[SciPy-User] scikits.timeseries on windows: invalid object used in slice

[Pierre GM]

On Apr 12, 2011, at 5:49 PM, Matt Knox wrote:

>>> Is this still the appropriate place to report bugs in the timeseries
>>> code? Where is the timeseries code hosted now?
>> 
>> Still on the SVN, officially. There's a github spot that I keep on intending
>> to use, but I'm a tad swamped those days...
>> 
> 
> Hi Pierre. Given that we have both been inactive on this project for a while,
> perhaps it is best that we expedite the move to github and grant commit / merge
> privileges to any interested maintainers and retire the svn repo. Then at least
> interested parties can pick up the slack if they want.

I created two different github projects. One is just the SVN ported to git last September, accessible here:
https://github.com/pierregm/scikits.timeseries 
I have a second one in parallel, timeseries-sandbox, which contains some cool additions (irregular frequencies) and some brilliant failures (DateArray as a pure subclass of ndarray in C. Nope, I can't speak good C enough...). My idea was to get something as close as the new dtype as possible, but, well. Fail.

> If you are ok with this, I can look after coordinating the details of the
> transition.

Yay! A game of kick-the-baby! Hop, your turn.