It sounds like an attempt to simplify document production by requiring encoding of characters that sometimes need to be encoded, regardless of whether the current context actually requires it. The attempt is misguided because it indicates that their document processors are not in compliance with the XML specification, which requires changes to the behavior of serializers (like Xerces) that encode only as necessary. You'd be doing a public service by pointing out to them how broken their approach is. As Alberto pointed out, if they refuse your documents, they're not actually supporting XML, but something derived from it. They lose all the benefits of using a standardized format and gain nothing.
If you really can't get them to accept the valid XML that Xerces generates, hacking the serializer as Alberto recommends is probably the easiest and least harmful workaround. -----Original Message----- From: Joerg Toellner [mailto:[EMAIL PROTECTED] Sent: Monday, March 06, 2006 9:31 AM To: [email protected] Subject: AW: Translation of some special characters in their entities won't work. Do i sth. Wrong or is it a bug? Dear Alberto, Let me first say thanks to you for your speedy reply. The problem is, that i'm using xerces to create some xml-documents where i have strict guidelines from outside my competence. The guidelines come from a german healthcare authority organization and i have to send in my documents for proofment of correctness to get a permission for our software to be used in medical healthcare here in germany. I can't change anything on the rules. They were for me as a law from government. I think starting a discussion that this characters won't hurt anything is useless. You know they are "officials" and not really thinking people. :-( I got my first sent in package back from them and they critize that i haven't encoded this characters in entities as forced by the documentation of the rules. They demand to encode to entities all 5 characters (<, >, ', " and &) and so i have to do it, if i want it or not. If you say, that this is intended behavior, i think i have to do a workaround like scanning for this two characters in my values before hand them over to xerces, and then xerces will do the rest for me. Is it legal/possible to give xerces a string like "rate is => 60% and <= 80%" for a attribute value? Will xerces then accept the already encoded entity and leave it as is while translating with fun the "lower than" char? Again, thanks very much for your time Greeting Joerg Toellner -----Ursprüngliche Nachricht----- Von: Alberto Massari [mailto:[EMAIL PROTECTED] Gesendet: Montag, 6. März 2006 13:59 An: [email protected] Betreff: Re: Translation of some special characters in their entities won't work. Do i sth. Wrong or is it a bug? Hi Joerg, that's the expected behavior; inside attribute values, apostroph (when the attribute value is delimited by quotes) and 'greater than' are not ambiguous symbols, and can be used directly. Anyhow, why does this trouble you? Alberto --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
