Jan Schulz

We have a webform which is processed by an agent. In this webform you can add a message. Every few times this agent failed to process the form (fortunately the raw message was copied in that case and so no message was lost). The problem is, that the messages were to long: @URLDecode can only process strings which are no longer than 2048 chars.

Unfortunately @URLDecode is used in several WebSession classes to access the form content, so have a look if this is a problem for you, too...

I opened a Service Request, as this wasn't specified in the designer help and the error message shown in the client sounds more like bug than a feature: "Type mismatch in method CoerStrToNum: STRING found, double expected".

Update 21.09.2009:
IBM looked at it and found that evaluate is the problem: it can't handle more than 2064 bytes. So a fix could be to add the string to decoded to a doc and pass that doc to evaluate and let @URLdecode get the string from there. Another would be to split the string, but then I would need to understand, where would be a good value to split it: it does not be in the middle of a multi byte decoded umlaut...