[
https://issues.apache.org/jira/browse/AXIOM-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15299941#comment-15299941
]
Boris Dushanov commented on AXIOM-483:
--------------------------------------
Sure.
I meant that there might be many 1.2.x users that rely on the fact that passing
an empty namespace e.g.
OMElementImpl.getChildrenWithName(new QName("","someLocalName"))
would mean that this will match children with "someLocalName" located in any
namespace.
In other words, passing an empty namespace is interpreted as a wildcard i.e.
any namespace.That's not the case in 1.3.x.
> OMElementImpl.getChildrenWithName() is too restrictive and not compatible
> with 1.2.x
> ------------------------------------------------------------------------------------
>
> Key: AXIOM-483
> URL: https://issues.apache.org/jira/browse/AXIOM-483
> Project: Axiom
> Issue Type: Bug
> Components: LLOM
> Affects Versions: 1.3.0
> Reporter: Boris Dushanov
> Priority: Critical
> Fix For: 1.3.0
>
>
> In 1.2.x org.apache.axiom.om.impl.llom.OMElementImpl.getChildrenWithName()
> contains the following comment:
> // The getChidrenWithName method used to tolerate an empty namespace
> // and interpret that as getting any element that matched the local
> // name. There are custmers of axiom that have hard-coded
> dependencies
> // on this semantic.
> // The following code falls back to this legacy behavior only if
> // (a) elementQName has no namespace, (b) the new iterator finds no
> elements
> // and (c) there are children.
> This seems to be not the case in 1.3.0 and the method is now restrictive.The
> current behavior is correct but the problem is that it's not backward
> compatible and will lead to many issues.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
