332293 - SETPOS does not work correctly

Status: VERIFIED DUPLICATE of bug 322458

(Calendar :: Internal Components, defect)

Description

[Andrew N Dowden]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8.0.1) Gecko/20060111 Firefox/1.5.0.1
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9a1) Gecko/20060329 Mozilla Sunbird/0.3a1+

Boxing Day (Canada)
Observed next weekday after Christmas Day observed (Canada)
RRULE:FREQ=YEARLY;BYMONTH=12;BYMONTHDAY=26;BYDAY=TU,WE,TH,FR;BYSETPOS=1
RRULE:FREQ=YEARLY;BYMONTH=12;BYMONTHDAY=27,28;BYDAY=MO,TU;BYSETPOS=1

In second rule, multiplier effect (2 month-days, and 2 weekdays == 4 possible occurrences per year) should be negated by 'BYSETPOS=1'.  Only first match (per year) should be returned.




Reproducible: Always

Steps to Reproduce:
1. load iCal file, including syntax as above
2. move calendar to December 2010


Actual Results:  
matches: 2010-12-27, and 2010-12-28
 (ALL possible matches)

Expected Results:  
match: 2010-12-27
 (first occurence ONLY)

work around: (split rules)

Boxing Day (Canada) - with workaround
Observed next weekday after Christmas Day observed (Canada)
RRULE:FREQ=YEARLY;BYMONTH=12;BYMONTHDAY=26;BYDAY=TU,WE,TH,FR;BYSETPOS=1
RRULE:FREQ=YEARLY;BYMONTH=12;BYMONTHDAY=27,28;BYDAY=MO
RRULE:FREQ=YEARLY;BYMONTH=12;BYMONTHDAY=27;BYDAY=TU

Tested, and fixes problem (for now).  But would be needed for any imported iCal file using this feature.

Comment 1

[Joey Minta]

Duplicate of bug 322458?

Comment 2

[Andrew N Dowden]

This might explain bug 322458, but the (PayDay.ics) iCal syntax looks Ok.

In rule 1 (my example), the 'BYSETPOS=1' with TU,WE,TH,FR works fine.  Any other instance, where multiple weekdays occur, has also tested Ok (so far).

Comment 3

[Andrew N Dowden]

I loaded and tested PayDay.ics from bug 322458.  It doesn't work.

I then reviewed ALL occurences where I have tested / used BYSETPOS.  I have been mostly focusing on creating the iCal syntax, and dialogs (for same), rather that testing exhaustively.

Conclusion: It may not work at all. (Need to continue testing.)

Comment 4

[Andrew N Dowden]

further testing:

//  first instance

Test 1:  'first Mon or Tue ONLY, after DTSTART'
RRULE:FREQ=MONTHLY;BYDAY=MO,TU;BYSETPOS=1
DTSTART;VALUE=DATE:20060403
DTEND;VALUE=DATE:20060404

result:  'any Mon or Tue, after DTSTART'
(equivalent to)
RRULE:FREQ=MONTHLY;BYDAY=MO,TU
DTSTART;VALUE=DATE:20060403
DTEND;VALUE=DATE:20060404


Test 2:  'first Fri or Sat ONLY, in April ONLY, after DTSTART'
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=FR,SA;BYSETPOS=1
DTSTART;VALUE=DATE:20060407
DTEND;VALUE=DATE:20060408

result:  'any Fri or Sat, in April ONLY, after DTSTART'
(equivalent to)
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=FR,SA
DTSTART;VALUE=DATE:20060407
DTEND;VALUE=DATE:20060408


//  last instance

Test 3:  'last Mon or Tue ONLY, after DTSTART'
RRULE:FREQ=MONTHLY;BYDAY=MO,TU;BYSETPOS=-1
DTSTART;VALUE=DATE:20060424
DTEND;VALUE=DATE:20060425

result:  'any Mon or Tue, after DTSTART'
(equivalent to)
RRULE:FREQ=MONTHLY;BYDAY=MO,TU
DTSTART;VALUE=DATE:20060424
DTEND;VALUE=DATE:20060425


Test 4:  'last Fri or Sat ONLY, in April ONLY, after DTSTART'
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=FR,SA;BYSETPOS=-1
DTSTART;VALUE=DATE:20060428
DTEND;VALUE=DATE:20060429

result:  'any Fri or Sat, in April ONLY, after DTSTART'
(equivalent to)
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=FR,SA
DTSTART;VALUE=DATE:20060428
DTEND;VALUE=DATE:20060429


(Also tested on Lightning, no noted variance)

Comment 5

[Michiel van Leeuwen (email: mvl+moz@)]

This really is bug 322458. bysetpos is simply ignored.

*** This bug has been marked as a duplicate of 322458 ***

Comment 6

[Reed Loden [:reed]]

The bugspam monkeys have struck again. They are currently chewing on default assignees for Calendar. Be afraid for your sanity!