Re: DateCalendar magic numbers

2013-05-08 Thread Justin Mclean
Hi, > As a non-native English speaker, I'd have used 'shiftDate()' Lease out the "f" and you have it right. :-) There's another bug in that code as it doesn't respect daylight saving boundaries. Your suggested changes worked fine btw, I've replaced the method and both the DataChooser and DateFi

Re: DateCalendar magic numbers

2013-05-08 Thread Erik de Bruin
As a non-native English speaker, I'd have used 'shiftDate()' or something along those lines, but that'd probably be subject to comments as well ;-) EdB On Wed, May 8, 2013 at 4:55 PM, Justin Mclean wrote: > Hi, > >> Yes, the 'incrementDate' method works as expected, AFAICT. > It may do but the

Re: DateCalendar magic numbers

2013-05-08 Thread Justin Mclean
Hi, > Yes, the 'incrementDate' method works as expected, AFAICT. It may do but there is still an issue - see this JIRA: https://issues.apache.org/jira/browse/FLEX-13423 > (+1) adds a day, adding -1 as the second argument subtracts a day. Sure but my point was that something called increment shoul

Re: DateCalendar magic numbers

2013-05-08 Thread Erik de Bruin
Yes, the 'incrementDate' method works as expected, AFAICT. The default (+1) adds a day, adding -1 as the second argument subtracts a day. What are the inputs you are testing with? EdB On Wed, May 8, 2013 at 3:03 PM, Erik de Bruin wrote: > As far as I can tell, the -1 increment causes 8640

Re: DateCalendar magic numbers

2013-05-08 Thread Erik de Bruin
As far as I can tell, the -1 increment causes 8640 milliseconds to be subtracted (?) from the time of the date you enter (1 day)... I'm gonna test this now. EdB On Wed, May 8, 2013 at 2:57 PM, Justin Mclean wrote: > Hi, > >> I think it's correct... > Including the "increment" by -1? There

Re: DateCalendar magic numbers

2013-05-08 Thread Justin Mclean
Hi, > I think it's correct... Including the "increment" by -1? There is an off by one error in there somewhere. It's reproducible. Thanks, Justin

Re: DateCalendar magic numbers

2013-05-08 Thread Erik de Bruin
Justin, I think it's correct... it sets the new range end date for the first half of the now split range. As both the start and the end date fall within the range, it needs to be split in two: the first one from the range start to the day before the start date (which is the one we're looking at) a

Re: DateCalendar magic numbers

2013-05-08 Thread Justin Mclean
Hi, > Just because I can't let something like this go... It was particularly special bit of code wasn't it :-) I forgot to mention one of my fav lines: _selectedRanges[n].rangeEnd = incrementDate(startDate,-1); (Which I'm fairly certain is incorrect) > I wasn't able to test this, so mil

Re: DateCalendar magic numbers

2013-05-08 Thread Erik de Bruin
Just because I can't let something like this go... I wasn't able to test this, so mileage may vary ;-) function removeRangeFromSelection(startDate:Date, endDate:Date):void { var n:int, rangeEnd:Date, rangeStart:Date; if (!endDate || endDate < startDate) return; for (n = 0; n