[jira] Commented: (RIVER-310) [patch] fix possible NPE due to incorrectly ordered if guard

["Peter Jones (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/RIVER-310?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12710457#action_12710457 ] 

Peter Jones commented on RIVER-310:
-----------------------------------

That patch looks fine to me, FWIW.  I think that the null check there is actually superfluous anyway, because this private method (mapPut) is only invoked by the private "read" method with values that cannot be null (either the result of a String.substring invocation, or the value of "name" itself when it has already been proven non-null).

> [patch] fix possible NPE due to incorrectly ordered if guard
> ------------------------------------------------------------
>
>                 Key: RIVER-310
>                 URL: https://issues.apache.org/jira/browse/RIVER-310
>             Project: River
>          Issue Type: Improvement
>         Environment: NA
>            Reporter: Dave Brosius
>            Priority: Trivial
>         Attachments: fix_possible_npe.diff
>
>
> Code improperly orders null length 0 guard, opening the possibility of a NPE
> patch fixes
> -	if ((name.length() == 0) || (name == null)) {
> +	if ((name == null) || (name.length() == 0)) {
>  	    throw new IOException("no name specified in preference" +
>  				  " expression");

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.