XML / XSL

Skip blocks of nodes with XSL

21-Dec-04 8:29

hi there,
so you want to represent graph in XML... nice homework, I guess Smile | :)

If you want to represent that matrix, what about:

<matrix>
<row>
<col connected="1" /><col/><col/><col/>
</row>
<row>
<col/><col/><col connected="1"/><col/>
</row>
<row>
<col/><col/><col/><col/>
</row>
<row>
<col/><col/><col connected="1"/><col/>
</row>
</matrix>
My first idea and indeed far from ideal structure for this task. However I gave you some ideas hopefully.

It's, of course, too much verbose representation of your graph.
What about
<graph>
<edge start="1" end="5" />
<edge start="4" end="2" />
.
.
.
</graph> ?

There you have some representations... Think about it (do not shoot like I do!) and choose the best for u Smile | :)

does it help?

best regards,
David 'DNH' Nohejl

Never forget: <i>"Stay kul and happy"</i> (I.A.)

Milby00720-Dec-04 0:22

Milby007

20-Dec-04 0:22

hi!!

i'm trying to process an xml file containing document tags & text, and skip certain blocks of the text based on condition tags. the problem i'm finding is that when i use apply-templates to move on and skip blocks of the document, it doesn't actually miss any of the text that i'm trying to avoid outputting to the resultant html document. i guess the question i have is, can you actually jump around an xml document using apply-templates?? i'm quite new to this stuff so please bear with me if i'm doing something wrong.

i've copied the code below just incase you want to see what i'm actually trying to do & run the transform for yourself. you'll need to create the following three files:

source.xml, "merge-data.xml" & test.xsl
(the file name in quotes has to have that exact name and be in the same directory as the stylesheet unless you want to change the code in the stylesheet).

source.xml:

<?xml version="1.0" encoding="UTF-8"?>
<br>
<!-- SOURCE DOCUMENT TO BE TRANSFORMED -->
<br>
<doc>
 <page>
  <content>
   <para>
    <run>
     <text>
      <condition eval="THIS = 'TEST'">&lt;&lt;IF THIS = 'TEST'&gt;&gt;</condition>
     </text>
    </run>
   </para>
   <para>
    <run>
     <text>IF EVALUATED TO TRUE!!!</text>
    </run>
   </para>
   <para>
    <run>
     <text>
      <condition eval="ELSE">&lt;&lt;ELSE&gt;&gt;&lt;&lt;IF THIS = 'TEST'&gt;&gt;</condition>
     </text>
    </run>
   </para>
   <para>
    <run>
     <text>IF EVALUATED TO FALSE!!!</text>
    </run>
   </para>
   <para>
    <run>
     <text>
      <condition eval="END">&lt;&lt;ENDIF&gt;&gt;&lt;&lt;IF THIS = 'TEST'&gt;&gt;</condition>
     </text>
    </run>
   </para>
  </content>
 </page>
</doc>

"merge-data.xml":

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

<!-- source document's condition information (merge-data.xml) -->

<dataset>
 <condition ref="&lt;&lt;IF WEATHER = 'BAD'&gt;&gt;">TRUE</condition>
 <condition ref="&lt;&lt;IF THIS = 'TEST'&gt;&gt;">TRUE</condition>
 <condition ref="&lt;&lt;IF THISDATA = 'RANDOM'&gt;&gt;">TRUE</condition>
</dataset>

test.xsl:

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

<!-- xslt stylesheet -->

<xsl:stylesheet version="1.0"
 xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
 xmlns:sdp="http://www.sectornet.co.uk/2004/DocPro">

 <xsl:output method="html"/>
 
 <xsl:variable name="mergeData" select="document('merge-data.xml')//dataset"/>
 
 <!-- ++++++++++++++++++++++++++++++++++ -->
 <!-- ++++++++++++++++++++++++++++++++++ -->
 <xsl:template match="/">
  <html>
   <head>
    <title>condition test output.</title>
   </head>
   <body>
    <center>
     <h1>Condition Test.</h1>
    </center>
    
    <xsl:apply-templates select="doc/page/content"/>
   </body>
  </html>
 </xsl:template>
 
 <!-- ++++++++++++++++++++++++++++++++++ -->
 <!-- ++++++++++++++++++++++++++++++++++ -->
 <xsl:template match="para">
  <p>
   <xsl:apply-templates/>
  </p>
 </xsl:template>
 
 <!-- ++++++++++++++++++++++++++++++++++ -->
 <!-- ++++++++++++++++++++++++++++++++++ -->
 <xsl:template match="text">
  <xsl:apply-templates/>
 </xsl:template>
 
 <!-- ++++++++++++++++++++++++++++++++++ -->
 <!-- ++++++++++++++++++++++++++++++++++ -->
 <xsl:template match="condition">
  <xsl:variable name="curNode" select="."/>
  
