Hi all,

I'm continuing work on the HTTP proxy I'm writing. The absolute
bare basics are working with Warp, Wai and http-enumerator as long
as the upstream web server doesn't send gzipped or chunked data. For
these later two cases, httpe-enumerator helpfully gunzips/unchunks
the data. That however causes a problem.

If my proxy simply passes the HTTP headers and data it gets from
http-enumerator and passes then on to the client, the client barfs
because the headers claim the data is gzipped or chunked and the
data actually isn't chunked/gzipped.

There are a number of possible solutions to this:

 a) Strip the content/transfer-encoding header and add a
   content-length header instead. I think this is probably
   possible with the API as it is, but I haven't figured out
   how yet.

 b) Rechunk or re-gzip the data. This seems rather wasteful
   of CPU resources.

 c) Modify the Network.Http.Enumerator.http function so that
   de-chunking/gunzipping is optional.

 d) Expose the iterHeaders function that is internal to the
   http-enumerator package so that client code can grab the
   headers before deciding how to handle the body.

Are there any other options I haven't thought of yet?

>From the options I have, I actually think d) makes the most sense,
Would a patch exposing iterHeaders be accepted?

Cheers,
Erik