[MacRuby] #537: Missing Time#strftime formatter: "%P"

MacRuby ruby-noreply at macosforge.org
Tue Dec 7 23:30:36 PST 2010


#537: Missing Time#strftime formatter: "%P"
-------------------------------+--------------------------------------------
 Reporter:  bburky@…           |       Owner:  lsansonetti@…        
     Type:  defect             |      Status:  new                  
 Priority:  minor              |   Milestone:  MacRuby Later        
Component:  MacRuby            |    Keywords:                       
-------------------------------+--------------------------------------------

Comment(by danielcavanagh@…):

 Okay, here it is. It also implements subsecs and Time as Rational in what
 is probably a very ugly way

 It passes all the rspec time specs except for the following 5:

 1) Time#+ adds a negative Float FAILED - Expected 700000 to equal 699999
 2) Time#strftime returns the date alone with %x FAILED - Expected
 "18/09/09" to equal "09/18/09"
 3) Time#strftime returns the time alone with %X FAILED - Expected "12:00
 PM" to equal "12:00:06"
 4) Time#to_r returns the rational number of seconds + usecs since the
 epoch FAILED - Expected (1209462791653111237151/1099511627776000000000) to
 equal (619244948763443/562949953421312)
 5) Time#usec returns the microseconds for time FAILED - Expected 899999 to
 equal 900000

 2 & 3 are invalid tests as far as I'm concerned, because %x and %X are
 defined as "preferred representation", which obviously differs depending
 on your region and system
 1 & 5 are off-by-ones involving floats. I don't know anything about floats
 but I'm guessing this is just a difference between our float
 implementations (perhaps ultimately because of hardware?), and considering
 floats are not designed to represent whole numbers this might be
 considered allowable
 4 is because of a difference between our Rational implementations. I'm not
 even going to go there...

 spec:ci aborts pretty early on with and without the patch, but it does
 abort a little bit earlier than normal with the patch so I guess that
 shows a regression. I have no idea how to get it to tell me what test it
 is running when it aborts though and don't feel like hacking up rspec at
 the moment just to work it out, so I'll leave that to someone else

 Hopefully you don't need to hack it into shape too much more :p Feel free
 to ask questions

-- 
Ticket URL: <http://www.macruby.org/trac/ticket/537#comment:9>
MacRuby <http://macruby.org/>



More information about the macruby-tickets mailing list