Re: [CVS ci] string_set is back for a faster live - not only

[Leopold Toetsch]

Brent Dax wrote:

Leopold Toetsch:
# The questions are (and this is IMHO the same problem with PMCs):
# - set vs assign
# - what should this program do

This idea may be totally on crack, but why do we even have S and P
registers as pointers?  What if the S registers were basically just
STRING[32] and the Ps were PMC[32], and 'set' was a COW alias-everything
operation?  This might boost performace, ease memory allocation, and end
poverty, disease and war.

Or not.

Whatever.

Interesting idea, I too had some time ago. I think, this wont work. How to you store a PMC/STRING in an aggregate then.

leo