Retry on network failure on windows

unread,

Feb 14, 2016, 11:08:04 AM2/14/16

to watchdog-python

Hi,

I needed to make a change so that the observer does not crash but keeps retrying if the

drive is disconnected, on Windows. So I am catching WindowsError with winerror == 64,

on get_directory_handle and read_events, and in such case I close the handle and retry

in, let's say 5s. With the current watchdog I have to clone the whole read_directory_changes.py,

so I am proposing a change:

to make the replacing of the windows api calls easier.

Would you guys care to have this retrying observer in watchdog? If yes I will prepare another

pull request. Should that replace current WindowsApi(Emitter|Observer) or should those 2 live

in parallel?

Best regards,

Petr

Message has been deleted

unread,

Jun 22, 2016, 2:45:24 PM6/22/16

to watchdog-python

This would be a real nice feature to have! But I'm totally new into this stuff, so I hope someone, who is competent finds this interesting too.