RE: Netscape Javascript Problem

["Jonathan Shore" <jshore@example.com>]

> From: Stephen J. Turnbull [mailto:turnbull@example.com]
> Well, lack of logic just seems logical[sic] to me; we _are_ talking
> about Javascript, aren't we?

heh - paper clips, rubber bands, and glue - too bad.

> Javascript does not understand locales.  ("Content-Type:
> ..;charset=iso-2022-jp" would have similar effect if Javascript did
> understand locales.  Seems unlikely, but you might check.)

Actually had tried working with each of the 8-bit formats (UTF-8, EUC,
SJIS), all with similar results. I can display these correctly when using
literals but not when programmatically generated.  [I also checked carefully
to make sure that the sequences were correct, displaying the ordinal values
of each character in the string].


> Use 7-bit ISO-2022, force designations of everything except the
> initial ASCII.  This will be mildly inefficient (lots of redundant
> escape sequences) but has some hope of working portably.

I was hoping to avoid 7-bit jis due to the messy escape sequences (from a
programming point of view) but will give it a shot.  It was easier to
manipulate EUC and unicode because could easily programmatically detect
double-byte versus single byte characters in the text.  At this point any
solution is better than none.

Thanks for the advice -

JS