Richard Kaye
2018-05-08 21:39:34 UTC
I owe you a Scotch next time you're in the area.
--
rk
-----Original Message-----
From: ProfoxTech <profoxtech-***@leafe.com> On Behalf Of Ted Roche
Sent: Tuesday, May 08, 2018 5:15 PM
To: ***@leafe.com
Subject: Re: Fun with date calculations in VFP
Did I mention I like date math? Brain-twisting stuff.
On Tue, May 8, 2018 at 5:14 PM, Ted Roche <***@gmail.com> wrote:
> How about:
>
> FUNCTION olddate(thedate, offset)
> * parameters:
> * the date: the source date
> * offset number of years offset, positive or negative
> * returns: the nearest date that is the same day of week as the source date
>
> LOCAL thediff, theoffset
> thediff = DOW(thedate) - DOW(GOMONTH(thedate, offset*12))
> theoffset = IIF(ABS(thediff)>3, 7-SIGN(thediff)*(thediff), thediff)
> RETURN GOMONTH(thedate, offset*12)+theoffset
>
> ABS() only picks up the high end of the date difference, 4,5,6
> SIGN() toggles whether you add or subtract the difference (minus a
> minus is a plus)
>
>
> On Tue, May 8, 2018 at 5:03 PM, Ted Roche <***@gmail.com> wrote:
>> An ICASE might do it.
>>
>> Basically the idea is get the day number (DOW) of the current date,
>> like 2 for Tuesday if SET FDOW is 1 ("the first day of the week is
>> Monday")
>>
>> Then create a new date with GOMONTH() and get the day number of that,
>> some number 1 through 7. So if the new date is a five, I want to go
>> back 3 to get to 2. Similarly, if it's a 6, I want to go forward 3 to
>> get to the next, closest 2.
>>
>> So, you want to find the nearest matching day number: they're either
>> the same day, so zero offset, or the matching day number are 1,2,3
>> days earlier or later.
>>
>> so you take the difference between the original date and the new date.
>> if the absolute difference is 3 or less, that's your offset and you
>> use that to move the new date to the matching day number.
>>
>> If the difference is -4, -5, -6 or 4, 5, 6 days, you need to look to
>> the next/previous weeks' matching day number, hence the magic 7
>> number.
>>
>>
>> On Tue, May 8, 2018 at 3:39 PM, Richard Kaye <***@invaluable.com> wrote:
>>> Or an ICASE? (I haven't quite wrapped my head around what you're doing there but it seems like there are 3 states?)
>>>
>>> m.theOffset=ICASE(theDiff>3, 7-m.theDiff,theDiff<-3, 7+m.theDiff, m.theDiff)
>>>
>>> --
>>>
>>> rk
>>>
>>> -----Original Message-----
>>> From: ProfoxTech <profoxtech-***@leafe.com> On Behalf Of Ted Roche
>>> Sent: Tuesday, May 08, 2018 3:25 PM
>>> To: ***@leafe.com
>>> Subject: Re: Fun with date calculations in VFP
>>>
>>> GOMONTH() just moves back to 2/28 on years that don't have a 2/29, so
>>> your olddate({^2016-02-29},-2) gives you the previous Tuesday,
>>> 2/24/2014.
>>>
>>> Hmmm. That's an OBO. It goes back 4 instead of forward 3. I think the
>>> nearest date ought to be 2014-03-03
>>>
>>> Ah, if thediff = -4, it remains -4. That's a bug. Add the second line
>>> below after the first:
>>>
>>> m.theOffset=IIF(theDiff>3, 7-m.theDiff, m.theDiff)
>>> m.theOffset=IIF(theDiff<-3, 7+m.theDiff, m.theDiff)
>>>
>>> I know there's a better algorithm, likely involving MOD() or CEILING()
>>> or FLOOR() but it's escaping me at the moment.
>>>
>>> There's 49 combinations of original date = (1,2,3,4,5,6,7) and
>>> resultant date = (1,2,3,4,5,6,7) and the offset should come out as
>>> {-3,-2,-1,0,1,2,3} , but I'm drawing a blank. Anyone? Beuller?
>>>
>>>
>>>
>>> On Tue, May 8, 2018 at 2:15 PM, Richard Kaye <***@invaluable.com> wrote:
>>>> Ted's solution works just fine. What I didn't test yet is using leap day itself as the starting point. Hmmm.....
