Support to Chunk-Encoded Requests
Hi !
When we could see in __roast_headers__, on each request `Content-Length is automatically append in the header request.
even if the roast file is like :
POST http://google.fr
PUT http://google.fr
DELETE http://google.fr
Note: for GET http://google.fr the Content-Length header is not added.
It's a big problem because if we want specify the header Transfer-Encoding: chunked it cause a conflict.
The HTTP specification says:
"Expert Info (Warning/Malformed): The Content-Length and Transfer-Encoding header must not be set together" (message from Wireshark).
See also https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Transfer-Encoding
Maybe it could be cool to have a way to not add automatically this HEADER field.
Thanks in advance.
With curl (therefore with https://github.com/diepm/vim-rest-console/ ) to unset a field header we could simply use something like
Accept:
For instance thanks Wireshark we could know that
curl http://google.com appends some headers automatically like Accept: */*. To remove it simply use
curl http://google.com -H 'Accept:. In Wireshark we see that the header Accept is not append.
Hi Julio, I'm not explicitly adding a Content-Length header. It might be the requests library that's doing it. Can you give an example with setting the Transfer-Encoding header that demonstrates this problem please?
@sharat87
If you want an example, use
Transfer-Encoding: chunk
POST http://localhost:3030
You could see in the __roast-headers__ window that an header with Content-Length was appended.
As I said in the first comment, chunked request have this problem. A chunked request should not have the Content-Length header (see https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Transfer-Encoding).
It seams that you must do something like http://2.python-requests.org/en/master/user/advanced/#chunk-encoded-requests
But probably it's not so important. Support Chunk-Encoded request is probably not a goal very important. Maybe nobody is interesting in this feature.