XML / XSL

XPath Q: How to select into a NodeList?

29-Jun-06 7:49

Instead of transforming with the StringWriter, wrap it in an XmlWriter:

XmlWriterSettings xws = new XmlWriterSettings();
xws.NewLineChars = "\n";  // Or whatever you want
XmlWriter xw = XmlWriter.Create(swTransformedXML, xws);

The XmlWriterSettings gives you more control over how the XML is written to the string.

Logifusion[^]

matthias s.29-Jun-06 5:24

Re: XPath Q: How to select into a NodeList?

29-Jun-06 5:24

Hi there,

I've got the following problem. I'd like to find a specific node using this XPath: /entry/comments/comment/[@author:"Jimmy" and @dateAdded:"29.06.2006 17:21:13"]. I use the following code to fill the XmlNodeList.

// sXPath contains the above mentioned path
XmlNodeList nodes = xml.SelectNodes(sXPath);

I receive an error (due to german framework here my attempt of a translation): System.Xml.XPath.XPathException : The expression passed into this method must result in a NodeSet.

Yes, this is what I expect. An XmlNodeSet. What else should I want to get?

Can anybody help me? Thanks in advance!

/matthias

I love deadlines. I like the whooshing sound they make as they fly by.
[Douglas Adams]

Dustin Metzgar29-Jun-06 5:50

Re: XPath Q: How to select into a NodeList?

29-Jun-06 5:50

matthias s. wrote:
/entry/comments/comment/[@author:"Jimmy" and @dateAdded:"29.06.2006 17:21:13"]

Do you mean /entry/comments/comment[@author:"Jimmy" and @dateAdded:"29.06.2006 17:21:13"] ?

Logifusion[^]

matthias s.29-Jun-06 6:20

where did the whitespace go? [modified]

29-Jun-06 6:20

yes, how did you know? Roll eyes | :rolleyes:

thanks!

/matthias

I love deadlines. I like the whooshing sound they make as they fly by.
[Douglas Adams]

Bluebamboo29-Jun-06 5:09

29-Jun-06 5:09

Hi all,

I'm having a problem with my XSL converting to HTML. I can view all content fine, except the white space is not preserved! I've tried all relevant attributes such as xml:space, xsl:preserve-space, bt don't work
Frown | :(

(

(

For example in my XML:

<catalog>
<content>
lines and lines of text
more text
</content>
</catalog>

When transforming I get:

lines and lines of text more text

It's all bunched up...where did the newline characters go?

my XSL contains this:

<xsl:value-of select="content"/>

Any ideas? Thanks.

-- modified at 11:55 Thursday 29th June, 2006

Dustin Metzgar29-Jun-06 5:48

29-Jun-06 5:48

In order for your message to show correctly you have to use < to create the < and > to create the > signs.

Anyways, this typically happens when doing an XSL transform. If you're coding the transformation yourself in .NET, you can put settings on the XmlTextWriter that will handle the spacing. If you're using a tool like XmlSpy, you can do a pretty print. There might be settings on the transformation software you're using to format the output.

Logifusion[^]

Bluebamboo29-Jun-06 6:01

29-Jun-06 6:01

thanks.

i think you might misunderstand my issue, the problem is about how to transform xml to html with all whitespace which the xml file has, not about the pretty print.;)

plz help me

Dustin Metzgar29-Jun-06 6:48

29-Jun-06 6:48

Well, I am doing the same thing as you. I pass some XML through a stylesheet to create HTML. The source is all cramped together, but I don't mind much as long as the HTML comes out ok. But sometimes it doesn't. For instance, if I want a space between two items like so:

<xsl:value-of select="$data/ch2_State"/> <xsl:value-of select="$data/vch10_ZipCode"/>

XSL will strip out the space between the state and the zip code so it looks like this: NY12345

That's a real pain. And, like you say, preserve space and whatever else you put there just doesn't work. So, what I ended up doing was this:

<xsl:value-of select="$data/ch2_State"/>
<xsl:text disable-output-escaping="yes">&</xsl:text>nbsp;
<xsl:value-of select="$data/vch10_ZipCode"/>

It's an extra annoyance, but it gets the job done.

Logifusion[^]

Bluebamboo29-Jun-06 9:59

29-Jun-06 9:59

it is also one of solutions, but what i'm aim to do is to write a xsl file, and i use this xsl to transform all articles which stored in xml file. So the positions of carriage return in each article are random.

do you have any solution to deal with it?

thanks?

Dustin Metzgar29-Jun-06 10:27

Re: where did the whitespace go? [modified]

29-Jun-06 10:27

In one situation where I needed the HTML to look exactly a certain way, I changed my output type from HTML to Text. This way, the stylesheet had to pay more attention to me. The stylesheet was harder to write because you have to enclose things in <xsl:text> tags, but it got the job done. To do a new line, you can do something like this:

<xsl:text>
</xsl:text>

<xsl:text>&#xD;&#xA;</xsl:text>

Even if the target is set to Text, XSL still likes to get rid of whitespace, so it's a lot of trial and error.

I hope that helps. If you find an easier solution, I'd be happy to hear it!

