What’s the best way to pull out multi-part form data from a Cohttp request, keeping headers for each content block? Are there any helpers implemented right now, or is it up to the user to split the body apart and reconstruct a higher-level type?
As far as I know (but I did not follow the development of Cohttp for a long time), Cohttp does not implement the
multipart. I sure than Ocsigen implements this feature. However, the
multipart is not easy as I tried with Mr. MIME (which one implement the
multipart but for e-mail) and it’s like a big recursive headache.
I don’t know the plan about Cohttp and, if httpaf will implement this feature too.
Ah, that’s great. Would love to see it integrated.
Having used it though, running into an issue with
\r\n being mangled in-between the multipart boundaries: https://github.com/cryptosense/multipart-form-data/issues/17
Right now, the individual body pieces are split on
\r\n, and each “line” is turned into a stream. I need to interpose streams of
\r\n before concatenating them. Is there an obvious way to do it?
Lwt_stream.concat doesn’t take the same argument as
String.concat for an in-between-elements separator.