AW: [jdom-interest] Problems with Mixed Content and XMLOutputter
jhunter at servlets.com
Fri Jul 27 10:12:22 PDT 2007
There's a TRIM_FULL_WHITE format option. It eliminates all pure
whitespace text nodes. If that's the extent of your problem, it could
work for you.
There's also the NORMALIZE option which is close to HTML rules but not
exactly. You're wanting NORMALIZE but without the trim, do you think?
Grimm, Markus wrote:
> that's my approach:
> xml ( with mixed-content whitespaces) --> open xml-text in a kind of text-editor --> load the <text>-content in a XML-Editor --> Save from the XML-editor (maybe there's already something to do protecting the spaces?!) to the text-editor --> Save from there to xml.
> And if I unsterstand your proposal, you will create the xml-result via XSLT at this point?
> I'm not so familiar with XSLT, but this could be a possible solution.
> But a native JDOM solution would be nice, too. I think it's more simple (ok, is't always relative ;) ...)
> -----Ursprüngliche Nachricht-----
> Von: Michael Kay [mailto:mike at saxonica.com]
> Gesendet: Freitag, 27. Juli 2007 09:54
> An: Grimm, Markus; 'Jason Hunter'
> Cc: jdom-interest at jdom.org
> Betreff: RE: [jdom-interest] Problems with Mixed Content and XMLOutputter
> If you need to trim some whitespace and leave other whitespace intact, then
> you're asking a bit much of a general-purpose serializer to do this for you.
> Write an XSLT transformation that implements your specific rules.
> Michael Kay
>> -----Original Message-----
>> From: jdom-interest-bounces at jdom.org
>> [mailto:jdom-interest-bounces at jdom.org] On Behalf Of Grimm, Markus
>> Sent: 27 July 2007 08:21
>> To: Jason Hunter
>> Cc: jdom-interest at jdom.org
>> Subject: AW: [jdom-interest] Problems with Mixed Content and
>> Basically the reason is caused in my application.
>> I prepare a xml-document for a view in a JTextPane. The
>> native xml of the <text>-Tag is load to this editor too.
>> Then, the text-content can be edited in an external
>> XML-Editor (like XMLSpy or Epic) through an external call of
>> such tool. The text-content is very complicated, a little bit
>> more than the "Hello World" example. After editing in Epic
>> f.e., the text-content is resaved in the TextPane. From there
>> the whole content (text-data and meta-data) can be saved
>> (compact-format!) to a xml-document.
>> During the save-action the xml is validated against a schema.
>> So I can't use CDATA for the mixed-content-areas. It must be
>> valid too.
>> With raw-format I adopted lots of empty text-nodes. Not
>> really nice an inoperative in the saved xml. And with
>> raw-format or mixed-content by xml:space preserve-Attribs I
>> would have the same or similar problems.
>> The solution is compact-format with its advantages of
>> whitespace-handling an trimming. Only in this described point
>> I have that problem with this format.
>> -----Ursprüngliche Nachricht-----
>> Von: Jason Hunter [mailto:jhunter at servlets.com]
>> Gesendet: Freitag, 27. Juli 2007 03:37
>> An: Grimm, Markus
>> Cc: jdom-interest at jdom.org
>> Betreff: Re: [jdom-interest] Problems with Mixed Content and
>>> We handle with XML-Documents including elements with mixed-content.
>>> an easy example:
>>> <Content><P>Hello <B>World</B> how are
>>> I have to use Compact-Format to output the document as string.
>> Why do you have to use the compact format? Why not leave the
>> whitespace alone?
>> To control your jdom-interest membership:
More information about the jdom-interest