Relative times containing years fail when computed from a leap day

[schoen]

When it is February 29 (like today!), parsedatetime.Calendar().parse("1 year") raises

ValueError: day is out of range for month

(It's trying to find February 29 in the following non-leap year, i.e., February 29, 2017.)

I think this is unexpected behavior because people would expect to be able to specify relative time intervals in terms of years on any day. (For example, we had written code that always computes parse("1 year") relative to the present day, and it crashes when run today!)

[idpaterson]

I'm getting Feb 28, 2017 – are you using a specific version, or master?

[schoen]

@idpaterson, 1.5 from PyPi.

[bear]

Last week I was thinking we should push a new version...

[mistercrunch]

+1 just hit this problem today

[mistercrunch]

I just confirmed that this is fixed in master who can push a version?!

[idpaterson]

@mistercrunch there are a lot of changes in master, so that's probably not going to happen today but I'm glad to hear that it's on your mind, @bear!

[mistercrunch]

Maybe cherry pick the commit that fixes this and push a hotfix to pypi?

[mistercrunch]

It master stable? I was going to install it on my production boxes.

[idpaterson]

It must have been an incidental fix; there are no commit messages, comments, or test cases specifically addressing the leap day. A pull request against version 1.5 may be appropriate if anyone affected has time to find and patch the issue.

I have been using parsedatetime master in production on an Alfred workflow since Feb 2015 with no problems (primarily using parse and nlp against today-relative user input dates) currently up-to-date with the latest master. The test coverage is excellent though datetime-related edge cases are inherently difficult to identify and test.

[bear]

I can push a version now if that is needed.

[bear]

v2.0 has been released

[mistercrunch]

@bear saves the day. See you in 4 years!

[bear]

👍 marking issue as closed!

[schoen]

@bear, thanks for the new release. I agree it fixes this problem, although please also see #156 because our project hopes to support Python 2.6 (which has a problem with 2.0).

[bear]

@schoen I will look at that now

[mistercrunch]

@schoen , many libs are dropping support for 2.6, you should consider moving up to 2.7 or go straight to 3.4