[DALI] error documents

[Norman Gray]

Pat, hello.

On 30 Oct 2012, at 19:45, Patrick Dowler <patrick.dowler at nrc-cnrc.gc.ca> wrote:

> On the topic of Accept headers, I tried using that in the past but it is hard 
> t make it work in practice. Some clients just have Accept: */* while others 
> specify the different types with weights and such... it's kind of a mess I'd 
> rather avoid in DALI :-)

I don't think it's really that bad.

I've implemented the whole server-side algorithm in under 170 lines of code, including weights and a from-scratch parser for the header.  It took a certain amount of thought, and an assortment of test-cases, but I think it was still only a day's work or so.  And if anyone's doing it in Java, there'd surely be pre-existing library support.

So I think I'd say 'slightly intricate' rather than a mess.  And in exchange, it gives lots of flexibility to the client.

Also, supporting the Accept headers needn't be mandatory -- a server can say 'this is what I've got, take it or leave it'.

All the best,

Norman


-- 
Norman Gray  :  http://nxg.me.uk
SUPA School of Physics and Astronomy, University of Glasgow, UK