neat.distributed: use socket module instead of multiprocessing.managers #125
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
resolved Conflicts: neat/distributed.py
remove the travis_wait command. IIRC, this was once added to fix a few bugs with distributed+multiprocessing on travis. Since i want to completely fix distributed, these commands should no longer be required. This should improve the runtime of the tests and allow realtime output of travis.
…t improvements and fixes
|
Apologies for the long silence--I've been overwhelmed with other stuff and just kind of let this fall idle. I appreciate the patch and will take a look as soon as I get some free time. Thanks! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I finally found time to complete the rewrite of
distributed.py.This PR replaces the usage of the
multiprocessing.managersmodule with thesocketmodule.Here are some advantages of this change:
These changes are mostly compatible with the old API, however the type of the
authkeyargument has changed frombytestostr/unicode.Also, this closes #101.
Edit: It seems like one of the tests still fails randomly. The travis build prior to the PR still succeeded.
Edit 2: Seems like the travis failure was caused by some process timeout. After restarting the build, the tests pass, but coverage is still reduced. However, most of the missing coverage seems to be from error catching code, which cant easily be triggered manually (e.g. handling of invalid network messages).