Logifusion[^]

Bluebamboo29-Jun-06 11:51

29-Jun-06 11:51

oh god!

i tried your way to do a new line with <xsl:text> tags:

<xsl:text>
</xsl:text>

it's just displayed a space instead of carriage return.

what's wrong?

-- modified at 17:51 Thursday 29th June, 2006

Dustin Metzgar29-Jun-06 14:03

29-Jun-06 14:03

Ya, unfortunately that way works with some XSL transformers and not with others. It's really quite annoying. You might want to use the other method instead. I'm not sure if you have to escape it or not, but this seemed to work pretty well for one guy at work. The only trouble is that it makes the XSL really ugly.

Logifusion[^]

Bluebamboo29-Jun-06 14:57

29-Jun-06 14:57

yes, i tried to enter <cr/> tag (my self-define tag )to substitute each CARRIAGE RETURN, when i am entering content of my article into xml file, and make relevant manipulation for this tag in xsl file.

however, i meet another problem Cry | :((

, how to find out all <cr/> tags in a xml file? because this tag can be existed in different hierarchy of xml file.

any ideas?

many thanks

Dustin Metzgar30-Jun-06 3:01

30-Jun-06 3:01

That sounds difficult. Is it that you have text in your XML document that has newlines in it and you want the formatting to be the same in the HTML? If you have the ability to, you might want to try putting it into a CDATA in the XML. After switching to text output, I was able to get my stylesheets working the way I wanted them to.

Logifusion[^]

Tuwing.Sabado3-Jul-06 19:59

Tuwing.Sabado

3-Jul-06 19:59

Define a template match for your <cr> node.

In the absence of a select attribute, the xsl:apply-templates instruction processes all of the children of the current node, including text nodes.

Tutorial for XSLT :
http://www.zvon.org/xxl/XSLTreference/Output/index.html[^]

Bluebamboo4-Jul-06 0:28

Longer text with special characters in xml files?

4-Jul-06 0:28

THanks everyone.

finally, I use <xsl:copy-of select="./node()"/>to address my problem.

Thanks for the Url you provided, it is very helpful. Smile | :)

matthias s.28-Jun-06 1:30

Re: Longer text with special characters in xml files?

28-Jun-06 1:30

Hi there,

I'm wondering whether there is a simple way to store longer text blocks containing any kind of character in an xml file. How can this be done? If this can't be done, where could I find information on what characters are allowed an not?

Thanks a ton in advance!

/matthias

I love deadlines. I like the whooshing sound they make as they fly by.
[Douglas Adams]

Igor Sukhov28-Jun-06 20:25

Igor Sukhov

28-Jun-06 20:25

You could create CDATA section in xml file to store text with any characters in that section.

Best regards,

-----------
Igor Sukhovhttp://sukhov.net

Re: Longer text with special characters in xml files?

matthias s.29-Jun-06 3:08

Of nodes or attributes [modified]

29-Jun-06 3:08

Exactly what I needed. Thanks!

Greetings from Berlin!

/matthias

I love deadlines. I like the whooshing sound they make as they fly by.
[Douglas Adams]

Malcolm Smart27-Jun-06 0:36

Malcolm Smart

27-Jun-06 0:36

Which is better / cleaner / more 'standard' or is it personal preference?

<DBLIST>
<Database Name="DB1" Owner="Jack SMITH" RequestedBy="" Description="Live Replication" Locked="Yes" Date="12/12/2005" Action="Requested by John SMITH"/>
</DBLIST>

or

<DBLIST>
<Database>Live Replication
<Name>DB1</Name>
<Owner>Jack SMITH</Owner>
<RequestedBy><RequestedBy>
<Locked>Yes</Locked>
<Date>12/12/2005</Date>
<Action>Requested by John SMITH"</Action>
</Database>
</DBLIST>

...and...how to format this post so I can use <pre> but keep the xml tags displayed!

Regards

Angel
*********************************************
The sooner you fall behind, the longer you have to catch up.

-- modified at 6:37 Tuesday 27th June, 2006 - typo

Igor Sukhov27-Jun-06 19:59

Igor Sukhov

27-Jun-06 19:59

What kind of operations expected to be performed on this xml ?

Best regards,

-----------
Igor Sukhovhttp://sukhov.net

Malcolm Smart27-Jun-06 22:02

Malcolm Smart

27-Jun-06 22:02

It is a very large file, thousands of entries, and xpath queries, reading every node, every attribute.

Regards

Angel
*********************************************
The sooner you fall behind, the longer you have to catch up.

Igor Sukhov28-Jun-06 14:11

Igor Sukhov

28-Jun-06 14:11

If the xml files are going to be huge - I would recommend running few tests. I've heard about speed difference processing elements and attributes - but tests are necessary to find out if this applies to your particular xml parser.

From the design prospective - I prefer to use attribute (rather that element) if I fee that the entity is not going to evolve in future to contain sub-entities.

If your example I would make all entities as attributes, except "Action" which I feel might require children in future (e.g. action parameters).

Best regards,

-----------
Igor Sukhovhttp://sukhov.net