I think we need use-cases for #: with negative arguments. I've probably hit some, over the years, but I can't recall any at the moment. Anyone else?
-Dan -----Original Message----- From: programming-boun...@jsoftware.com [mailto:programming-boun...@jsoftware.com] On Behalf Of Raul Miller Sent: Monday, December 12, 2011 9:34 AM To: Programming forum Subject: Re: [Jprogramming] How #: should have been designed Then again, the current behavior is not all that useful: 2 #. #: i: 6 2 3 4 5 6 7 0 1 2 3 4 5 6 2 #. #: i: 7 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 2 #. #: i: 8 8 9 10 11 12 13 14 15 0 1 2 3 4 5 6 7 8 -- Raul On Mon, Dec 12, 2011 at 9:31 AM, Raul Miller <rauldmil...@gmail.com> wrote: > Ok... > > This, combined with a few other issues, is making me think that #: is > not the right word to be used here. Perhaps #:: but probably just a > library word. > > But the problem is that 2 #. #: i: 9 would no longer be equivalent to i: 9. > > That, and my original suggestion breaks if #: gets updated... > > -- > Raul > > On Sun, Dec 11, 2011 at 11:43 PM, Kip Murray <k...@math.uh.edu> wrote: >> Below I present tcpl "two's-complement plus". It works with the >> transpose of the arrangement below for adding two's-complement numbers: >> >> 1 1 0 0 (value _4) >> 1 1 0 1 (value _3) >> ------- >> 1 1 0 0 (carries shifted to the right) >> 1 0 0 1 (answer, value _7) >> >> Working above from right to left, you say 0 plus 1 is 0 1 and write 0 1 >> vertically below the line. Next you say 0 plus 0 plus 0 is 0 0 (the >> first 0 is the carry from the preceding step) and write 0 0 vertically >> below the line. Continuing, you say 0 plus 1 plus 1 is 1 0 and 1 plus 1 >> plus 1 is 1 1 . The answer 1 0 0 1 is in the bottom line. >> >> hc =: {.@#:@(,: 2 * |) NB. Raul's improved #: (hash colon) >> >> hcinv =: ([: -/ [: #. (,: [: +: 1 {. ]))"1 NB. Henry Rich >> >> Table =: 2 2 2 2 $ 0 0,0 1,0 1,1 0,0 1,1 0,1 0,1 1 >> >> (< 1 1 1) { Table NB. sum of three bits >> 1 1 >> >> stack =: ,.&.|: NB. stacks x over y >> >> hv =: (0 {:: <"1) :: ] NB. returns head vector >> >> op =: ] stack~ (2 {. [) , Table {~ [: < (2 {. [) , 2 { [: hv ] >> >> ba =: 0 ,~ 0 ,.~ 0 ,.~ ,. NB. build argument >> >> 1 1 0 0 ba 1 1 0 1 NB. starting table >> 1 1 0 0 >> 1 1 0 0 >> 0 0 0 0 >> 0 1 0 0 >> 0 0 0 0 >> >> op/ 1 1 0 0 ba 1 1 0 1 >> 1 1 1 1 >> 1 1 1 0 >> 0 0 0 0 >> 0 1 0 1 >> 0 0 0 0 >> >> NB. above see transpose of arrangement at top, with an added >> NB. row of 0's >> >> tcpl =: [: }: [: {:"1 [: op/ ba NB. two's-complement plus >> >> 1 1 0 0 tcpl 1 1 0 1 NB. _4 plus _3 is _7 >> 1 0 0 1 >> >> hc _4 _3 _7 >> 1 1 0 0 >> 1 1 0 1 >> 1 0 0 1 >> >> On 12/10/2011 6:48 PM, Kip Murray wrote: >>> Cool. I think it is an improvement because it neatly avoids a case >>> statement. Now I wonder if we could implement two's-complement addition >>> and multiplication with overflow, basing these on bitwise operations >>> >>> 0 + 0 is 0, 0 + 1 is 1 + 0 is 1, and 1 + 1 is 1 0 >>> >>> 0 * 0 is 0 * 1 is 1 * 0 is 0, and 1*1 is 1 >>> >>> That is, I do not want hc and hcinv to be used except to produce data >>> and check answers, and I want n-bit two's-complement answers for n-bit >>> two's-complement data so some of them will be wrong because of overflow. >> >> ADDED LATER -- ACCEPT 2n-bit multiplication answers from n-bit data >> >>> >>> Here for Linda is Henry's hcinv expressed without conjunctions other >>> than " . >>> >>> hcinv =: ([: -/ [: #. (,: [: +: 1 {. ]))"1 >>> >>> On 12/10/2011 10:00 AM, Henry Rich wrote: >>>> Same idea, better implementation >>>> >>>> hcinv =. -/@:#.@:(,: +:@(1&{.))"1 >>>> >>>> Henry Rich >> ---------------------------------------------------------------------- >> For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
