[
https://issues.apache.org/jira/browse/OGNL-173?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13434629#comment-13434629
]
Senen de Diego commented on OGNL-173:
-------------------------------------
I agree completely with Tony Stevenson.
If I'm comparing two objects of the same class (not Number) and o1.equals(o2)
returns false, I want OgnlOps.isEqual(o1,o2) to return false. And even if the
class is Comparable and o1.compareTo(o2) returns 0, OgnlOps.isEqual() should be
consistent with equal(), not with compareTo(), and also return false.
Of course, throwing an Exception, which is what happens with the current
implementation when the class is not Comparable, is the worst election.
I believe the call to compareWithConversion has to be done only after checking
that both objects are instances of Number, otherwise o1.equals(o2) should
suffice.
> OgnlOps.equal fails for two non-numeric objects that are not Comparable
> ------------------------------------------------------------------------
>
> Key: OGNL-173
> URL: https://issues.apache.org/jira/browse/OGNL-173
> Project: Commons OGNL
> Issue Type: Bug
> Reporter: David Mansfield
>
> OgnlOps.equal calls OgnlOps.isEqual, which, for two objects of some class X
> which are not equal according to X.equals() method, will then call
> OgnlOps.compareWithConversion. This method will throw an
> IllegalArgumentException on or around line 92 unless the class is a
> Comparable. This exception is wrong or is not caught properly in the isEqual
> method.
> How about wrapping the compareWithConversion in isEqual in a try/catch?
> I'm currently using ognl 3.0.1 but I checked latest code in SVN and it
> doesn't appear any different.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
