Sure, but what does nothing/left/right/both mean? If we are talking physical movement, "both" and "nothing" would be equivalent.
I was asking for a non-contrived example, if you recall? Thanks, -- Raul On Thu, Mar 2, 2017 at 9:20 AM, 'Pascal Jasmin' via Programming <[email protected]> wrote: > The need comes up pretty naturally in branching. A 4 range code that > indicates do "nothing, left,right, both" can be implemented with > pcase 1;3 > case 2;3 > > > when "do nothing" actually involves some code, it also permits a case "else" > condition too. > > > > > ----- Original Message ----- > From: Raul Miller <[email protected]> > To: Programming forum <[email protected]> > Sent: Thursday, March 2, 2017 3:11 AM > Subject: Re: [Jprogramming] fcase. > > That sounds sort of neat, but I cannot think of a non-contrived case > where this would be useful. > > -- > Raul > > > > On Thu, Mar 2, 2017 at 2:29 AM, 'Pascal Jasmin' via Programming > <[email protected]> wrote: >> The difference with the fcase. example at >> http://www.jsoftware.com/help/dictionary/csel.htm >> >> is the "case 2;3" line below pcase. where the fcase. example that line is >> "case 3" >> >> fcase triggers "2 conditions", but the 2nd is triggered automatically even >> though the input doesn't match that conditions' test. >> >> pcase. would require that the input match a subsequent condition to be >> further triggered. >> >> an example of the basic case statement >> >> >> 3 : 'select. y case. 1;2 do. 11 case. 2;3 do. 12 end.' 2 >> 11 >> 3 : 'select. y fcase. 1;2 do. 11 case. 3 do. 12 end.' 2 >> 12 >> >> only the first condition gets triggered with case. with fcase. the 2nd non >> matching condition gets returned, though first has chance to impose side >> effects. With pcase. the 2nd condition would get returned only if it also >> matched the input. >> >> >> >> >> ----- Original Message ----- >> From: Henry Rich <[email protected]> >> To: [email protected] >> Sent: Wednesday, March 1, 2017 10:44 PM >> Subject: Re: [Jprogramming] fcase. >> >> ? >> >> change your pcase. to fcase.; doesn't it do what you say? Is there >> something your proposed idea would do that the current syntax doesn't do? >> >> Henry Rich >> >> On 3/1/2017 10:14 PM, 'Pascal Jasmin' via Programming wrote: >>> Do not see the usefulness of the fcase. option. see example here: >>> http://www.jsoftware.com/help/dictionary/csel.htm >>> >>> >>> I would see a "passthrough case" pcase. usefulness >>> >>> f2=: 3 : 0 >>> t=. '' >>> select. y >>> case. 1 do. t=.t,'one ' >>> pcase. 2 do. t=.t,'two ' >>> case. 2;3 do. t=.t,'three ' >>> case. 2;3;4 do. t=.t,'four ' >>> end. >>> ) >>> >>> f2 2 >>> two three >>> >>> a pcase. statement would not abort search after a match. >>> ---------------------------------------------------------------------- >>> 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 > ---------------------------------------------------------------------- > 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
