Re: [Xmlunit-general] Test cases of XMLs for all the DifferenceConstants...

[Stefan B. <bo...@ap...>]

On Wed, 9 May 2007, Michael Schumacher <MSc...@ca...>
wrote:

> I'm using the DetailedDiff to compare XML's to find out what the
> differences are.  I was wondering if there is some XML's Test cases
> out there that would hit all the cases in the DifferenceConstants?

I don't think there is.  Some of them are tricky to create.  For
example the parser must be in validating mode in order to ever receive
a ATTR_VALUE_EXPLICITLY_SPECIFIED_ID difference.

Note that XMLUnit 1.1 adds a type of Difference and will - under
certain circumstances create another set of differences than 1.0 did.

> For example, if I have the following XML's, it will produce a
> difference of "presence of child nodes to be"
> 
> XML Document 1
> 
> <?xml version="1.0" encoding="UTF-8"?>
> <Root>
>    <OneElementDeep>
>       <BLAH/>
>    </OneElementDeep>
> </Root>
> 
>  
> 
> XML Document 2
> 
> <?xml version="1.0" encoding="UTF-8"?>
> <Root>
>    <OneElementDeep>
>       <BLAH>
>          <Presence>Child</Presence>
>       </BLAH>
>    </OneElementDeep>
> </Root>

What else would you expect?  I think in XMLUnit 1.1 you should get
HAS_CHILD_NODES_ID, CHILD_NODELIST_LENGTH_ID and
CHILD_NODE_NOT_FOUND_ID.

> This would help me handle all the cases there are produced from the
> DetailedDiff and handle them accordingly.  Right now I'm building up
> a java class that will generate a Delta between the two XML's.

Seems to be the easiest thing to do for your tests.

Stefan