  <xsl:text> (CONDITION) </xsl:text>
  
  <xsl:choose>
   <xsl:when test="$curNode/@eval = 'ELSE'">
    <!-- process else statement -->
    <xsl:variable name="prevIfNode" select="$mergeData/condition[@ref = substring-after($curNode, '&lt;&lt;ELSE&gt;&gt;')]"/>
    
    <xsl:if test="$prevIfNode = 'TRUE'">
     <!-- go to end tag -->
     <xsl:apply-templates select="ancestor::*/following::*/condition[text() = concat('&lt;&lt;ENDIF&gt;&gt;', $prevIfNode/@eval)]"/>
    </xsl:if>
    
   </xsl:when>
   <xsl:when test="$curNode/@eval != 'END'">
    <!-- process if statement -->
    <xsl:variable name="curIfNode" select="$mergeData/condition[@ref = $curNode]"/>
    
    <xsl:if test="$curIfNode != 'TRUE'">
     <xsl:choose>
      <xsl:when test="ancestor::*/following::*/condition/text() = concat('&lt;&lt;ELSE&gt;&gt;', $curNode)">
       <!-- go to else tag -->
       <xsl:apply-templates select="ancestor::*/following::*/condition[text() = concat('&lt;&lt;ELSE&gt;&gt;', $curNode)]"/>
      </xsl:when>
      <xsl:otherwise>
       <!-- go to end tag -->
       <xsl:apply-templates select="ancestor::*/following::*/condition[text() = concat('&lt;&lt;ENDIF&gt;&gt;', $curNode)]"/>
      </xsl:otherwise>
     </xsl:choose>
    </xsl:if>
   </xsl:when>
  </xsl:choose>
 </xsl:template>
</xsl:stylesheet>

thanks!!!

Re: Skip blocks of nodes with XSL

Milby00720-Dec-04 3:13

Milby007

20-Dec-04 3:13

I've just realised that by changing the conditions return value from "TRUE" to "FALSE" (see merge-data.xml above), you can see it skipping the block. this is great, but after it's skipped the block it returns and continues processing from it's original position. e.g. it meets the first condition tag (if statement), then the template causes the jump to the else block because the if evaluates to false, but once that call has been completed it continues from the original condition tag (if statement). so basically the apply-templates is put on a call stack as it is in most programming languages' function calls. something i hadn't realised!! the problem is how to prevent it from continuing along nodes that i want to skip. i'm pretty sure it's down to the initial <xsl:apply-templates select="doc/page/content"/> found in the root template. it's obviously going to end up calling a template for all implemented tags under the content node. can anyone see a way to change this statement to be a bit more selective in which nodes it processes? or any other suggestions would be great!!

thanks!!

Re: Skip blocks of nodes with XSL

Milby00720-Dec-04 23:46

Milby007

20-Dec-04 23:46

Just incase anyone would find this helpful, i've got it working. it's not really how i wanted it to be written, so if anyone can see a better way of doing it then please say. i've also added another special document control called a dataname (basically a variable), so by adding text to the document enclosed in <dataname> tags with a reference to that and its value in the merge-data.xml file (e.g. <dataname ref="<<CFNAME>>">RICH</dataname> then you can effectively merge data into the document. the stylsheet code is copied below:

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

<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:sdp="http://www.sectornet.co.uk/2004/DocPro">
	<xsl:output method="html"/>
	<xsl:variable name="mergeData" select="document('merge-data.xml')//docpro-print-entry/dataset"/>
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<xsl:template match="/">
		<html>
			<head>
				<title>condition test output.</title>
			</head>
			<body>
				<center>
					<h1>Condition Test.</h1>
				</center>
				<xsl:apply-templates select="doc/page/content/para[1]"/>
			</body>
		</html>
	</xsl:template>
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<xsl:template match="para">
		<p>
			<xsl:apply-templates/>
		</p>
	</xsl:template>
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<xsl:template match="text">
		<xsl:choose>
			<xsl:when test="count(self::*/*) &gt; 0">
				<!-- we have controls in this block -->
				<xsl:apply-templates/>
			</xsl:when>
			<xsl:otherwise>
				<!-- output text and move on to next block -->
				<xsl:value-of select="."/>
				<xsl:apply-templates select="ancestor::*/following-sibling::*[1]"/>
			</xsl:otherwise>
		</xsl:choose>
	</xsl:template>
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<xsl:template match="dataname">
		<xsl:param name="sourceContext" select="$mergeData"/>
		<xsl:variable name="curNode" select="."/>
		
