected] > Date: Fri, 27 May 2011 01:57:16 +0800 > > Send Programming mailing list submissions to > [email protected] > > To subscribe or unsubscribe via the World Wide Web, visit > http://jsoftware.com/cgi-bin/mailman/listinfo/programming > or, via email, send a message with subject or body 'help' to > [email protected] > > You can reach the person managing the list at > [email protected] > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Programming digest..." > > > Today's Topics: > > 1. Re: Programming Digest, Vol 68, Issue 43 - array limit (Dan Bron) > 2. Re: roll (Andrew Nikitin) > 3. Re: roll (Raul Miller) > 4. [JHS] how to start jum (bill lam) > 5. Re: [JHS] how to start jum (Lettow, Kenneth) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 26 May 2011 00:03:19 -0400 > From: "Dan Bron" <[email protected]> > Subject: Re: [Jprogramming] Programming Digest, Vol 68, Issue 43 - > array limit > To: "'Programming forum'" <[email protected]> > Message-ID: <002201cc1b59$d959ebd0$8c0dc370$@us> > Content-Type: text/plain; charset="us-ascii" > > Well, 65000*65000 is 4,225,000,000 ~= 4 billion. If your distances are > integral and integers are 4 bytes, then that's already >16GB. If your > distances are floating point or integers are 8 bytes wide (I think they are > on 64-bit architectures?), then that's >32GB. > > Even if you had enough RAM to make this feasible, it would probably be a > waste, given that the output is only plotted. So probably easiest to do > some sampling to reduce the data set. Our local mathematicians can advise > you on that better than I can. > > One question though: you indicate the plot is 3D, but your array only has 2 > dimensions (65e3 x 65e3). Where's the 3rd dimension? > > -Dan > > -----Original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of Charles Holder > Sent: Wednesday, May 25, 2011 7:38 PM > To: [email protected] > Subject: Re: [Jprogramming] Programming Digest, Vol 68, Issue 43 - array > limit > > > I ask because I am shaping an array 65000 by 65000 for ploting LY distances > in Boeing Graph (3D). > > > From: [email protected] > > Subject: Programming Digest, Vol 68, Issue 43 > > To: [email protected] > > Date: Wed, 25 May 2011 22:22:39 +0800 > > > > Send Programming mailing list submissions to > > [email protected] > > > > To subscribe or unsubscribe via the World Wide Web, visit > > http://jsoftware.com/cgi-bin/mailman/listinfo/programming > > or, via email, send a message with subject or body 'help' to > > [email protected] > > > > You can reach the person managing the list at > > [email protected] > > > > When replying, please edit your Subject line so it is more specific > > than "Re: Contents of Programming digest..." > > > > > > Today's Topics: > > > > 1. J arrays (Charles Holder) > > 2. Re: J arrays (Roger Hui) > > 3. Sorting (Gian Medri) > > 4. Re: Sorting (Pablo Landherr) > > 5. Re: Sorting (bill lam) > > 6. Re: Sorting (R.E. Boss) > > 7. Re: Sorting (bill lam) > > 8. Re: J arrays (Dan Bron) > > 9. Re: Sorting (Devon McCormick) > > > > > > ---------------------------------------------------------------------- > > > > Message: 1 > > Date: Wed, 25 May 2011 05:23:23 +0000 > > From: Charles Holder <[email protected]> > > Subject: [Jprogramming] J arrays > > To: <[email protected]> > > Message-ID: <[email protected]> > > Content-Type: text/plain; charset="iso-8859-1" > > > > > > What is the max items an array can have? > > > > > > ------------------------------ > > > > Message: 2 > > Date: Tue, 24 May 2011 22:47:47 -0700 > > From: Roger Hui <[email protected]> > > Subject: Re: [Jprogramming] J arrays > > To: Programming forum <[email protected]> > > Message-ID: <[email protected]> > > Content-Type: text/plain; charset=us-ascii > > > > _1+2^n-1 where n is the number