Ludovic Courtes wrote:
> > Implementations of the R6RS may use variable-width
> > representations while achieving O(1) amortized time for
> > string-length, string-ref, and string-set!.
>
> Forgive my ignorance, but could you point me to algorithms you have in
> mind? (I don't remember reading them on this list.)
The representations that appear most attractive for Larceny
are described at
http://larceny.ccs.neu.edu/larceny-trac/wiki/StringRepresentations
Many other representations are possible, including many
that would work well in interpreted systems but were not
considered for Larceny because the string-ref and string-set!
procedures would become a little too complex to be inlined.
Will
Received on Wed Aug 29 2007 - 08:28:47 UTC