>>>>
>>>> **************************************
>>>> * Program: datestuff3.prg
>>>> * Date: 05/08/2018 11:09 AM
>>>> * VFP Version: Visual FoxPro 09.00.0000.7423 for Windows
>>>> * Notes:
>>>> **************************************
>>>> ACTIVATE SCREEN
>>>> CLEAR
>>>> ? [Today: ]+CHR(9)+CHR(9)
>>>> ?? DATE()
>>>> ?? CHR(9)+CDOW(DATE())
>>>> ? [Same day last year: ]
>>>> ?? olddate(DATE(),-1)
>>>> ?? CHR(9)+CDOW(olddate(DATE(),-1))
>>>> ? [Same day 2 years ago: ]
>>>> ?? olddate(DATE(),-2)
>>>> ?? CHR(9)+CDOW(olddate(DATE(),-2))
>>>> ? [Same day 3 years ago: ]
>>>> ?? olddate(DATE(),-3)
>>>> ?? CHR(9)+CDOW(olddate(DATE(),-3))
>>>> ? [Same day 10 years ago: ]
>>>> ?? olddate(DATE(),-10)
>>>> ?? [ ]+CDOW(olddate(DATE(),-10))
>>>> ? [Yesterday: ]+CHR(9)
>>>> ?? DATE()-1
>>>> ?? CHR(9)+CDOW(DATE()-1)
>>>> ? [Same day last year: ]
>>>> ?? olddate(DATE()-1,-1)
>>>> ?? CHR(9)+CDOW(olddate(DATE()-1,-1))
>>>> ? [Same day 2 years ago: ]
>>>> ?? olddate(DATE()-1,-2)
>>>> ?? CHR(9)+CDOW(olddate(DATE()-1,-2))
>>>> ? [Same day 3 years ago: ]
>>>> ?? olddate(DATE()-1,-3)
>>>> ?? CHR(9)+CDOW(olddate(DATE()-1,-3))
>>>> ? [Same day 10 years ago: ]
>>>> ?? olddate(DATE()-1,-10)
>>>> ?? [ ]+CDOW(olddate(DATE()-1,-10))
>>>> ? [Tomorrow: ]+CHR(9)
>>>> ?? DATE()+1
>>>> ?? CHR(9)+CDOW(DATE()+1)
>>>> ? [Same day last year: ]
>>>> ?? olddate(DATE()+1,-1)
>>>> ?? CHR(9)+CDOW(olddate(DATE()+1,-1))
>>>> ? [Same day 2 years ago: ]
>>>> ?? olddate(DATE()+1,-2)
>>>> ?? CHR(9)+CDOW(olddate(DATE()+1,-2))
>>>> ? [Same day 3 years ago: ]
>>>> ?? olddate(DATE()+1,-3)
>>>> ?? CHR(9)+CDOW(olddate(DATE()+1,-3))
>>>> ? [Same day 10 years ago: ]
>>>> ?? olddate(DATE()+1,-10)
>>>> ?? [ ]+CDOW(olddate(DATE()+1,-10))
>>>>
>>>> FUNCTION oldDate(theDate AS Date, offset AS Integer)
>>>> * compliments of Ted Roche
>>>> * parameters: thedate: date to start from
>>>> * offset: number of years offset, positive or negative
>>>> * returns: nearest date to the offset that falls on the same day of the week
>>>> LOCAL m.theDiff, m.theOffset
>>>> m.theDiff=DOW(m.theDate)-DOW(GOMONTH(m.theDate, m.offset*12))
>>>> m.theOffset=IIF(theDiff>3, 7-m.theDiff, m.theDiff)
>>>> RETURN GOMONTH(m.theDate, m.offset*12) + m.theOffset
>>>>
>>>> ENDFUNC
>>>>
>>>> --
>>>>
>>>> rk
>>>>
>>>> -----Original Message-----
>>>> From: ProfoxTech <profoxtech-***@leafe.com> On Behalf Of Gene Wirchenko
>>>> Sent: Tuesday, May 08, 2018 12:46 PM
>>>> To: ***@leafe.com
>>>> Subject: RE: Fun with date calculations in VFP
>>>>
>>>> At 06:27 2018-05-08, Richard Kaye <***@invaluable.com> wrote:
>>>>>The latter. And that was the approach I was just working through.
>>>>
>>>> What about the week every few years that you will be disregarding?
>>>>
>>>> A year has 52 weeks plus one or two days.
>>>>
>>>> [snip]
>>>>
>>>> Sincerely,
>>>>
>>>> Gene Wirchenko
>>>>
>>>>
[excessive quoting removed by server]
_______________________________________________
Post Messages to: ***@leafe.com
Subscription Maintenance: http://mail.leafe.com/mailman/listinfo/profox
OT-free version of this list: http://mail.leafe.com/mailman/listinfo/profoxtech
Searchable Archive: http://leafe.com/archives/search/profox
This message: http://leafe.com/archives/byMID/profox/***@DM5PR10MB1244.namprd10.prod.outlook.com
** All postings, unless explicitly stated otherwise, are the opinions of the author, and do not constitute legal or medical advice. This statement is added to the messages for those lawyers who are too stupid to see the obvious.
