Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

AttributeError: 'unicode' object has no attribute 'rpartition' #87

Closed
jonathanunderwood opened this issue Apr 1, 2016 · 7 comments
Closed
Labels

Comments

@jonathanunderwood
Copy link

If the remote end doesn't have python 2.7, sshuttle will bail with:

Traceback (most recent call last):
File "", line 1, in ?
File "assembler.py", line 18, in ?
AttributeError: 'unicode' object has no attribute 'rpartition'
client: fatal: server died with error code 1

Ideally, sshuttle would support python >=2.4 on the remote end in order to support RHEL5 and RHEL6 remotely. Failing that, a more user friendly error message informing the user that the remote end python is too out of date would be nice.

A downstream bug report is here

@jonathanunderwood
Copy link
Author

PR #81 is progress towards fixing this issue.

@vieira
Copy link
Contributor

vieira commented Apr 1, 2016

I opened #37 with some fixes to make it work but it broke some tests and there wasn't much interest at the time. I thought I was the only remaining Python 2.4 user in the world. 😜

I rebased sometime in January and opened #56 and @brianmay was kind enough to fix most of the tests in #81 even though I think he is not affected by this nor very excited about maintaining Python 2.4 code.

#83 should have fixed the last few problems and one or two tests. I have also tested manually against a range of Python versions from 2.4 to 3.5 and it seems to be working nicely.

Do you still have any problems with #83?

@jonathanunderwood
Copy link
Author

I can understand the lack of excitement about maintaining python 2.4 code :). I wonder if a half way compromise might be to support python 2.6 - that covers rhel6 era distros, and tools such as six.py and pasteurize make it fairly painless.

Is the only blocker on merging #81 the fixing of the tests? If so, when I am back from travels, I could have a look at pitching in there.

@vieira
Copy link
Contributor

vieira commented Apr 2, 2016

AFAIK in #83 (which includes all commits from #81 plus some extra fixes) all tests are OK. It should work with all Python versions from 2.4 to 3.5. @brianmay already fixed the tests which I think was one of the blockers. He might have other considerations/reasons for not wanting to merge, I'm not sure.

Although I have tested #83 against Python 2.4-3.5 on a Debian server, if you can also test and report any problems it would be nice (clone vieira/sshuttle, checkout old_python).

@jonathanunderwood
Copy link
Author

OK, I have taken current git from Brian and added the vieira/old_python commits on top and built a new Fedora package for testing, so hopefully that'll help shake out any problems. In case you're interested, the build is here:

http://koji.fedoraproject.org/koji/taskinfo?taskID=13540955

@racker-tyler
Copy link

I also noticed this issue when installing it fresh from brew for mac. It appears that the server I am connecting to is using python2.4 :( I doubt that itll get updated anytime soon either. Would I need to reach out to the owners of the homebrew repo's to copy the commits and build a new repo or is this going to be incorporated into a newer version?

@brianmay
Copy link
Member

brianmay commented Apr 8, 2016

Released version 0.78.0; closing.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

4 participants