[
https://issues.apache.org/jira/browse/GROOVY-7781?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15224196#comment-15224196
]
Jason Winnebeck commented on GROOVY-7781:
-----------------------------------------
Not sure if GH comments get copied to here, but I did check out the pull
request, and it looks like the changes I would expect and the unit test looks
like it would properly reproduce the bug as well as pass when the fix is good.
> GPathResult: Cannot retrieve XML attribute with namespace
> ---------------------------------------------------------
>
> Key: GROOVY-7781
> URL: https://issues.apache.org/jira/browse/GROOVY-7781
> Project: Groovy
> Issue Type: Bug
> Components: XML Processing
> Affects Versions: 2.4.5, 2.4.6
> Reporter: Jason Winnebeck
>
> When using GPathResult with XmlSlurper, if an attribute is in a namespace,
> you must retrieve it using the namespace prefix defined in the document and
> not that defined by declareNamespace. This issue prevents safe retrieval of
> namespaced XML attributes since the document generator is free to choose
> namespace prefix.
> {code}
> def text = """<x:root xmlns:x="blah">
> <x:child x:id='1'>c</x:child>
> </x:root>"""
> def xml =
> new XmlSlurper()
> .parseText(text)
> .declareNamespace(x:'blah')
> // .declareNamespace(t:'blah')
> println xml.child.text() //"c" always
> println xml.'x:child'.text() //"c" when declareNamespace x, else ""
> println xml.'t:child'.text() //"c" when declareNamespace t, else ""
> println xml.child.'@x:id' //"1" always
> println xml.child.'@t:id' //"" always
> {code}
> Looking in the debugger, there is namespaceMap and namespaceTagHints in
> GPathResult. namespaceMap is updated by declareNamespace but
> namespaceTagHints is not, and namespaceTagHints is used when looking up
> attributes. I don't see a way to update namespaceTagHints and namespaceMap
> doesn't really even seem to be used.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