--
rk
-----Original Message-----
From: ProfoxTech <profoxtech-***@leafe.com> On Behalf Of Ted Roche
Sent: Tuesday, May 08, 2018 5:15 PM
To: ***@leafe.com
Subject: Re: Fun with date calculations in VFP
Did I mention I like date math? Brain-twisting stuff.
On Tue, May 8, 2018 at 5:14 PM, Ted Roche <***@gmail.com> wrote:
> How about:
>
> FUNCTION olddate(thedate, offset)
> * parameters:
> * the date: the source date
> * offset number of years offset, positive or negative
> * returns: the nearest date that is the same day of week as the source date
>
> LOCAL thediff, theoffset
> thediff = DOW(thedate) - DOW(GOMONTH(thedate, offset*12))
> theoffset = IIF(ABS(thediff)>3, 7-SIGN(thediff)*(thediff), thediff)
> RETURN GOMONTH(thedate, offset*12)+theoffset
>
> ABS() only picks up the high end of the date difference, 4,5,6
> SIGN() toggles whether you add or subtract the difference (minus a
> minus is a plus)
>
>
> On Tue, May 8, 2018 at 5:03 PM, Ted Roche <***@gmail.com> wrote:
>> An ICASE might do it.
>>
>> Basically the idea is get the day number (DOW) of the current date,
>> like 2 for Tuesday if SET FDOW is 1 ("the first day of the week is
>> Monday")
>>
>> Then create a new date with GOMONTH() and get the day number of that,
>> some number 1 through 7. So if the new date is a five, I want to go
>> back 3 to get to 2. Similarly, if it's a 6, I want to go forward 3 to
>> get to the next, closest 2.
>>
>> So, you want to find the nearest matching day number: they're either
>> the same day, so zero offset, or the matching day number are 1,2,3
>> days earlier or later.
>>
>> so you take the difference between the original date and the new date.
>> if the absolute difference is 3 or less, that's your offset and you
>> use that to move the new date to the matching day number.
>>
>> If the difference is -4, -5, -6 or 4, 5, 6 days, you need to look to
>> the next/previous weeks' matching day number, hence the magic 7
>> number.
>>
>>
>> On Tue, May 8, 2018 at 3:39 PM, Richard Kaye <***@invaluable.com> wrote:
>>> Or an ICASE? (I haven't quite wrapped my head around what you're doing there but it seems like there are 3 states?)
>>>
>>> m.theOffset=ICASE(theDiff>3, 7-m.theDiff,theDiff<-3, 7+m.theDiff, m.theDiff)
>>>
>>> --
>>>
>>> rk
>>>
>>> -----Original Message-----
>>> From: ProfoxTech <profoxtech-***@leafe.com> On Behalf Of Ted Roche
>>> Sent: Tuesday, May 08, 2018 3:25 PM
>>> To: ***@leafe.com
>>> Subject: Re: Fun with date calculations in VFP
>>>
>>> GOMONTH() just moves back to 2/28 on years that don't have a 2/29, so
>>> your olddate({^2016-02-29},-2) gives you the previous Tuesday,
>>> 2/24/2014.
>>>
>>> Hmmm. That's an OBO. It goes back 4 instead of forward 3. I think the
>>> nearest date ought to be 2014-03-03
>>>
>>> Ah, if thediff = -4, it remains -4. That's a bug. Add the second line
>>> below after the first:
>>>
>>> m.theOffset=IIF(theDiff>3, 7-m.theDiff, m.theDiff)
>>> m.theOffset=IIF(theDiff<-3, 7+m.theDiff, m.theDiff)
>>>
>>> I know there's a better algorithm, likely involving MOD() or CEILING()
>>> or FLOOR() but it's escaping me at the moment.
>>>
>>> There's 49 combinations of original date = (1,2,3,4,5,6,7) and
>>> resultant date = (1,2,3,4,5,6,7) and the offset should come out as
>>> {-3,-2,-1,0,1,2,3} , but I'm drawing a blank. Anyone? Beuller?
>>>
>>>
>>>
>>> On Tue, May 8, 2018 at 2:15 PM, Richard Kaye <***@invaluable.com> wrote:
>>>> Ted's solution works just fine. What I didn't test yet is using leap day itself as the starting point. Hmmm.....