		<xsl:value-of select="$sourceContext//dataname[@ref = $curNode]"/>
		
		<!-- move on to next block -->
		<xsl:apply-templates select="ancestor::*/following-sibling::*[1]"/>
	</xsl:template>
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<!-- ++++++++++++++++++++++++++++++++++ -->
	<xsl:template match="condition">
		<xsl:variable name="curNode" select="."/>
		
		<xsl:choose>
			<xsl:when test="$curNode/@eval = 'ELSE'">
				<!-- process else statement -->
				<xsl:variable name="prevIfNode" select="$mergeData/condition[@ref = substring-after($curNode, '&lt;&lt;ELSE&gt;&gt;')]"/>
				
				<xsl:choose>
					<xsl:when test="$prevIfNode != 'TRUE'">
						<!-- continue through the block -->
						<xsl:apply-templates select="ancestor::*/following-sibling::*[1]"/>
					</xsl:when>
					<xsl:otherwise>
						<!-- go to end tag -->
						<xsl:apply-templates select="ancestor::*/following::*/condition[text() = concat('&lt;&lt;ENDIF&gt;&gt;', $prevIfNode/@ref)]"/>
					</xsl:otherwise>
				</xsl:choose>
			</xsl:when>
			<xsl:when test="$curNode/@eval != 'END'">
				<!-- process if statement -->
				<xsl:variable name="curIfNode" select="$mergeData/condition[@ref = $curNode]"/>
				
				<xsl:choose>
					<xsl:when test="$curIfNode = 'TRUE'">
						<!-- continue through the block -->
						<xsl:apply-templates select="ancestor::*/following-sibling::*[1]"/>
					</xsl:when>
					<xsl:otherwise>
						<xsl:choose>
							<xsl:when test="ancestor::*/following::*/condition/text() = concat('&lt;&lt;ELSE&gt;&gt;', $curNode)">
								<!-- go to else tag -->
								<xsl:apply-templates select="ancestor::*/following::*/condition[text() = concat('&lt;&lt;ELSE&gt;&gt;', $curNode)]"/>
							</xsl:when>
							<xsl:otherwise>
								<!-- go to end tag -->
								<xsl:apply-templates select="ancestor::*/following::*/condition[text() = concat('&lt;&lt;ENDIF&gt;&gt;', $curNode)]"/>
							</xsl:otherwise>
						</xsl:choose>
					</xsl:otherwise>
				</xsl:choose>
			</xsl:when>
			<xsl:when test="$curNode/@eval = 'END'">
				<!-- continue past the block -->
				<xsl:apply-templates select="ancestor::*/following-sibling::*[1]"/>
			</xsl:when>
		</xsl:choose>
	</xsl:template>
	
</xsl:stylesheet>

Basic XML

sharkbait16-Dec-04 3:55

sharkbait

16-Dec-04 3:55

I am an XML rookie and need some help. I am creating a project to poll web applications and other apps and of there is an error I want to send an email. I would like to poll the different apps listed in my XML. I want to pull back a application and its attributes. I have tried the XPathNavigator methods
with little success. Any help would be appreciated as to how to approach this problem or some good basic XML instructions. Here is the XML I created maybe I need to do it differently.

<applications xmlns="http://tempuri.org/Applications.xsd">
<application>
<name>Volunteer Email
<url>http://ftwweb02/volemail/report4.asp
<apptype>web
<databaseconnectionstring>

<database>

<emailaddresstonotify>BobSmith@texashealth.org

<fullpath>

<testtime>60000
<timeout>20000

<application>
<name>XBox App
<url>none
<apptype>database
<databaseconnectionstring>pwd=214;uid=me

<database>

<emailaddresstonotify>GeraldStanford@texashealth.org

<fullpath>

<testtime>60000
<timeout>20000

Xslt generated XML string format problem.

titabonbon15-Dec-04 19:27

15-Dec-04 19:27

Hi,all

I have few issues with Xslt transfer xml to xml

1: xml is from a string, how xslt transfer a dynamcally generated string instead of a file;

2: I transfer xml to xml string by xslt, but the XML
string can't be validated by schema. the thing is that if I load the generated XML string into VS studio and save it,which become a file. it can be validated. or it is valid if I just add a blank
space anywhere.
It might be the format issue with file and string, but I think it is weired.

DavidNohejl17-Dec-04 12:17

17-Dec-04 12:17

add 1) System.Xml.XsltTransform.Transform() method accepts TextWriter as an argument (see MSDN[^]). StringWriter is derived from TextWriter, so... Smile | :)

hi,

add 2) similarly, System.Xml.XmlValidatingReader constructor takes XmlReader as argument. Xmlreader can read from stream. and string can be written into stream (e.g System.IO.MemoryStream).

