Cory Benfield
Cory Benfield
Requests makes no promises about leaving headers in the form they are received from the server. Right now, I highly recommend you use the cookiejar Requests uses to store cookies...
I've taken a quick look at the code in urllib3 and I can't find any reference to the `*_PROXY` environment variables there. Can you point to where in the code...
It does ignore `trust_env` and the original proxies dict (both bugs), but you're not running that code unless you're using the requests release from GitHub. =) Fixing both of those...
To be clear for those who aren't sure, the way @ouroborus' suggestion differs from the current logic is that we take the proxies from the request, then apply proxies from...
No, and IIRC (I'm on my phone) it's been that way for a while. Looked like an oversight to me.
@sigmavirus24 Not that I can see. =)
This is an entirely reasonable suggestion. Honestly, I'm not averse to doing it: there are definitely worse things to do than this. I'd be open to providing a default timeout....
@kuraga No. Per @sigmavirus24, and in many many previous discussions: > I think I could actually still implement it using hooks - create a method which uses the prepared request...
Argh, sorry, copy-paste fail: > I'm still pretty strongly against sessions having a timeout attribute though and I think that shouldn't muddy the waters of this discussion because it will...
It's extremely unclear to me what "wrong" means here. They aren't wrong: they work, as designed. The introductory examples are all clearly typed into the Python interactive interpreter, so they...