Re: Additional Header(s) -- is 1024 bytes a brick wall?

[hbquikcomjamesl]

I've recently run into a situation in which I've had to expand the size of what I'm sending with the "ADDL_HDR" hook. And worse, my web service people haven't given me an absolute ceiling on what I might have to send.

So is the 1024-byte parameter length for the http_xproc hook identified as "HTTP_POINT_ADDL_HEADER" a brick wall?

[Scott Klement]

No, the first parameter to the callback for HTTP_POINT_ADDL_HEADER can be a VARCHAR(32767) parameter.

Will that work for you?

[hbquikcomjamesl]

Thanks. It is actually entirely possible that 1024 works just fine; the problem is that our web service developers are in another time zone, and I don't have an absolute ceiling (I just know that bumping up a particular header value from 64 characters to 80 solved the immediate problem). And they weren't aware that changing the size of a required additional header, without warning me, would break anything (it broke everything, and over the course of about 6 hours yesterday, I had to tweak close to 100 separate programs to get them to process that additional header properly).

[Scott Klement]

When planning the size of variables, it's a good idea to think about the largest size something can ever be (vs. just what it is at the moment.)