469,323 Members | 1,567 Online
Bytes | Developer Community
New Post

Home Posts Topics Members FAQ

Post your question to a community of 469,323 developers. It's quick & easy.

Storing HTML in an XML file and rendering it in the browser...

So I am quite upset that after working for a few hours on getting an
XML file format and XSL file that formats the XML data appropriatly,
only to find that if you store HTML code in your XML file (even in a
CDATA block), after the XML file is rendered, the HTML that was stored
in the XML file is not rendered, essentially put into the page as if it
had <pre> tags around it.

I know that someone is going to yell at me and say that the XML file
should only contain the content of the page, and that the formatting
should be done by the XSL, but I have a particluar use case for this.
Essentially I store HTML in my database, and want it to be transmitted
to the browser and rendered on the fly. As brief as I can say it is
like this:

I plan on using Javascript to download small XML files from my server
and place their contents, formatted accordingly, onto a page. The trick
here is that my XML file is basically a serialized dataset.
Additionally, I store HTML in my database for some fields. I want to
use XSL because I don't really care about the data in the XML file, I
just want it displayed (which is a perfect application for XML/XSL). I
have read that I could do something cheesy like matching the
A|P|BR|FONT tags, but that seems REALLY cheesy. Here is a bare bones of
my XML file:

<Data>
<RowDefinition>
<Field name=foo displaytype"Text|HTML|Dollars|Link|Image" />
... etc ...
</RowDefinition>
<Rows>
<Row>
<Value name=foo><![CDATA[A Value or maybe some HTML
code]]></Value>
... etc ...
</Row>
... There may or may not be more than one Row tag ...
</Rows>
</Data>

(I just typed this so it may not really be well formed BTW)

Is there something besides CDATA that I could use in this case? I can't
seem to find much usefull information on such a topic.

I have tried encoding the chars first, IE &lt; and &gt; , both with and
without the CDATA declerations, and I do get something different. With
the CDATA declaration, the &gt; comes through as is, but without the
CDATA declaration, &gt; comes through as >. I was really hopin that
something like this would work, damn...

My other (and more dreaded option) is to use a javascript XMLDom and
parse the contents out manually and set certain parts of a
spans.InnerHTML property to the value from the DOM, but I really don't
want to have to write recursive javascript functions, and design HTML
that responds that well to this type of thing. I guess I would have had
a hard time getting the XSL to render the XML via javascript, so
writing some XMLDom javascript code may be my only option...

Any ideas on this, or maybe if you don't think that this is possible,
if someone could send me a good, browser concious, reference guide to
the various Javascript XML Doms out there, it would be greatly
appreciated...

Thanks for your time in advance!

Andy Baldwin

Nov 12 '05 #1
7 8828


An************@gmail.com wrote:
I plan on using Javascript to download small XML files from my server
and place their contents, formatted accordingly, onto a page. The trick
here is that my XML file is basically a serialized dataset.
Additionally, I store HTML in my database for some fields. I want to
use XSL because I don't really care about the data in the XML file, I
just want it displayed (which is a perfect application for XML/XSL).


While XSLT is a nice tool to transform XML it is not designed to process
HTML unless you can ensure that your HTML is XHTML or is transformed to
XHTML before the XSLT processor sees it.
So while you can choose to store some HTML tag soup in a CDATA section
of an XML document you have to live with the consequences, any XML
application (like XSLT) sees the stuff in the CDATA section as plain
text and not as structured markup.

--

Martin Honnen --- MVP XML
http://JavaScript.FAQTs.com/
Nov 12 '05 #2
> While XSLT is a nice tool to transform XML it is not designed to process
HTML unless you can ensure that your HTML is XHTML or is transformed to
XHTML before the XSLT processor sees it.
So while you can choose to store some HTML tag soup in a CDATA section
of an XML document you have to live with the consequences, any XML
application (like XSLT) sees the stuff in the CDATA section as plain
text and not as structured markup.


I love your phrase "HTML tag soup", classic...

I agree, I DON'T want it to process the underlying HTML, however I want
it to ultimitly be "rendered" in the browser with the bold, LI and
other tags rendered into the appropriate text decorations.

After thinking about it some more overnight, I realized that this may
be a moot point. Considering how I am applying the translation to my
XML file, I have not actually run the code through an XML Dom to do the
translation and save the resulting HTML to a file. Instead I
shortcutted it and placed the following tag at the top of the XML file,
so when it is opened in a browser the transformation is done inline,
and the resulting HTML is never actually saved and rendered.

