[Calypso] Incorrect parsing of rrule?

Petter Reinholdtsen pere at hungry.com
Tue Jan 26 00:27:45 PST 2016

[Guido G√ľnther]
> I checked that callers and they don't distinguish False and None so this
> looks good.

Good.  I checked it too, and hope this part can be commited right away
to reduce the number of patches I need to track. :)

> Since this all ends up in answering the report request from a client I
> think we should set a proper response code (400) in report() for that
> particular imtem of the MultiStatus repose. just as we unconditionally
> set 200 atm. Does this make sense?

Make sense to me.  Throwing any exception will cause do_REPORT to send

> We don't have calypso specific exception's yet. Is it time to introduce
> these or do we just raise ValueError?

In this case, I guess ValueError will do just fine.  Rewriting.

>> +        # Tried calling do_REPORT() directly, but lacked the arguments
>> +        # needed to get the CollectionHTTPHandler class working.  Use
>> +        # match_filter() directly instead.
> Hmm...either we want to teaat do_REPORT than we should try harder or we
> just want to exercise match_filter in which case i would leave the
> comment out.

I'll drop the comment.  This test started out with an intention to test
do_REPORT(), but now it is testing match_filter() instead.  We can (and
probably should) write another script to test do_REPORT(). :)

> Given we do the exception raising from above we should also test if we
> can catch that here.

Done. :)

Attached is a new and improved version of

Happy hacking
Petter Reinholdtsen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-Allow-time-range-query-filters-to-be-open-ended.patch
Type: text/x-diff
Size: 7530 bytes
Desc: not available
URL: <http://keithp.com/pipermail/calypso/attachments/20160126/c93c5e43/attachment.patch>

More information about the Calypso mailing list