Skip to content

Conversation

@ghost
Copy link

@ghost ghost commented Jan 29, 2014

More fixes to @network to handle the incredibly unreliable yahoo servers.
This should do it.

The mind boggles that a billion-dollar company can have a QoS this bad.

N.B. the default socket timeout setting.
It's not required for this because I believe travis has some firewall rules that
timeout net connections, but it's nicer when running locally where connections
can hang for much longer. 15 seconds is very generous so I hope it's not disruptive.

cc @jreback, @jtratner

ghost pushed a commit that referenced this pull request Jan 29, 2014
@ghost ghost merged commit 1f3efa1 into pandas-dev:master Jan 29, 2014
@ghost ghost deleted the PR_more_network branch January 29, 2014 17:31
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nice way to hack around this :)

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

desperate you mean. I can't believe how screwed up this corner of python is,
you literraly have to do all this to catch these network errors. Timeouts can bubble
up in at least 5 different ways, diff exceptions, with errno or with out. nightmare.

The code for @network, and the kwarg lists were really nice to see after
wallowing in the muck of indexing.py. kudos.

This pull request was closed.
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