( IE <?xml-stylesheet type='text/xsl' href='DataSet.xsl'?> )

I think that programatically getting the HTML that results from this
translation, then either saving it to an actual HTML file, or using
javascript to set a span's innerHTML property to the resulting
translated HTML code would actually work. I just hadn't gotten that
far, but this may be a "non-issue" after all (at least I hope :D)

Andy Baldwin

Nov 12 '05 #3
Hello!
<Data> .... <Rows>
<Row>
<Value name=foo><![CDATA[A Value or maybe some HTML
code]]></Value>
... etc ...
</Row>
... There may or may not be more than one Row tag ...
</Rows>
</Data>


Just use

<xslt:value-of
disable-output-escaping="yes"
select="/Data/Rows/Row[1]/Value[@name = 'foo']/text()"
/>

This will copy the plain text from the CDATA to the output, but I don't
know if it works with not well formed HTML.

But there should be no reason to create Tag Soups these days!
--
Pascal Schmitt
Nov 12 '05 #4
Pascal Schmitt wrote:
Hello! ....
....
....
Just use

<xslt:value-of
disable-output-escaping="yes"
select="/Data/Rows/Row[1]/Value[@name = 'foo']/text()"
/>

This will copy the plain text from the CDATA to the output, but I don't
know if it works with not well formed HTML.

But there should be no reason to create Tag Soups these days!

Cool, I will try it out, thanks for the tip!

Just curious, how else could I handle this situation? I mean, I am
basically trying to use AJAX to get data from the DB sent to the client
in an XML file. The data in the DB could contain raw HTML, so there
really isn't much other choice that I have, right? The only other way
is to parse the XML and then pulling out the data individually, I could
stick the values into appropriate span tags on the page. However, it
would seem to be much easier to just XSL it and use that result.

Any other suggestions would be greatly appreciated! :D

AB

Nov 12 '05 #5
Ohh yeah, and disable-output-escaping="yes" worked like a champ!

But yeah, it seems that I have to create "tag soup" I can't think of
any other way unfortunatly...

Thanks for the help!

AB

Nov 12 '05 #6


An************@gmail.com wrote:
and disable-output-escaping="yes" worked like a champ!


But be aware that disable-output-escaping is an optional feature that an
XSLT processor does not need to implement at all or might not support in
certain scenarios (like not serializing the result tree).
It can help you if you know you use a certain processor and output
format where it is supported but if you want to write XSLT stylesheets
that rely on disable-output-escaping and you want to throw them at
various processors in the wild you will face problems. For instance the
XSLT processor in Mozilla/Firefox does not support
disable-output-escaping thus if you want to use that feature in
client-side XSLT Mozilla is out.
--

Martin Honnen --- MVP XML
http://JavaScript.FAQTs.com/
Nov 12 '05 #7

Martin Honnen wrote:
But be aware that disable-output-escaping is an optional feature that an
XSLT processor does not need to implement at all or might not support in
certain scenarios (like not serializing the result tree).
It can help you if you know you use a certain processor and output
format where it is supported but if you want to write XSLT stylesheets
that rely on disable-output-escaping and you want to throw them at
various processors in the wild you will face problems. For instance the
XSLT processor in Mozilla/Firefox does not support
disable-output-escaping thus if you want to use that feature in
client-side XSLT Mozilla is out.

Yes, I figured that would be problematic, I have given up on the XSL
transform and just started researching how to use the various XMLDoms
that Firefox and IE have available, seems that if GMAIL uses them for
AJAX types of things, then it must be possible in both of those
browsers...

AB

Nov 12 '05 #8

This discussion thread is closed

Replies have been disabled for this discussion.

Similar topics

71 posts views Thread by tomy_baseo | last post: by
81 posts views Thread by sinister | last post: by
6 posts views Thread by bissatch | last post: by
3 posts views Thread by Clay Black | last post: by
10 posts views Thread by George | last post: by
32 posts views Thread by Next | last post: by
2 posts views Thread by Giedrius | last post: by
reply views Thread by suresh191 | last post: by
reply views Thread by harlem98 | last post: by
reply views Thread by listenups61195 | last post: by
By using this site, you agree to our Privacy Policy and Terms of Use.