a variation on Rob's method:
dat=: 3 42$'John Smith WA 418-Y11-4111' (restore the blanks)
msk=: (i.42) e. +/\0,20 10 12
('';msk) <;.1 dat
On Thu, Nov 16, 2017 at 9:24 PM, Daniel Lyons <[email protected]>
wrote:
> Thanks Rob!
>
> I did feel that the general idea of the rolling sum generating the indices
> of the frets was not terrible, it just seemed like I was doing too much
> work to get there.
>
> I have noticed that almost every time I ask for something, it's right
> there in the core documentation!
>
> > On Nov 16, 2017, at 10:20 PM, Rob Hodgkinson <[email protected]> wrote:
> >
> > I meant to add … which is very close to what you derived.
> >
> > The key is the generation of the fret, another way here, but the group
> may suggest better phrases;
> > 1(0,+/\}.20 10 12)}42$0
> > 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0
> 0 0 0 0 0 0
> >
> > …/Regards Rob
> >
> >> On 17 Nov 2017, at 4:03 pm, Rob Hodgkinson <[email protected]> wrote:
> >>
> >> Hi Daniel, there is a primer article on cut (using cut2) which may help
> you;
> >> http://www.jsoftware.com/help/primer/cut.htm
> >>
> >> The last example shows fixed width cuts using a boolean mask, so your
> same example could be generated as so;
> >>
> >> NB. This generates the fret as a boolean list
> >> (i.42) e. +/\0,20 10 12
> >> 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0
> 0 0 0 0 0 0
> >>
> >> NB. Now apply the fret to the data using the ‘cut’ conjunction and
> argument is < (enclose)
> >> ((i.42) e. +/\0,20 10 12) <;.1 'John Smith WA 418-Y11-4111’
> >> ┌────────────────────┬──────────┬────────────┐
> >> │John Smith │WA │418-Y11-4111│
> >> └────────────────────┴──────────┴────────────┘
> >>
> >> NB. Now use rank “1 to apply the function on vectors of the table
> argument (as in your example)
> >> ((i.42) e. +/\0,20 10 12) <;.1"(1) 1 42$'John Smith WA
> 418-Y11-4111’ NB.
> >> ┌────────────────────┬──────────┬────────────┐
> >> │John Smith │WA │418-Y11-4111│
> >> └────────────────────┴──────────┴────────────┘
> >>
> >> Hope this helps, Regards Rob
> >>
> >>
> >>> On 17 Nov 2017, at 3:36 pm, Daniel Lyons <[email protected]>
> wrote:
> >>>
> >>> My problem is essentially to give an input file and the column widths
> and get it split up properly. I asked about it on the code review stack
> exchange site, but there doesn't seem to be much traffic there, so I
> figured maybe I could get a few eyes without annoying too many people
> reposting here.
> >>>
> >>> https://codereview.stackexchange.com/questions/
> 180633/parsing-fixed-width-format-with-j <https://codereview.
> stackexchange.com/questions/180633/parsing-fixed-width-format-with-j>
> >>>
> >>> The solution I hacked up is this:
> >>>
> >>> parseFW =: dyad def '(+/ (0,+/\x) ="(0 1) i. (+/x)) <;.1 y'"(_ _1)
> >>>
> >>> Usage then looks like this:
> >>>
> >>> (20 10 12) parseFW 1 42 $ 'John Smith WA 418-Y11-4111'
> >>>
> >>> I find this works, at least for the simple cases I have thrown at it
> so far, but I can't help but feel there's too much happening here for
> something fairly simple. I know where the frets are the rest is just
> encoding them. In particular it felt like the inverse of I. should be
> helpful but I didn't get that working.
> >>>
> >>> If there's a library that does this, I'd probably rather use that, but
> I didn't see one at first blush.
> >>>
> >>> Input greatly appreciated! Thanks,
> >>>
> >>> --
> >>> Daniel Lyons
> >>>
> >>>
> >>>
> >>>
> >>> ----------------------------------------------------------------------
> >>> 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
>
> --
> Daniel Lyons
>
>
>
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
