ERXFormat based on headers about wonder HOT 5 CLOSED

my original version did and IIRC (it's been a couple years) i ran into problems because everything claims to accept HTML and XML, so it was never clear wtf they ACTUALLY wanted back ... if they say they accept html, do you route them to html page? at the end of the day, it ended up just being way less error prone to just use the file extension. i'm not opposed to supporting accepts headers if it's well-behaved across all the browsers, though.

from wonder.

About .html vs the other formats, I guess that if the query was made with XMLHttpRequest (X-Requested-With: XMLHttpRequest), it will not be HTML (unless people want to send HTML fragments?). But I agree that it can be a PITA because of the browsers.

For Content-Type, I did it with:

        if (type == null) {
            String contentType = request().headerForKey("Content-Type");
            if (contentType != null) {
                String[] types = contentType.split("/");
                if (types.length == 2) {
                    type = types[1];
                }
            }
        }

from wonder.

I think formatNamed will return for both the mime type and the file extension ... it's sort of a weird dictionary.

also, does this let file extensions file if you define them? i'd be fine with this, then -- backwards compatible, etc.

from wonder.

Yes, the content-type stuff is done if the file extension was not specified. I added this because I'm playing with JBoss's RestEasy to consume REST services and it's using Content-Type instead of Accept, which is kind of crazy.

from wonder.

Closing as an ancient ticket: if someone wants to adopt this idea, by all means re-open this ticket.

from wonder.