Closed benanderman closed 7 years ago
@benanderman, thanks for the report. Looks like idna 2.6 was released yesterday -- yay for upstream breaking my release on the same day.
This looks like a sympom of https://github.com/pypa/pip/issues/988. basically what's happening is:
I've pushed a workaround where I just list requests first, which fixed the problem on my local machine. If it passes travis and works for you, I'll tag it as 0.5.1 and release it. I'd like a less precarious solution, but off the top of my head I can't really think of one short of freezing the entire transitive dependency graph (note that just doing exact versions one level deep wouldn't prevent this), and I don't like that.
It works for me — thanks for the quick response!
You're welcome! I've tagged/pushed/uploaded a new release with the fix. I'm going to close this.
FYI, the conflict seems to be between cryptography
and requests
requirements.
cryptography
has requirement of "idna >= 2.1"
which pulls the most recent version
Until recently requests
had requirement of "idna<2.7,>=2.5"
, and that caused the issue.
pip install 'idna<2.9,>=2.5' This work for me!
pip install 'idna<2.9,>=2.5' This work for me!
On Ubuntu 16.04.2 LTS, I did a fresh clone,
./bootstrap.sh
, and./venv.sh
, and I got this error when trying to runsimp_le
:It was fixed like this:
But this seems like it shouldn't be necessary (note,
idna-2.6
was installed in the venv, not in the system).