Recently, an OpenInsight Web Deployment customer of mine posed a problem. Their website is accepting business card orders from around the world. A bunch of orders from Hungary appeared where many characters appeared to the U.S. customer as question marks when viewed in Advanced Revelation.

I understand that Unicode is how you have to build a multi-language product in software like OpenInsight, but the problem here seems different. The user's probably have a local font that shows the characters in their language set (code page?) so the order characters look fine to them. When it is saved in a Linear Hash file we have no idea what language set was in use by the orderer and can't readily present them.

It's not feasible for this small business to provide pages in every language that may be ordering.

I'm looking for suggestions on how we might get the proper characters, which must be printed on their business card and stationery orders? My client thinks my solution, "use a graphics program and e-mail the picture" is not feasible.

Thanks, Dave (see you in New Orleans)

Run javascript over the form contents to encode (ICONV) each character into its corresponding escape code. Trap the domain (javascript location.href) and even the IP address of the source session (IP ranges run by country). Include a field saying "the base language for this brochure to be printed in is……(a)Swahili (b) Latvian ©Klingon".

It may be that a web page using VBscript could interrogate the local registry data for base language. Failing that, you could use webdav to submit the file from the local PC and have the user specify the encoding default.

The lack of a suitable Hungarian phrase book may mean your hovercraft is full of eels.

Steve

View this thread on the forum...