"you cannot avoid referring to the name" That is precisely the problem. If one intends to use ". , to solve the original request, then the linear representation of the argument is required and you have to name it. Hence,
lrep=: 3 :'5!:5<''y''' lrep 1 2 3 1 2 3 works. However, lrep=: [: ". '5!:5 <''','''',~] lrep 1 2 3 |domain error: lrep | lrep 1 2 3 does not. Therefore, as far as I can see, "it is not possible to produce of that kind of verbs without involving an explicit verb (or an adverb or a conjunction) which refer to its argument." Perhaps you misunderstood my quoted remark; I meant to say that an explicit verb which refers to its argument is required, (or an explicit adverb which refers to its argument is required, or an explicit verb which refers to one of its argument is required)." Moreover, some important primitives and foreigns are not tacit friendly because of the naming requirement. On Wed, Jul 5, 2017 at 2:21 PM, Raul Miller <[email protected]> wrote: > you cannot avoid referring to the name, but there's no need to use the > : conjunction for this. > > For example: > > lrep=: [: ". '5!:5 <''','''',~] > lrep 'lrep' > [: ". '5!:5 <''' , '''' ,~ ] > > Thanks, > > -- > Raul > > > On Wed, Jul 5, 2017 at 1:20 PM, Jose Mario Quintana > <[email protected]> wrote: > > What a coincidence that this type of question was posed a day after I > > suggested to incorporate the unofficial Jx interpreter verb extension =:: > > into the official interpreter! > > > > I am afraid that it is not possible to produce of that kind of verbs > > without involving an explicit verb (or an adverb or a conjunction) which > > refer to its argument (such as 3 :'5!:5<''y''' which is not tacit) . > > > > I would be glad if someone can prove me wrong. > > > > > > > > On Tue, Jul 4, 2017 at 8:54 PM, Nikolas Drosdek < > [email protected]> > > wrote: > > > >> This is exactly what I was looking for, thanks a lot! > >> > >> I am slightly confused by this however: > >> > >> 13 : '5!:5<''y''' > >> (,'1')"_ > >> > >> as it seems to replace y with 1 instead of giving 3 : '5!:5<''y'''. > >> > >> Thanks, > >> N.D. > >> > >> 2017-07-05 3:28 GMT+03:00 Raul Miller <[email protected]>: > >> > >> > Well, that depend on what you mean by a "tacit expression". > >> > > >> > If your concept of tacit is one which does not use names, then no, > >> > because this expression uses names. > >> > > >> > If you prefer a more relaxed concept of tacit programming, however, > >> > you might try something like this: > >> > > >> > lrep=: 3 :'5!:5<''y''' > >> > V=: [: ". 'R=:R,' , lrep > >> > > >> > Thanks, > >> > > >> > -- > >> > Raul > >> > > >> > > >> > > >> > On Tue, Jul 4, 2017 at 8:14 PM, Nikolas Drosdek > >> > <[email protected]> wrote: > >> > > Explicitly: > >> > > > >> > > R=: 0 > >> > > V=: 3 : 'R=: R , ...' > >> > > > >> > > Is there a way to write V in tacit form? > >> > > > >> > > Thanks, > >> > > N.D. > >> > > ------------------------------------------------------------ > ---------- > >> > > For information about J forums see http://www.jsoftware.com/forum > s.htm > >> > ------------------------------------------------------------ > ---------- > >> > For information about J forums see http://www.jsoftware.com/forum > s.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
