[patch] Return of SimpleTimeZone fix

Jerry Quinn jlquinn@optonline.net
Mon Jun 7 12:39:00 GMT 2004


Michael Koch writes:
 > -----BEGIN PGP SIGNED MESSAGE-----
 > Hash: SHA1
 > 
 > Am Montag, 7. Juni 2004 05:27 schrieb Jerry Quinn:
 > > I finally had time to track down my bug in this patch.  The patch
 > > introduces a new failure in
 > > gnu.testlet.java.util.GregorianCalendar.first, which is from a bug
 > > somewhere in GregorianCalendar feeding bad input to
 > > SimpleTimeZone.getOffset.
 > >
 > >
 > > 2004-06-06  Jerry Quinn  <jlquinn@optonline.net>
 > >
 > > 	* java/util/SimpleTimeZone.java (serialVersionOnStream): Bump to
 > > 2. (setStartRule,setEndRule): Don't take abs of day number.
 > > 	(getOffset): Clarify docs.  Add argument checks.
 > > 	(isBefore): Take abs of day number in DOW_LE_DOM_MODE.
 > > 	(equals,hasSameRules,toString,readObject): Use startTimeMode and
 > > 	endTimeMode.
 > 
 > Somehow your patch doesn't apply to CVS HEAD. Can you please update 
 > your tree and re-diff ?

I'm confused.  I just updated and I still get exactly the same diff
against HEAD.  I noticed the diff doesn't have the path to the file.
Is that the problem?

I could post the whole file if you want.

Jerry



More information about the Java-patches mailing list