Georg Holzmann wrote:
Hallo!
of course this (like [l2s]) will polute the symbol-hash and (what is more important) will add a memory-leak (as symbols that once appear in a pd-sessions will stay there forever)
yes ... :(
it is not a very big problem, i think (of course a memory-leak, or what appears to be a memory-leak, is always a problem)
in most cases you will not manage to fill up your memory (only if you really generate a lot of symbols on the fly)
so a string-type would really be nice.
anyway, would it make much sense, if we use lists for that? e.g.
[list this is a long text over more lines( | [maxchar 10] (or a more meaningful name ...) | [list this is a \n long text \n over more \n lines( | [text2d]
otoh, TextBase generates a string out of the incoming list anyhow. it does not care whether the "\n" is a separate symbol or hidden within a symbol: "bi \n ba" is the same string (on the TexBase-side) as "bi" "\n" "ba" (using " " as a separator). both should result in: "bi ba"
(but none is "bi\nba"!)
and i do think [linebreak] is a better name than [maxchar] ;-)
mfga.sd.-r IOhannes