Re: What is the time anyways?

[gersh@aplpy.jhuapl.edu (John R. Gersh)]

In article <92Dec1.110128est.144049@explorer.dgp.toronto.edu> 
Russell Owen writes:
>Hi,
>    This may be a novice question but I am really stumped.  I am trying to
>write a function that returns the current time read from the mac's
>real-time clock.  It would seem to be a simple task.
>

	[Provides code using Mac time traps that exhibits strange
	behavior]


This problem raises some interesting questions:

The simple answer to the basic need, as has been mentioned, is to use Lisp
time functions rather than access the clock directly:

	(defun get-current-hms ()
	   (multiple-value-bind (sec min hr) (get-decoded-time)
	      (format nil "~2,'0D:~2,'0D:~2,'0D" hr min sec)))

will provide the desired functionality of the initial code.

But here's the weird part: the original code works fine if the
definition is evaluated, rather than compiled!

Stripping things to essentials:

	(defun time-test ()
	  (rlet ((time datetimerec))
	     (#_gettime time)
	     (rref time datetimerec.minute)))

works fine if the defun is evaluated with *compile-definitions* nil,
but does does not work correctly if *compile-definitions* is true.
This behavior shows up whether an rlet is used or the datetime record is
explicitly created and disposed in the function.

What's happening?




---------------------------------------------------------------------
John Gersh                                     gersh@aplpy.jhuapl.edu
The Johns Hopkins University Applied Physics Laboratory
Johns Hopkins Rd., Laurel, MD 20723		       (301) 953-5503