of bits in a machine word. > > (n is 32 or 64). > > > > > > > > ----- Original Message ----- > > From: Charles Holder <[email protected]> > > Date: Tuesday, May 24, 2011 22:24 > > Subject: [Jprogramming] J arrays > > To: [email protected] > > > > > What is the max items an array can have? > > > > > > > > ------------------------------ > > > > Message: 3 > > Date: Wed, 25 May 2011 10:00:42 +0200 > > From: Gian Medri <[email protected]> > > Subject: [Jprogramming] Sorting > > To: [email protected] > > Message-ID: <[email protected]> > > Content-Type: text/plain; charset=UTF-8 > > > > Hi! > > 1) I have a problem with sorting with the Swedish alphabet that contains 3 > > extra letters after z. > > Example: > > /:~'qwe?rta?as?sdf' > > aadefqrsstw?????? > > > > Perhaps it depends on the limitation of "a. "? > > > > 2) Often I encounter the necessity to replace element x by element y in an > > array. > > I solved the problem in an APL fashion such as: > > repl=: 4 : '($y)$ (1{x) (((0{x)=,y)#i.*/$y)},y' NB. replace x1 in array y > > by x2 > > > > This verb works OK for any array, but I wonder if there is a better J-way > > to do it. > > > > Thanks > > Gian Medri > > > > > > ------------------------------ > > > > Message: 4 > > Date: Wed, 25 May 2011 10:38:04 +0200 > > From: Pablo Landherr <[email protected]> > > Subject: Re: [Jprogramming] Sorting > > To: Programming forum <[email protected]> > > Message-ID: <[email protected]> > > Content-Type: text/plain; charset=UTF-8 > > > > And I get > > > > 'qwe?rta?as?sdf'/:'abcdefghijklmnopqrstuvwxyz???' i. 'qwe?rta?as?sdf' > > aadefqrsstw? > > 9!:14 '' > > j602/2008-03-03/16:45 > > > > On Wed, May 25, 2011 at 10:00 AM, Gian Medri <[email protected]> wrote: > > > > > Hi! > > > 1) I have a problem with sorting with the Swedish alphabet that contains > 3 > > > extra letters after z. > > > Example: > > > /:~'qwe?rta?as?sdf' > > > aadefqrsstw?????? > > > > > > Perhaps it depends on the limitation of "a. "? > > > > > > 2) Often I encounter the necessity to replace element x by element y in > an > > > array. > > > I solved the problem in an APL fashion such as: > > > repl=: 4 : '($y)$ (1{x) (((0{x)=,y)#i.*/$y)},y' NB. replace x1 in array > y > > > by x2 > > > > > > This verb works OK for any array, but I wonder if there is a better > J-way > > > to do it. > > > > > > Thanks > > > Gian Medri > > > ---------------------------------------------------------------------- > > > For information about J forums see http://www.jsoftware.com/forums.htm > > > > > > ------------------------------ > > > > Message: 5 > > Date: Wed, 25 May 2011 16:42:28 +0800 > > From: bill lam <[email protected]> > > Subject: Re: [Jprogramming] Sorting > > To: [email protected] > > Message-ID: <[email protected]> > > Content-Type: text/plain; charset=utf-8 > > > > You may try converting it to wide character first > > > > /:~ucp'qwe?rta?as?sdf' > > aadefqrsstw??? > > > > or if you want utf8 back > > > > /:~&.ucp'qwe?rta?as?sdf' > > aadefqrsstw??? > > > > if x is a rank-1 vector of 2 atom, I think it can be (untested) > > > > ($y) $ (1{x) (I.(0{x)=,y) } ,y > > > > please also refer to the verb charsub , that amend idiom should be > supported by > > special code. > > > > ???, 25 ??? 2011, Gian Medri ?????(?): > > > Hi! > > > 1) I have a problem with sorting with the Swedish alphabet that contains > 3 > > > extra letters after z. > > > Example: > > > /:~'qwe?rta?as?sdf' > > > aadefqrsstw?????? > > > > > > Perhaps it depends on the limitation of "a. "? > > > > > > 2) Often I encounter the necessity to replace element x by element y in > an > > > array. > > > I solved the problem in an APL fashion such as: > > > repl=: 4 : '($y)$ (1{x) (((0{x)=,y)#i.