SORTBREAK (v1.0, nov 2008)

sort elements of a vector between break points

Y = SORTBREAK(X,BP) sorts the elements within sections of the vector X
in ascending order. The sections are defined by the breakpoint
indices BP. Each section is sorted independently.

If BP is a single index, the two sections are X(1:BP) and X((BP+1):end).
Example:

Y = sortbreak([5 3 7 1 4 2 8 6],4)
% Y -> [1 3 5 7 2 4 6 8]

If BP contains more indices, the indices are used in sorted order, and
the first section contains the elements X(1:BP(1)), the second
section contains the elements X(BP(1)+1:BP(2)), etc. The last
section contains the elements X(BP(end)+1:end). Example:

Y = sortbreak([5 3 7 1 4 2 8 6],[2 6])
% Y -> [3 5 1 2 4 7 6 8]

SORTBREAK(X,BP,'fixed') excludes the break points from sorting. Example:

% sort between zeros
X = [3 1 2 0 3 1 -4 0 6 4 7 0 5 3] ;
Y = sortbreak(X,find(X==0),'fixed')
% Y -> [1 2 3 0 -4 1 3 0 4 6 7 0 3 5]
% whereas Y = sortbreak(X,find(X==0))
% would yield [0 1 2 3 -4 0 1 3 ...]

[Y,I,J] = SORTBREAK(X,BP) returns sorting indices I and J such that
that Y equals X(I) and X equals Y(J).

Notes:
- Break point indices larger than the number of elements in X are
ignored.
- If BP is empty, the whole vector is sorted.
- For arrays, X is treated as a vector, but Y, I and J will have the
same size as X.
- To sort between NaNs, the third argument 'fixed' is not needed as
NaNs are always sorted at the end of the sublist
X = [3 1 2 NaN 3 1 -4 NaN 6 4 7 NaN 5 3] ;
Y = sortbreak(X,find(isnan(X)))
% Y -> [1 2 3 NaN -4 1 3 NaN 4 6 7 NaN 3 5]

See also : sort, sortrows, issorted
randpermbreak (File Exchange)