Mark et al,
The browsers are fairly reliable in converting data input in an HTML form into
the declared charset of the form. A hidden field with the charset value works
very well in this case - or - if the form is dynamically generated, so too is
the HTTP header and/or META tag with the charset, which means the program
already knows the charset of the form. If the CGI is persistent, this
information is simply accessed from a stored variable. If the CGI is not
persistent, the hidden field can be generated as part of the form.
We have used this technique with success in a couple of client applications,
both with persistent and non-persistent CGIs.
Andrea
-- Andrea Vine, avine@eng.sun.com Sun-Netscape Alliance i18n architect Necessity is the mother of strange bedfellows. -- Dr. Dave Farber (father of SNOBOL and one of the creators of Token Ring)mark.davis@us.ibm.com wrote: > > Unfortunately, as has been noted on this list, there is no explicit > reliable way to determine the character encoding of a posted form in HTML. > A common technique for handling this is to have a hidden field with known, > fixed contents. When these contents are analysed, the encoding can be > heuristically determined. > > What I would like to see is if anyone has developed a set of minimal > contents of one or more hidden fields that suffice to distinguish all major > character sets? > > Mark
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:58 EDT