*/$y)},y' NB. replace x1 in array > y > > > by x2 > > > > > > This verb works OK for any array, but I wonder if there is a better > J-way > > > to do it. > > > > > > Thanks > > > Gian Medri > > > ---------------------------------------------------------------------- > > > For information about J forums see http://www.jsoftware.com/forums.htm > > > > -- > > regards, > > ==================================================== > > GPG key 1024D/4434BAB3 2008-08-24 > > gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 > > > > > > ------------------------------ > > > > Message: 6 > > Date: Wed, 25 May 2011 11:11:14 +0200 > > From: "R.E. Boss" <[email protected]> > > Subject: Re: [Jprogramming] Sorting > > To: "'Programming forum'" <[email protected]> > > Message-ID: <[email protected]> > > Content-Type: text/plain; charset="UTF-8" > > > > rpl2a=: ] - (-/ , 0:)@[ {~ {.@[ i. ] > > > > by Hui from > http://www.jsoftware.com/pipermail/programming/2007-July/007303.html is the > most elegant known (to me). > > > > If x and y are from a small domain (e.g. characters), then rpl3a is faster > and leaner: > http://www.jsoftware.com/pipermail/programming/2007-July/007302.html > > > > rpl3a=: 4 : 0 > > 'x0 x1'=. x > > ((x1,a.) {~ (x0,a.) i. ]) y > > ) > > > > > > R.E. Boss > > > > > > > -----Oorspronkelijk bericht----- > > > Van: [email protected] [mailto:programming- > > > [email protected]] Namens Gian Medri > > (...) > > > > > 2) Often I encounter the necessity to replace element x by element y in > an > > > array. > > > I solved the problem in an APL fashion such as: > > > repl=: 4 : '($y)$ (1{x) (((0{x)=,y)#i.*/$y)},y' NB. replace x1 in array > y by x2 > > > > > > This verb works OK for any array, but I wonder if there is a better > J-way to do > > > it. > > > > > > > > > > ------------------------------ > > > > Message: 7 > > Date: Wed, 25 May 2011 18:50:57 +0800 > > From: bill lam <[email protected]> > > Subject: Re: [Jprogramming] Sorting > > To: [email protected] > > Message-ID: <[email protected]> > > Content-Type: text/plain; charset=utf-8 > > > > What you see in jconsole or ide is _not_ the same as that inside J engine. > For > > example, the character ? is one character in ide but it is two bytes in J > > engine, furthermore, J engine has no idea that these two bytes should be > > regarded as a single character, or know that you want them to be > > interpreted in utf8 encoding, or anything special. So nearly all string > > operations on these non-ascii characters will fail. > > > > There are some articles on unicodes in jwiki, please enter 'unicode' in > the > > search box to locate them. They should provide some background materials. > > > > ???, 25 ??? 2011, Pablo Landherr ?????(?): > > > And I get > > > > > > 'qwe?rta?as?sdf'/:'abcdefghijklmnopqrstuvwxyz???' i. 'qwe?rta?as?sdf' > > > aadefqrsstw? > > > 9!:14 '' > > > j602/2008-03-03/16:45 > > > > > > On Wed, May 25, 2011 at 10:00 AM, Gian Medri <[email protected]> > wrote: > > > > > > > Hi! > > > > 1) I have a problem with sorting with the Swedish alphabet that > contains 3 > > > > extra letters after z. > > > > Example: > > > > /:~'qwe?rta?as?sdf' > > > > aadefqrsstw?????? > > > > > > > > Perhaps it depends on the limitation of "a. "? > > > > > > > > 2) Often I encounter the necessity to replace element x by element y > in an > > > > array. > > > > I solved the problem in an APL fashion such as: > > > > repl=: 4 : '($y)$ (1{x) (((0{x)=,y)#i.