I only made some searches at MSDN and didn't try it. Let me know if there is any problem Smile | :)

Does it help?

best regards,
David 'DNH' Nohejl

Never forget: "Stay kul and happy" (I.A.)

titabonbon22-Dec-04 15:25

22-Dec-04 15:25

Hi,David:

1. yes, you are right for the first problem. I am using a StringReader to read the string and then use the XMLtextReader to read that stringReader,
2. NO. it wasn't that issue, the issue is that the decodeing problem, I need to change it to UTF-8, then done. thanks.

DavidNohejl23-Dec-04 0:06

23-Dec-04 0:06

hi,
I don't understand to your problem than. You have XML in string, right? and you want to validate it with XmlValidatingReader.. And there is problem with encoding? How you get that xml string? from file? in which encoding it is? See this MSDN[^]
which encodings are supported by XmlTextReader. If it's not your case see System.Text.Encoding etc.. sry I must go now I'll look at it later.

David

Never forget: "Stay kul and happy" (I.A.)

titabonbon23-Dec-04 12:58

23-Dec-04 12:58

when the xml string was generated, it was UTF-16, the XmlValidatingReader can't take it. ( For some reason, but might not this issue), I just changed it by UTF-8 using String.replace, and it is working now.

thank you very much.

DavidNohejl23-Dec-04 13:28

23-Dec-04 13:28

titabonbon wrote:
thank you very much

lol you bring up question, force me to look at it, than you post answer (so I learned something I didn't know before) and than you thank me! Wonderful Big Grin | :-D

marry christmas!

FYI: "All the encoding standards that the underlying operating system supports are supported." MSDN[^]

David

Never forget: "Stay kul and happy" (I.A.)

titabonbon26-Dec-04 12:14

26-Dec-04 12:14

oh, sorry.

because my project is on due, I have to do whatever I can, fortunately I solved it. but I want to thank you for your help sincerely. and I just gave you some reference maybe it can help sometime.

happy holiday.

DavidNohejl26-Dec-04 14:16

Does anybody know why XmlAttribute has ChildNodes property?

26-Dec-04 14:16

titabonbon wrote:
oh, sorry.

No I didn't mean it ironic! Blush | :O

happy holiday to you too

David

Never forget: "Stay kul and happy" (I.A.)

DavidNohejl15-Dec-04 11:46

Re: Does anybody know why XmlAttribute has ChildNodes property?

15-Dec-04 11:46

Hi,
I have one question about System.Xml.XmlAttribute class. Why it has ChildNodes property? Attributes cannot have child nodes, right? Is it bad design or did I miss something? Plz don't post answers like it's because XmlAttribute is derived from XmlNode. I know. I just don't know WHY? There is probably no case when it can be useful, since when I tried to append element to attribute (weird, I know Blush | :O

) it throws an exception

Unhandled Exception: System.InvalidOperationException: The specified node cannot<br />
 be inserted as the valid child of this node, because the specified node is the<br />
wrong type.<br />

What is your opinion/explanation? Or give me some links to people who will answer this (.NET designers maybe?)...

Thx a lot for some ideas or suggestions!

best regards,
David 'DNH' Nohejl

Never forget: "Stay kul and happy" (I.A.)

J4amieC20-Dec-04 3:09

J4amieC

20-Dec-04 3:09

Readup on Abstracting base functionality into a Base implementation, together with polymorphism and you may understand why EVERYTHING in the Xml tree derives from XmlNode.

Re: Does anybody know why XmlAttribute has ChildNodes property?

DavidNohejl20-Dec-04 3:46

20-Dec-04 3:46

hi,
thanks for your reply. I think I understand why everything derives from same base class... I am working on html parser these days and that made me think about it afer all. Bt I think it's either wrong to give XmlNode ability to have child nodes, or it's wrong to make XmlNode as base class. According to W3C recomendation, attribute is name-value pair. Well I don't understand to tokens, entities etc. (anyway I'll hopefully study it ) .. does it allow node* as value? I don't think so. I may be wrong and in case I am wrong, plz correct me.
IMO XmlNode has too much abilities. I'd make even more basic class as base for XmlNode, and derive XmlAttribute from this base class. (actually I do in my project) I am really at the beginning of true understanding to OOP, so plz tell me if it's ok so and why/why not. I simply can't see why atrribute could have child nodes. It's (slightly) bad design IMO.

thank you
David

*I think of value like something atomic, not with hierarchy - thats why i don't like child nodes there

Never forget: "Stay kul and happy" (I.A.)

Asp.net - Xml

Anonymous13-Dec-04 23:54

Anonymous

13-Dec-04 23:54

Hi all,
I want to trap all the error messages in my application to a Xml file format.So that it goes on appending the nodes whenever a new error occurs.How to accomplish this.Can anyone help me.....

Thanks in advance
Smile | :)

Re: Asp.net - Xml

DavidNohejl14-Dec-04 2:15