On 20 September 2011 12:20, Olivier Lefevre <lefevrol@xxxxxxxxx> wrote:
> Hi,
>
> I am using javax.xml.transform.Transformer to pretty-print DOM
> documents or fragments thereof, pulled with XPath. The relevant
> code is:
>
>  transformer.setOutputProperty(OutputKeys.INDENT, "yes");
>  transformer.setOutputProperty("{http://xml.apache.org/xslt}indent-amount",
> "2");
>
> It works fine for the root node, i.e., for a full document, but
> if I try that on a node retrieved with XPath the indentation is
> all messed up. For instance if I pull the location node from
> this document:
>
> <GeocodeResponse>
>  <status>OK</status>
>  <result>
>    <geometry>
>      <location>
>        <lat>51.3398300</lat>
>        <lng>12.3627600</lng>
>      </location>
>    </geometry>
>  </result>
> </GeocodeResponse>
>
> using "//location" and print it I get
>
> <location>
>        <lat>51.3398300</lat>
>        <lng>12.3627600</lng>
>      </location>
>
> Is this a known issue or am I doing something wrong? This is with
> Java 1.6.0_27; I am unsure which exact version of the XML
> libraries that implies.

That's correct... you will probably want to strip whitespace only text
nodes from the input (which is fine provided you don't have any 'mixed
content' elements).


--
Andrew Welch
http://andrewjwelch.com