[Freeassociation-devel] default INTERVAL parameter

[Patrick Ohly]

On Fri, 2009-08-07 at 10:09 -0400, dothebart wrote:
> having run the test-suite this is what I found:
> 
> ########## Test recur parser                        (8) ##########
> not ok 209 -
> FREQ=YEARLY;UNTIL=20000131T090000Z;INTERVAL=1;BYDAY=-1TU,3WE,-4FR,SA,SU;BYYEARDAY=34,65,76,78;BYMONTH=1,2,3,4,8
> # test failed: ""
> #          at: regression.c:2291
> #      got:
> FREQ=YEARLY;UNTIL=20000131T090000Z;BYDAY=-1TU,3WE,-4FR,SA,SU;BYYEARDAY=34,65,76,78;BYMONTH=1,2,3,4,8
> # expected:
> FREQ=YEARLY;UNTIL=20000131T090000Z;INTERVAL=1;BYDAY=-1TU,3WE,-4FR,SA,SU;BYYEARDAY=34,65,76,78;BYMONTH=1,2,3,4,8
> not ok 210 -
> FREQ=DAILY;COUNT=3;INTERVAL=1;BYDAY=-1TU,3WE,-4FR,SA,SU;BYYEARDAY=34,65,76,78;BYMONTH=1,2,3,4,8
> # test failed: ""
> #          at: regression.c:2296
> #      got:
> FREQ=DAILY;COUNT=3;BYDAY=-1TU,3WE,-4FR,SA,SU;BYYEARDAY=34,65,76,78;BYMONTH=1,2,3,4,8
> # expected:
> FREQ=DAILY;COUNT=3;INTERVAL=1;BYDAY=-1TU,3WE,-4FR,SA,SU;BYYEARDAY=34,65,76,78;BYMONTH=1,2,3,4,8
> 
> thats the Interval missing we're talking about, right?
> 
> So it seems as if our own test-suite says Benoit might be right?

It certainly isn't wrong to send the INTERVAL. But that the test suite
fails without it only shows that at some point, libical used to add the
default parameter. It doesn't mean that the parameter is required.

I don't mind sending the default parameter, although it makes the output
slightly larger than necessary. What I wanted to point out was that
although adding the parameter back will solve this particular problem,
it is not the best solution and may lead to conflicting requirements for
libical.

Did the Lightning developers contact Synchronica before asking for
changes in libical? At the very least this should pursued in parallel to
changing libical.

-- 
Bye, Patrick Ohly
--  
Patrick.Ohly at gmx.de
http://www.estamos.de/