Skip to content

Test py36, spark 2.1.1 on travis#33

Merged
parente merged 2 commits intoadtech-labs:masterfrom
parente:spark-2.1.1
May 12, 2017
Merged

Test py36, spark 2.1.1 on travis#33
parente merged 2 commits intoadtech-labs:masterfrom
parente:spark-2.1.1

Conversation

@parente
Copy link
Contributor

@parente parente commented May 12, 2017

Fixes #27.

@codecov
Copy link

codecov bot commented May 12, 2017

Codecov Report

Merging #33 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master      #33   +/-   ##
=======================================
  Coverage   85.35%   85.35%           
=======================================
  Files           6        6           
  Lines         403      403           
=======================================
  Hits          344      344           
  Misses         59       59

@parente
Copy link
Contributor Author

parente commented May 12, 2017

@basnijholt the jupyter kernel test and unit test suites pass with py35 and py36 on travis. I'm updating the Makefile to create a py36 environment for development and will poke around a bit in live notebooks.

@parente
Copy link
Contributor Author

parente commented May 12, 2017

Running locally, it looks like there's an issue in metakernel's shell_magic when spawning a spylon-kernel in Python 3.6.

[MetaKernelApp] ERROR | Can't load '/Users/parente/miniconda3/envs/spylon-kernel-dev/lib/python3.6/site-packages/metakernel/magics/shell_magic.py': error: Timeout exceeded.
<pexpect.pty_spawn.spawn object at 0x109706f98>
command: /bin/bash
args: [b'/bin/bash', b'--rcfile', b'/Users/parente/miniconda3/envs/spylon-kernel-dev/lib/python3.6/site-packages/metakernel/bashrc.sh']
buffer (last 100 chars): '\x1b[?1034h(spylon-kernel-dev) '
before (last 100 chars): '\x1b[?1034h(spylon-kernel-dev) '
after: <class 'pexpect.exceptions.TIMEOUT'>
match: None
match_index: None
exitstatus: None
flag_eof: False
pid: 88201
child_fd: 54
closed: False
timeout: 30
delimiter: <class 'pexpect.exceptions.EOF'>
logfile: None
logfile_read: None
logfile_send: None
maxread: 2000
ignorecase: False
searchwindowsize: None
delaybeforesend: 0.05
delayafterclose: 0.1
delayafterterminate: 0.1
searcher: searcher_re:
    0: re.compile("[$#]")

The kernel eventually comes up, but there's a long delay. I'll need to produce a smaller reproducible case and file upstream.

@parente
Copy link
Contributor Author

parente commented May 12, 2017

The exception above is not due to py36. Something changed in my OSX setup that is causing pexpect to fail. I can reproduce it on py35 now too, but not when I switch to my Linux dev environment. So ... not important right now.

@parente parente merged commit 6f62b12 into adtech-labs:master May 12, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant