Hello David,
Curiosity let me solve the riddle.
I have not used the extra node method by Ralf, but instead, I looked into
the API of the DocumentBuilderFactory. Precisely, googled for
"DocumentBuilder" and "Namespace".
The trick is to make the DocumentFactory namespace-aware and in addition,
add a namespace to your text snippet.
In detail this is being changed:
OLD:
String fragment = "<text:p text:style-name=\"P13\"><text:span
text:style-name=\"T1\">Test</text:span></text:p>";
Node node =
DocumentBuilderFactory.newInstance().newDocumentBuilder().parse(new
InputSource(new StringReader(fragment ))).getDocumentElement();
NEW:
String fragment = "<text:p
xmlns:text=\"urn:oasis:names:tc:opendocument:xmlns:text:1.0\"
text:style-name=\"P13\"><text:span
text:style-name=\"T1\">Test</text:span></text:p>";
DocumentBuilderFactory dbf =
DocumentBuilderFactory.newInstance();
dbf.setNamespaceAware(true);
And then the test produces an ODF document with the desired output.
Cheers,
Svante
PS: Ralf we should continue to discuss on the dev list as Dave suggested, I
might be offline until Monday, so don't worry, if my response is late. ;-)
ᐧ
2018-05-10 22:34 GMT+02:00 Svante Schubert <[email protected]>:
> Hello David,
>
> to reproduce your problem, I have added a test case in one of the existing
> tests of the Simple API.
>
> @Test
> public void testXMLasString() {
> try {
> TextDocument resDoc = (TextDocument) TextDocument.loadDocument(
> ResourceUtilities.getTestResourceAsStream("section.odt"));
> Section section = resDoc.getSectionByName( "Section1" ); //
> this section does exist in the document
>
> // create new node form String
> String fragment = "<text:p text:style-name=\"P13\"><text:span
> text:style-name=\"T1\">Test</text:span></text:p>";
> Node node =
> DocumentBuilderFactory.newInstance().newDocumentBuilder().parse(new
> InputSource(new StringReader(fragment ))).getDocumentElement();
> node = section.getOdfElement().getOwnerDocument().importNode(
> node, true );
> // append new node into section
> section.getOdfElement().appendChild( node );
> resDoc.save(ResourceUtilities.newTestOutputFile("
> sectionWithText.odt"));
> } catch (Exception e) {
> Logger.getLogger(ListItemTest.class.getName()).log(Level.SEVERE, null, e);
> Assert.fail(e.getMessage());
> }
> }
>
> It is not REALLY a test as it always succeeds, but by this, I am able to
> debug in the IDE and first of all get a document from your procedure.
> You might want to do the same.
>
> An OpenDocument File is a ZIP with XML files within. You can unzip it and
> take a look into the content.xml
> I usually use the JEdit editor with the "archive" and "XML" plugin to open
> the content.xml file without unpacking and make an indent for the XML.
>
> If you do so it reveals the following:
> <text:section text:name="Section1" text:style-name="Sect1">
> <text:p text:style-name="Standard"/>
> <p style-name="P13"><span style-name="T1">Test</span></p></text:section>
>
> The string is inserted, but the prefix, the namespace was not found. You
> might get the DOM element and insert the elements one by one with DOM
> functions or search for a solution.
> Uncertain how it works by String insertion. If you find a way, please come
> back and tell :)
>
> Good luck!
> Svante
>
> 2018-05-10 12:08 GMT+02:00 David Michal <[email protected]>:
>
>> Hi,
>>
>> I have a code:
>>
>> // -----------------------------------------------------------------
>>
>> TextDocument resDoc = TextDocument.loadDocument( someInputStream );
>>
>> Section section = resDoc.getSectionByName( "Section1" ); // this section
>> does exist in the document
>>
>> // create new node form String
>>
>> String fragment = "<text:p text:style-name=\"P13\"><text:span
>> text:style-name=\"T1\">Test</text:span></text:p>";
>>
>> Node node = DocumentBuilderFactory.newInstance().newDocumentBuilder().parse(
>> new InputSource( new StringReader( fragment ) ) ).getDocumentElement();
>> node = section.getOdfElement().getOwnerDocument().importNode( node, true
>> );
>>
>> // append new node into section
>>
>> section.getOdfElement().appendChild( node );
>>
>> // -----------------------------------------------------------------
>>
>> The code runs without a problem. But nothing does appear in the section
>> in the result document. Please any idea how can I add new nodes created
>> from string into the odf document?
>>
>> Thanks,
>>
>> David
>>
>>
> ᐧ
>
