Hi Eitan, perhaps there is a misunderstanding concerning CDATA sections?

In case you are not aware of it: 1) XQuery cannot see CDATA sections - their content is just part of the text content. Regard it as an evelope which does not influence the letter contained. As a consequence, it is principally not possible to *select* any text based on whether or not they are in CDATA sections. 2) XQuery output does not use CDATA sections, unless you supply the serialization parameter "cdata-section-elements", which contains a whitespace-separated list of element names. The contents of these elements will then by serialized as CDATA sections.

Example: basex -i input.xml -s "cdata-section-elements=package folder" myquery.xq

Kind regards, Hans-Juergen

Dirk Kirsten dk@basex.org schrieb am 9:43 Montag, 24.Februar 2014:

Hi Eitan,

works for me. What is your expected output and what do you get instead?

General advise for your query: Using / instead of // would be sufficient in this case and is in general is much more performant.

Cheers, Dirk

On 24/02/14 08:23, eitan elmalam wrote:

Hi,

I try to extract the Cdata by the follow query:

for $x in doc('input.xml')//App//package//text() return $x

On the follow xml

<?xml version="1.0" encoding="UTF-8"?>

<App name="BasexTest_1">   <package name="main/resources/xml">     <![CDATA[ -- dd --]]>     <!-- ff -->   <file name="input.xml"></file>   </package>   <!-- ff -->   <package name="org/basex/examples/local">   <file name="RunCommands.java"></file>     <!-- dd -->     <!-- ff -->   <folder>lib</folder></package> </App>

and  it is not return the cdata ,why ?

Thanks Eitan

---

BaseX-Talk mailing list BaseX-Talk@mailman.uni-konstanz.de https://mailman.uni-konstanz.de/mailman/listinfo/basex-talk