cache.incr may raise ValueError about durabledict

[Fluxx]

Hmmmm, yes that's a good point. I will fix this either today or tomorrow.

[DXist]

By the way, why does DurableDict get method call __dict.pop? If cache is not expired __dict will not have popped key.

[Fluxx]

@DXist can you link to the specific line in question?

[DXist]

Here you are

[Fluxx]

Ah, that appears to be a typo/bug. I'll also fix that too.

On Thu, Apr 11, 2013 at 1:17 PM, Rinat Shigapov [email protected]:

Herehttps://github.com/disqus/durabledict/blob/master/durabledict/base.py#L41you are

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-16258570
.

[Fluxx]

@DXist this is fixed in 0.7.1. Please note that there is a fundamental distributed computing impossibility in how the ModelDict uses the cache to maintain last_updated so this solution is not 100% foolproof, but should be good enough for most use cases, especially ones with a low write volume.

[DXist]

Thank you! Why do you use ">" condition in https://github.com/disqus/durabledict/blob/master/durabledict/base.py#L23 .
Why not just require inequality between persistance_last_updated and last_synced?