ruleset.count() not changing after applying ruleset.exrule()

[anandkr07]

Seems like the ruleset.count() is getting cached after first call to it. An Example of this behavior -

>>> rules = rrule.rruleset()
>>> rules.rrule(rrule.rrule(rrule.DAILY, until=datetime.datetime(2038,1,1,0,0,0), dtstart=datetime.datetime(now.year,now.month,now.day,0,0,0)))
>>> rules.count()
8179
>>> rules.exrule(rrule.rrule(rrule.DAILY, until=datetime.datetime(2038,1,1,0,0,0), dtstart=datetime.datetime(now.year,now.month,now.day,0,0,0)))
>>> rules.count()
8179

Only after iterating over the ruleset , do we get the correct value , Example -

>>> list(rules)
[]
rules.count()
0

[pganssle]

The issue here is that rrulebase caches the length after the first call to count(), and this cache is not invalidated when changes are made to the rruleset:

from datetime import datetime as dt
from dateutil import rrule as rr

dts = dt(2015, 8, 11)
dte = dt(2016, 8, 11)

crr = rr.rrule(rr.DAILY, dtstart=dts, until=dte)

rules = rr.rruleset()
print(rules.count())
>>> 0

rules.rrule(crr)
print(rules.count())
>>> 0

rules._len = None
print(rules.count())
>>> 367

This should be easy enough to fix, I can make sure it's fixed in 2.5.0. I am in general loathe to recommend building anything upon private methods or attributes, but in the meantime, anything that iterates over the rruleset should update the count, so list(rules) or something like: for _ in rules: pass (this is what rruleset.count() would be doing anyway if the cache were properly invalidated anyway, so it shouldn't degrade performance).

Thanks for the bug report!

[pganssle]

Pushing fix to 2.5.1 to speed release of 2.5.0.