Mathematica 9 is now available
Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
1996
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*December
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 1996

[Date Index] [Thread Index] [Author Index]

Search the Archive

Enhancements of Combinatorica, Re:List help needed

  • To: mathgroup at smc.vnet.net
  • Subject: [mg5114] Enhancements of Combinatorica, Re:[mg5065] List help needed
  • From: Allan Hayes <hay at haystack>
  • Date: Wed, 30 Oct 1996 22:04:22 -0500
  • Sender: owner-wri-mathgroup at wolfram.com

John Rowney <jrowney at inetg1.Arco.COM>
[mg5065] List help needed
writes

> What I would like to do is the following:
> N by combining adjacent elements of the original list into sub
> lists. This might become clearer with an example.
>
> Given {a,b,c,d,e} of length 5, ALL possible length 4 lists  
subject > to the conditions above are:
> {{a,b},c,d,e}, {a,{b,c},d,e}, {a,b,{c,d},e} and {a,b,c,{d,e}}

> two of the possible length 3 lists are
> {a,{b,c,d},e} and {{a,b},c,{d,e}}

John,

The function Segments, below does what you seem to want.
I include new versions of two functions from the package  
DiscreteMath`Combinatorica` :

Allan Hayes
hay at haystack.demon.co.uk
http://www.haystack.demon.co.uk/


***********	

CombinatoricaPlus

Enhancements to the package  DiscreteMath`Combinatorica`
Copyright Allan Hayes 1996

(*Segments*)

Segments::usage =
"Segments[s, k] for a list s and integer k with 0<k<=Length[s] gives 
a list of k segments of s ,{s1,s2,...},with Join[s1,s2,...] = s.
Segments also works when s is of the form h[x1, ...].
Segments[h[0],0] = {h[0]}.
"

splits[{x___,s_}] :=
Segments[h_[],0] := {h[]};
Segments[s:h_[__], k_Integer]/;0<k<=Length[s] :=
Examples:
Segments[Range[4],2]

Segments[Range[12],6];//Timing//First

Segments[h[1,2,3,4],2]
(*Compositions*)

<<DiscreteMath`Combinatorica`

Segments::usage =
"Compositions2[n, k] for positive integer k,and non negative
integer n gives a list of all lists of k non-negative integers
{n1,n2,...},with n1+n2 + = n (each of these lists is called a
composition of n).
Compositions2[0,0] = {{}}
";

comp[{x___,s_}] := Thread[{x,#,Reverse[#]}]&[Range[0,s]];
Compositions2[0,0] = {{}};
Compositions2[n_Integer?NonNegative,k_Integer?Positive] :=

Compositions[4,2]
Compositions2[4,2]
Compositions[12,6];//Timing//First
Compositions2[12,6];//Timing//First


(*KSubsets*)

KSubsetsList::usage =
"KSubsetsList[l, {m, n}] for non negatives integers m, n with m <= n 
gives {Sm,S(m+1),...,Sn} where Si is the list of all subsets of
the list l with exactly k elements, ordered lexicographically.
\nKSubsetsList[l, {m}] = KSubsetsList[l, {m,m}];
\nKSubsetsList[l, m] = KSubsetsList[l, {m,m}][[1]];
\nKSubsetsList[l] = KSubsetsList[l, {0,Length[l]}] = all subsets.
";

KSubsetsList[_,{0,0}] := {{}}
KSubsetsList[{},___] := {}
KSubsetsList[l_,{n_Integer?Positive,n_}]/;n== Length[l] := {{l}};
KSubsetsList[l_,{1,1}] := {List/@l};
KSubsetsList[l_,{n_Integer?NonNegative}] := KSubsetsList[l,{n,n}]
KSubsetsList[l_,n_Integer?NonNegative] := KSubsetsList[l,{n,n}]//First
KSubsetsList[l_] := KSubsetsList[l, {0, Length[l]}]
KSubsetsList[l_,{n_}] := KSubsetsList[l, {n,n}]
KSubsetsList[l_,{m_Integer?NonNegative,n_Integer?NonNegative}]/;m<=n:=  
  Module[{c = 0,step,ln = Length[l]},	
      step[d:{a___,_}, p_] :=
         Apply[Join[Thread[{p,#2}],#1]&,
      Map[
         Flatten,
      ]
   ]

KSubsetsList[Range[4],2]
KSubsetsList[Range[4],2]
KSubsetsList[Range[4],{1,3}]

{{1, 2}, {1, 3}, {1, 4}, {2, 3}, {2, 4}, {3, 4}}
{{1, 2}, {1, 3}, {1, 4}, {2, 3}, {2, 4}, {3, 4}}
{ {{1}, {2}, {3}, {4}},
  {{1, 2}, {1, 3}, {1, 4}, {2, 3}, {2, 4}, {3, 4}},
  {{1, 2, 3}, {1, 2, 4}, {1, 3, 4}, {2, 3, 4}}
}

KSubsets[Range[12],6];//Timing//First
KSubsetsList[Range[12],6];//Timing//First
3.31667 Second
0.433333 Second
******************




  • Prev by Date: Re: Solve results
  • Next by Date: Re: arguments
  • Previous by thread: Integration..Constants
  • Next by thread: Re: arguments