Thanks to everyone for the interesting approaches. I enjoyed
dissecting the different thinking processes.
As Marshall points out, not all the solutions will work on lists that
contain non-unique characters which I think is a desirable
characteristic. His development of Raul's approach seems to have the
best combination of generality, performance and simplicity.
findinRows=: [: *@#@> [: ([ #~ (< {:))&.>/ <@I.@(=/~)
On Fri, Aug 19, 2011 at 7:40 AM, R.E. Boss <[email protected]> wrote:
> This has the gist of Millers solution.
>
>
> R.E. Boss
>
>
>> -----Oorspronkelijk bericht-----
>> Van: [email protected] [mailto:programming-
>> [email protected]] Namens Marshall Lochbaum
>> Verzonden: donderdag 18 augustus 2011 19:19
>> Aan: Programming forum
>> Onderwerp: Re: [Jprogramming] Search for ordered letters within lines of
>> matrix
>>
>> Also, this is a nice version based on your original strategy.
>>
>> findinrows =. [: *@#@> [: ([#~(<{:))&.>/ [: <@I. =/~
>> A findinrows 'TAG'
>> 0 1 0 1 0
>>
>> Marshall
>>
>> On Thu, Aug 18, 2011 at 1:05 PM, Marshall Lochbaum
>> <[email protected]>wrote:
>>
>> > Here is a version that's morally equivalent to R.E. Boss's, but a bit
>> > cleaner and more efficient. Note that it only works if the string you
>> > are finding has all unique characters.
>> >
>> > findinrows =. 0 = (,{~1~:-~)/@(#@[,~i.)"1
>> > 'TAG' findinrows A
>> > 0 1 0 1 0
>> >
>> > Marshall
>> >
>> > On Thu, Aug 18, 2011 at 10:51 AM, R.E. Boss <[email protected]> wrote:
>> >
>> >>
>> >> 'TAG' (i.@>:@#@[ (-:~.)"1 [:(]`[@.(1=-~))/\."(1) #@[ ,.~ i."1) A
>> >> 0 1 0 1 0
>> >>
>> >>
>> >> R.E. Boss
>> >>
>> >>
>> >> > -----Oorspronkelijk bericht-----
>> >> > Van: [email protected] [mailto:programming-
>> >> > [email protected]] Namens Ric Sherlock
>> >> > Verzonden: donderdag 18 augustus 2011 15:25
>> >> > Aan: Programming forum
>> >> > Onderwerp: [Jprogramming] Search for ordered letters within lines
>> >> > of
>> >> matrix
>> >> >
>> >> > I saw the following thread on comp.lang.apl
>> >> >
>> http://groups.google.com/group/comp.lang.apl/browse_thread/thread/8
>> >> > 9c
>> >> > 585c9d1a7bd3a#
>> >> > and was trying to put together a (non-regex) J solution.
>> >> >
>> >> > Given the matrix:
>> >> > ]A=:
>> >> > 'CTGGTTGAT','GTAGTCATA','CATGTCTAA','TCGAAAGTT',:'CCGGAGAAG'
>> >> > CTGGTTGAT
>> >> > GTAGTCATA
>> >> > CATGTCTAA
>> >> > TCGAAAGTT
>> >> > CCGGAGAAG
>> >> >
>> >> > and the list 'TAG'
>> >> >
>> >> > Identify the rows of the matrix where the letters in the list occur
>> >> > in
>> >> the
>> >> order
>> >> > they appear in the list, but not necessarily adjacent to each other.
>> >> > So the result of a verb findinRows would be:
>> >> > A findinRows 'TAG'
>> >> > 0 1 0 1 0
>> >> >
>> >> > I have an answer below but am thinking there may be other
>> >> > approaches I'm missing
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> > Spoiler alert!!
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> > findinRows=: [: (1 e. ,@:(</))&>/ 2 (([: ($ #. ]) 4 $. [: $.
>> >> > </)&.>)/\
>> >> > [: <@I."1 ="_ 0
>> >> > -------------------------------------------------------------------
>> >> > --- 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