>>>>
>>>> **************************************
>>>> * Program: datestuff3.prg
>>>> * Date: 05/08/2018 11:09 AM
>>>> * VFP Version: Visual FoxPro 09.00.0000.7423 for Windows
>>>> * Notes:
>>>> **************************************
>>>> ACTIVATE SCREEN
>>>> CLEAR
>>>> ? [Today: ]+CHR(9)+CHR(9)
>>>> ?? DATE()
>>>> ?? CHR(9)+CDOW(DATE())
>>>> ? [Same day last year: ]
>>>> ?? olddate(DATE(),-1)
>>>> ?? CHR(9)+CDOW(olddate(DATE(),-1))
>>>> ? [Same day 2 years ago: ]
>>>> ?? olddate(DATE(),-2)
>>>> ?? CHR(9)+CDOW(olddate(DATE(),-2))
>>>> ? [Same day 3 years ago: ]
>>>> ?? olddate(DATE(),-3)
>>>> ?? CHR(9)+CDOW(olddate(DATE(),-3))
>>>> ? [Same day 10 years ago: ]
>>>> ?? olddate(DATE(),-10)
>>>> ?? [ ]+CDOW(olddate(DATE(),-10))
>>>> ? [Yesterday: ]+CHR(9)
>>>> ?? DATE()-1
>>>> ?? CHR(9)+CDOW(DATE()-1)
>>>> ? [Same day last year: ]
>>>> ?? olddate(DATE()-1,-1)
>>>> ?? CHR(9)+CDOW(olddate(DATE()-1,-1))
>>>> ? [Same day 2 years ago: ]
>>>> ?? olddate(DATE()-1,-2)
>>>> ?? CHR(9)+CDOW(olddate(DATE()-1,-2))
>>>> ? [Same day 3 years ago: ]
>>>> ?? olddate(DATE()-1,-3)
>>>> ?? CHR(9)+CDOW(olddate(DATE()-1,-3))
>>>> ? [Same day 10 years ago: ]
>>>> ?? olddate(DATE()-1,-10)
>>>> ?? [ ]+CDOW(olddate(DATE()-1,-10))
>>>> ? [Tomorrow: ]+CHR(9)
>>>> ?? DATE()+1
>>>> ?? CHR(9)+CDOW(DATE()+1)
>>>> ? [Same day last year: ]
>>>> ?? olddate(DATE()+1,-1)
>>>> ?? CHR(9)+CDOW(olddate(DATE()+1,-1))
>>>> ? [Same day 2 years ago: ]
>>>> ?? olddate(DATE()+1,-2)
>>>> ?? CHR(9)+CDOW(olddate(DATE()+1,-2))
>>>> ? [Same day 3 years ago: ]
>>>> ?? olddate(DATE()+1,-3)
>>>> ?? CHR(9)+CDOW(olddate(DATE()+1,-3))
>>>> ? [Same day 10 years ago: ]
>>>> ?? olddate(DATE()+1,-10)
>>>> ?? [ ]+CDOW(olddate(DATE()+1,-10))
>>>>
>>>> FUNCTION oldDate(theDate AS Date, offset AS Integer)
>>>> * compliments of Ted Roche
>>>> * parameters: thedate: date to start from
>>>> * offset: number of years offset, positive or negative
>>>> * returns: nearest date to the offset that falls on the same day of the week
>>>> LOCAL m.theDiff, m.theOffset
>>>> m.theDiff=DOW(m.theDate)-DOW(GOMONTH(m.theDate, m.offset*12))
>>>> m.theOffset=IIF(theDiff>3, 7-m.theDiff, m.theDiff)
>>>> RETURN GOMONTH(m.theDate, m.offset*12) + m.theOffset
>>>>
>>>> ENDFUNC
>>>>
>>>> --
>>>>
>>>> rk
>>>>
>>>> -----Original Message-----
>>>> From: ProfoxTech <profoxtech-***@leafe.com> On Behalf Of Gene Wirchenko
>>>> Sent: Tuesday, May 08, 2018 12:46 PM
>>>> To: ***@leafe.com
>>>> Subject: RE: Fun with date calculations in VFP
>>>>
>>>> At 06:27 2018-05-08, Richard Kaye <***@invaluable.com> wrote:
>>>>>The latter. And that was the approach I was just working through.
>>>>
>>>> What about the week every few years that you will be disregarding?
>>>>
>>>> A year has 52 weeks plus one or two days.
>>>>
>>>> [snip]
>>>>
>>>> Sincerely,
>>>>
>>>> Gene Wirchenko
>>>>
>>>>
[excessive quoting removed by server]
_______________________________________________
Post Messages to: ***@leafe.com
Subscription Maintenance: http://mail.leafe.com/mailman/listinfo/profox
OT-free version of this list: http://mail.leafe.com/mailman/listinfo/profoxtech
Searchable Archive: http://leafe.com/archives/search/profox
This message: http://leafe.com/archives/byMID/profox/***@DM5PR10MB1244.namprd10.prod.outlook.com
** All postings, unless explicitly stated otherwise, are the opinions of the author, and do not constitute legal or medical advice. This statement is added to the messages for those lawyers who are too stupid to see the obvious.