Closed rogerleite closed 11 years ago
i can't simulate this error because webmock's magic. See this commit for example: https://github.com/bblimke/webmock/commit/553496851cebbef505708c91f99ae1d00151a9d6
@rubiii Any idea? I started to work on this branch: https://github.com/rogerleite/httpi/tree/fix_response_headers
no idea right now, but i'll take a look at it later.
thanks @rogerleite, but i need specs for this. i'll look into getting rid of webmock. i was planning to replace it with a simple server anyway.
@rubiii i kind of extracted MinionServer to help with it. I'm using on integration tests of http_monkey. If you want to help, i can add you as collaborator. What do you think?
I need this for http_monkey, i can try to simulate this bug on integration tests. Ok with that?
have to take a look at this "after work" ;) i kind of started something similar here. don't know if we're aiming for the same tool here.
Your IntegrationServer
is really cool. The main goal is to provide a integration server that supports "rack app" and i can control the start and stop. You server is cool because also supports SSL. Many projects needs this, and always use some kind of "home made" solution.
yes, the idea was to have a basic server that can run rack apps (just like your minion server) and add n applications for every kind of http test case one can imagine (ssl, redirects, errors, timeouts, cookies, etc).
Great! The invite as collaborator is still open! :smile:
thanks :) i would really like to hear what your plans are for minion server and whether we can combine efforts. but this issue may not be the right place for that ;)
@rogerleite i removed webmock from master. hope this solves your problem.
Thanks @rubiii I'll write some tests for this issue!
Some evidence:
Using a simple script (httpi 2.0.0.rc1), i realized that only curb adapter return all "Set-Cookie" headers.
Results:
I'm planning to work on net_http and httpclient adapters to return headers like curb do. Is that ok?