*/$y)},y' NB. replace x1 in > array y > > > > by x2 > > > > > > > > This verb works OK for any array, but I wonder if there is a better > J-way > > > > to do it. > > > > > > > > Thanks > > > > Gian Medri > > > > ---------------------------------------------------------------------- > > > > For information about J forums see http://www.jsoftware.com/forums.htm > > > ---------------------------------------------------------------------- > > > For information about J forums see http://www.jsoftware.com/forums.htm > > > > -- > > regards, > > ==================================================== > > GPG key 1024D/4434BAB3 2008-08-24 > > gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 > > > > > > ------------------------------ > > > > Message: 8 > > Date: Wed, 25 May 2011 12:02:44 +0000 > > From: "Dan Bron" <[email protected]> > > Subject: Re: [Jprogramming] J arrays > > To: "J Programming" <[email protected]> > > Message-ID: > > > <1182149784-1306324965-cardhu_decombobulator_blackberry.rim.net-1490800668-@ > b18.c25.bise6.blackberry> > > > > Content-Type: text/plain > > > > Which is to say: exactly as much as your machine can handle (note the > formula below also defines the limits of memory pointers on modern > architectures). > > > > Which should also tell you that the max number of items depends on the > size (in memory) of 1 item. The smallest J item is one byte (a literal > atom), so you could have 4 billion of those on a 32 bit machine (modulo OS > taxes). But if your items were, say, 4-byte integers, you'd max out at 1 > billion. And if your items were tuples of 3 integers (R,G,B pixels, for > example) you'd top out at ~300 million, etc. > > > > > > Of course, J also supports sparse arrays, which have the same theoretical > limit (2^n etc) but much higher practical limits, as they only store the > relevant parts of arrays (of course, if all the items of your array are > relevant, you're back to the same memory limit). > > > > Why do you ask? > > > > -Dan > > > > PS: Roger's formula refers to the limit of a single dimension of a J array > (the number of "items", as opposed to "atoms"). But J arrays are > multidimensional, and each dimension could (in theory) have that many items, > and of course the array's total volume is the product of all its dimensions. > The max # of dimensions for a J array also shares Roger's limit, so the max > shape J's internals can handle is > > > > shape =. $~ _1+2^n-1 > > > > And correspondingly its volume would be */shape which is shape^shape or > about 2^(n-1)^2 atoms. Which of course no machine can handle. But it does > let you play silly games like carrying payload data around in the shape of > an (empty) array :) > > > > And while we're discussing theory, it's worth pointing out that the > specification of J (the Dictionary) imposes no limits on array size. So in > theory, a different implementation of J could have arrays that stretched > back onto disk (like a RDBMS), or across mutliple disks, or across multiple > networked machines, or across the globe.... > > > > The notation wouldn't change. Just the implementation. > > > > > > > > Please excuse typos; composed on a handheld device. > > > > -----Original Message----- > > From: Roger Hui <[email protected]> > > Sender: [email protected] > > Date: Tue, 24 May 2011 22:47:47 > > To: Programming forum<[email protected]> > > Reply-To: Programming forum <[email protected]> > > Subject: Re: [Jprogramming] J arrays > > > > _1+2^n-1 where n is the number of bits in a machine word. > > (n is 32 or 64). > > > > > > > > ----- Original Message ----- > > From: Charles Holder <[email protected]> > > Date: Tuesday, May 24, 2011 22:24 > > Subject: [Jprogramming] J arrays > > To: [email protected] > > > > > What is the max items an array can have? > > > > ---------------------------------------------------------------------- > > For information about J forums see http://www.jsoftware.com/forums.htm > > > > ------------------------------ > > > > Message: 9 > > Date: Wed, 25 May 2011 10:22:03 -0400 > > From: Devon McCormick <[email protected]> > > Subject: Re: [Jprogramming] Sorting > > To: Programming forum <[email