[Jprogramming] Interweave

Martin Kreuzer info at airkreuzer.com
Fri Mar 11 11:39:05 UTC 2016


Trying to word the question which originally came to mind when I 
looked at this problem (converting the "list notation" of a c.f.) to 
a decimal value:

How would one go about throwing in the additional "1"s into the 
spaces (gaps) of the paling fence..? (There is one less "1" compared 
to the number of integers in the "list notation" [if I always write 
out the integer part, even if it is "0").

If the two lists I like to "interweave" don't agree in number of 
items, I obviously get a lenght error.

    ] v=. i.4
0 1 2 3
    ] w=. 3 $ 1
1 1 1
    ,v ,. w
|length error
|   ,v    ,.w

Would there be a (similar, slightly modified) way to do this, or does 
one need a (very) different approach..?

-M

At 2016-03-11 10:18, you wrote:
>To keep students organized in dealing with fractions I sometimes (on 
>a good day) have them work out intermediate results of continued 
>fractions like phi = [1; 1, 1, 1, ...] or sqrt(2) = [1; 2, 2, 2, 
>...]. This example I found at 
>http://www.maths.surrey.ac.uk/hosted-sites/R.Knott/Fibonacci/cfINTRO.html 
>45/16 = [2; 1, 4, 3] I then went on clumsily, further down using the 
>"strategy" (as Linda Alford put it):     ln=. 2 1 4 3  NB. list 
>notation (shorthand)     ] ip=. {. ln  NB. get integer part 2     ] 
>fp=. }. ln  NB. get fractional part 1 4 3     ] nn=. (# fp) $ 1  NB. 
>produce additional numerators (all 1) 1 1 1     ] fpn=. ,nn 
>,.fp  NB. "interweave" fractional part with added numerators to 
>yield 1 1 1 4 1 3     ] lne=. ip ,fpn  NB. "list notation 
>(expanded)" value for the Cont. Fraction function 2 1 1 1 4 1 
>3     cf=. +`% /     cf lne 2.8125     ^.^ 45r16 2.8125 -M At 
>2016-03-11 08:54, you wrote: >Maybe...  The new list interweaves the 
>two original lists. The >"strategy" also applies to lists of 
>numbers. Linda -----Original >Message----- From: 
>Programming >[mailto:programming-bounces at forums.jsoftware.com] On 
>Behalf Of Henry >Rich Sent: Thursday, March 10, 2016 6:58 PM 
>To: >programming at jsoftware.com Subject: Re: [Jprogramming] 
>Interweave >You're right... Now, how would you describe the 
>order?  Would 'index >order" be enough?  The Dictionary says "normal 
>order" and then goes >on to define it.  I have looked for years for 
>the right words.  It >has to be short and intuitive - not a 
>mathematical definition. How >about "in the order in which they 
>appear in y"? Henry Rich On >3/10/2016 6:30 PM, Nicholas Spies 
>wrote: > This is perhaps a >quibble, but... > > Looking at the NuVoc 
>definition of , y Ravel > > >The following statement and the example 
>directly below it... > >-------- > Produces a 
>list ><http://code.jsoftware.com/wiki/Vocabulary/AET#List> > 
>containing >all the atoms > 
><http://code.jsoftware.com/wiki/Vocabulary/AET#Atom> >in y, in 
>order. > > An 
>atom ><http://code.jsoftware.com/wiki/Vocabulary/AET#Atom> is > 
>converted >to a 1-atom 
>list > ><http://code.jsoftware.com/wiki/Vocabulary/AET#List>, a 
>list > ><http://code.jsoftware.com/wiki/Vocabulary/AET#List> is 
>left > >unchanged, and any 
>noun > ><http://code.jsoftware.com/wiki/Vocabulary/AET#Noun> > of 
>higher >rank <http://code.jsoftware.com/wiki/Vocabulary/AET#Rank> 
>is > >flattened into a 
>list > ><http://code.jsoftware.com/wiki/Vocabulary/AET#List> > > 
>]z=: 0 >10 20 +/ i.   NB. sample noun >   0  1  2  3  4 > 10 11 12 
>13 14 > >20 21 22 23 24 > >     ,z > 0 1 2 3 4 10 11 12 13 14 20 21 
>22 23 >24 > ------------------- > > 
>...and >also... > >     ,'ace',.'bdf' > > abcdef > > > ...give the 
>first >impression that Ravel puts the items into lexical > 
>order, > > which >is misleading, as shown by the example 2 under 
>Common Uses: > > > ]a >=: 4 4 ?@$ 100 > > 89 91 1 24 > > 88 43 43 
>32 > > 85 84 27 31 > > 10 >11 49 90 > > +/ 50 < , a > > 6 > > ,a > > 
>89 91 1 24 88 43 43 32 85 >84 27 31 10 11 49 90 > > > Since NuVoc is 
>targeting new users, the >'in order' statement should. > IMHO, 
>specify explicitly the order as >being <whatever description is 
>appropriate>. > > > >Nick > > > > > > > > On Thu, Mar 10, 2016 at 
>5:48 PM, Roger Hui ><rogerhui.canada at gmail.com> > 
>wrote: > >>     ,'ace',.'bdf' >> >abcdef >>     'ace',@,.'bdf' >> 
>abcdef >> >> >> On Thu, Mar 10, 2016 >at 2:47 PM, Roger Hui >> 
><rogerhui.canada at gmail.com> >> >wrote: >> >>> 
>,|:'ace',:'bdf' >>> abcdef >>>     'ace'([: , >|:@,:)'bdf' >>> 
>abcdef >>> >>> >>> On Thu, Mar 10, 2016 at 2:42 PM, >Kip Murray 
><thekipmurray at gmail.com> >>> wrote: >>> >>>> A quickie -- >how do 
>you interweave two equal-length lists, so >that >>>> >>>>     'ace' 
>itw 'bdf' >>>> abcdef >>>> >>>> --Kip >Murray >>>> >>>> >>>> >>>> 
>-- >>>> Sent from Gmail 
>Mobile >>>> >-------------------------------------------------------- 
>----------- > > >>> --- 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



More information about the Programming mailing list