protected]> > > Message-ID: <[email protected]> > > Content-Type: text/plain; charset=UTF-8 > > > > I like the boolean variant of replace: > > > > replace1by1=: 3 : 0 > > 'toreplace replacement target'=. y > > (toreplace=target)}target,:replacement > > ) > > > > replace1by1 'e';'X';'Replace one element by another' > > RXplacX onX XlXmXnt by anothXr > > > > $mat=. ><;._1 ' Replace one element by another' > > 5 7 > > > > replace1by1 'e';'X';mat > > RXplacX > > onX > > XlXmXnt > > by > > anothXr > > > > On Wed, May 25, 2011 at 6:50 AM, bill lam <[email protected]> wrote: > > > > > What you see in jconsole or ide is _not_ the same as that inside J > engine. > > > For > > > example, the character ? is one character in ide but it is two bytes in > J > > > engine, furthermore, J engine has no idea that these two bytes should be > > > regarded as a single character, or know that you want them to be > > > interpreted in utf8 encoding, or anything special. So nearly all string > > > operations on these non-ascii characters will fail. > > > > > > There are some articles on unicodes in jwiki, please enter 'unicode' in > the > > > search box to locate them. They should provide some background > materials. > > > > > > ???, 25 ??? 2011, Pablo Landherr ?????(?): > > > > And I get > > > > > > > > 'qwe?rta?as?sdf'/:'abcdefghijklmnopqrstuvwxyz???' i. 'qwe?rta?as?sdf' > > > > aadefqrsstw? > > > > 9!:14 '' > > > > j602/2008-03-03/16:45 > > > > > > > > On Wed, May 25, 2011 at 10:00 AM, Gian Medri <[email protected]> > > > wrote: > > > > > > > > > Hi! > > > > > 1) I have a problem with sorting with the Swedish alphabet that > > > contains 3 > > > > > extra letters after z. > > > > > Example: > > > > > /:~'qwe?rta?as?sdf' > > > > > aadefqrsstw?????? > > > > > > > > > > Perhaps it depends on the limitation of "a. "? > > > > > > > > > > 2) Often I encounter the necessity to replace element x by element y > in > > > an > > > > > array. > > > > > I solved the problem in an APL fashion such as: > > > > > repl=: 4 : '($y)$ (1{x) (((0{x)=,y)#i.*/$y)},y' NB. replace x1 in > > > array y > > > > > by x2 > > > > > > > > > > This verb works OK for any array, but I wonder if there is a better > > > J-way > > > > > to do it. > > > > > > > > > > Thanks > > > > > Gian Medri > > > > > > ---------------------------------------------------------------------- > > > > > For information about J forums see > http://www.jsoftware.com/forums.htm > > > > ---------------------------------------------------------------------- > > > > For information about J forums see http://www.jsoftware.com/forums.htm > > > > > > -- > > > regards, > > > ==================================================== > > > GPG key 1024D/4434BAB3 2008-08-24 > > > gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 > > > ---------------------------------------------------------------------- > > > For information about J forums see http://www.jsoftware.com/forums.htm > > > > > > > > > > > -- > > Devon McCormick, CFA > > ^me^ at acm. > > org is my > > preferred e-mail > > > > > > ------------------------------ > > > > ---------------------------------------------------------------------- > > For information about J forums see http://www.jsoftware.com/forums.htm > > > > End of Programming Digest, Vol 68, Issue 43 > > ******************************************* > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > > > > ------------------------------ > > Message: 2 > Date: Thu, 26 May 2011 11:02:55 -0400 > From: Andrew Nikitin <[email protected]> > Subject: Re: [Jprogramming] roll > To: J programming <[email protected]> > Message-ID: <[email protected]> > Content-Type: text/plain; charset="iso-8859-1" > > > Since we are on ? #. and x: subject, here is a trick question > > require 'plot' > t=:?10^100x > plot x;((100) 6!:2 ' #.inv t' ,~ ":)"> x=.10000+i:20 > > Why there is a performance hit at 10000 and not 10001, as one would expect if > extended integers were represented base 10000? > > My guess is another < vs <: bug, possibly in #./#: family. > > > ------------------------------ > > Message: 3 > Date: Thu, 26 May 2011 11:17:16 -0400 > From: Raul Miller <[email protected]> > Subject: Re: [Jprogramming] roll > To: Programming forum <[email protected]> > Message-ID: <[email protected]> > Content-Type: text/plain; charset=ISO-8859-1 > > On Thu, May 26, 2011 at 11:02 AM, Andrew Nikitin <[email protected]> wrote: > > Since we are on ? #. and x: subject, here is a trick question > > > > ? require 'plot' > > ? t=:?10^100x > > ? plot x;((100) 6!:2 ' #.inv t' ,~ ":)"> x=.10000+i:20 > > > > Why there is a performance hit at 10000 and not 10001, as one would expect > > if > > extended integers were represented base 10000? > > I am not sure why I should expect that. > > Ok, hypothetically speaking, if extended integers were represented in > base 10000, the implementation could optimize for the 10000 #.inv case > when dealing with extended integers. But it is not clear to me that > that kind of optimization would result in people writing better > programs in J. > > -- > Raul > > > ------------------------------ > > Message: 4 > Date: Fri, 27 May 2011 00:23:09 +0800 > From: bill lam <[email protected]> > Subject: [Jprogramming] [JHS] how to start jum > To: jprogramming <[email protected]> > Message-ID: <[email protected]> > Content-Type: text/plain; charset=us-ascii > > I use the following script to start jhs, and start browser to browse > localhost:65002/jum > > but it show the jlogin page, What is the correct way to let it show the jum > page? > > ----------------------------- > #!/home/bill/share/jsoftware/j701/bin32/jconsole > > load'~addons/ide/jhs/core.ijs' > load'~addons/ide/jhs/jum.ijs' > createjum_jum_ 65002;'1234' NB. only if new PORT or PASS > init_jhs_'jum' > NB. browse to jum task /jijx and login > NB. startjum_jum_ 65003 65004 65005;'localhost';'buzz' > > -- > regards, > ==================================================== > GPG key 1024D/4434BAB3 2008-08-24 > gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 > > > ------------------------------ > > Message: 5 > Date: Thu, 26 May 2011 13:56:41 -0400 > From: "Lettow, Kenneth" <[email protected]> > Subject: Re: [Jprogramming] [JHS] how to start jum > To: "Programming forum" <[email protected]> > Message-ID: > <[email protected]> > Content-Type: text/plain; charset="us-ascii" > > Untested, but wasn't the page jumjum? > > -----Original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of bill lam > Sent: Thursday, May 26, 2011 12:23 PM > To: jprogramming > Subject: [Jprogramming] [JHS] how to start jum > > I use the following script to start jhs, and start browser to browse > localhost:65002/jum > > but it show the jlogin page, What is the correct way to let it show the > jum > page? > > ----------------------------- > #!/home/bill/share/jsoftware/j701/bin32/jconsole > > load'~addons/ide/jhs/core.ijs' > load'~addons/ide/jhs/jum.ijs' > createjum_jum_ 65002;'1234' NB. only if new PORT or PASS > init_jhs_'jum' > NB. browse to jum task /jijx and login > NB. startjum_jum_ 65003 65004 65005;'localhost';'buzz' > > -- > regards, > ==================================================== > GPG key 1024D/4434BAB3 2008-08-24 > gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3 > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > > > ------------------------------ > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > > End of Programming Digest, Vol 68, Issue 45 > ******************************************* ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
