Re: Node list operations

["G. Ken Holman" <gkholman@xxxxxxxxxxxxxxxxxxxx>]

At 99/05/09 15:23 +0700, James Clark wrote:
>You can do it (albeit inefficiently) like this:

Thank you James ... I hadn't considered walking over the entire list each
time and only looking at a single entry of the list when doing so.  I think
your approach exemplifies an approach to a class of problems that people
will come across in their work.

I've taken the time to add some comments to your code to illustrate for
others how your solution works:

<xsl:stylesheet xmlns:xsl="http://www.w3.org/XSL/Transform/1.0";>

                    <!--show 'x' only when first in group of all 'x's-->
<xsl:template match="v">
  <xsl:text>X Y Z
</xsl:text>
  <xsl:for-each select="w">        <!--walk the items in sorted order-->
    <xsl:sort select="x"/>       <!--context node list becomes sorted-->
    <xsl:sort select="y"/>
    <xsl:choose>
      <xsl:when test="position()=1">
        <xsl:value-of select="x"/><!--first in list is first in group-->
      </xsl:when>
      <xsl:otherwise>              <!--get value for current position-->
        <xsl:variable name="pos" expr="position()"/>
        <xsl:variable name="x" expr="x"/>
        <xsl:for-each select="../w">       <!--re-walk the list again-->
                 <!--(causing us to go over the 'n' items in the list
                          a total of n times; hence the inefficiency)-->
          <xsl:sort select="x"/>     <!--must do it in the same order-->
          <xsl:sort select="y"/>
             <!--stop at the position in list before current position;
                 skipping processing for every other position in list-->
          <xsl:if test="position() = $pos - 1">
             <xsl:choose>         <!--examine position before current-->
               <xsl:when test="not(x=$x)">  <!--not same, so is first-->
                 <xsl:value-of select="$x"/>           <!--show value-->
               </xsl:when>
               <xsl:otherwise>  <!--is same, so is not first in group-->
                 <xsl:text> </xsl:text>
               </xsl:otherwise>
             </xsl:choose>
          </xsl:if>
        </xsl:for-each>            <!--finish the re-walk of the list-->
      </xsl:otherwise>
    </xsl:choose> 
    <xsl:text> </xsl:text>        <!--display other associated values-->
    <xsl:value-of select="y"/>
    <xsl:text> </xsl:text>
    <xsl:value-of select="z"/>
    <xsl:text>&#xA;</xsl:text>                   <!--go to a new line-->
  </xsl:for-each>
</xsl:template>

</xsl:stylesheet>

>"G. Ken Holman" wrote:
>> Produce:
>> 
>>  X   Y   Z
>>  1   1   a
>>      2   b
>>      3   c
>>  2   1   d
>>      2   e
>>  3   1   f
>> 
>> ..... suppressing redundant values for X.
>
>> Now, consider the problem where I want the same output from an unsorted
>> collection:
>> 
>> <?xml version="1.0"?>
>> <v>
>> <w><x>1</x><y>2</y><z>b</z></w>
>> <w><x>3</x><y>1</y><z>f</z></w>
>> <w><x>1</x><y>3</y><z>c</z></w>
>> <w><x>2</x><y>2</y><z>e</z></w>
>> <w><x>2</x><y>1</y><z>d</z></w>
>> <w><x>1</x><y>1</y><z>a</z></w>
>> </v>
>
>> So, my question is, can this problem be solved with the proposed working
>> draft in a single pass, and if not, is it important enough a problem that

>> it should be solved with version 1.0?

Thank you again, James, for your tutelage.

........... Ken 

--
G. Ken Holman                  mailto:gkholman@xxxxxxxxxxxxxxxxxxxx
Crane Softwrights Ltd.           http://www.CraneSoftwrights.com/s/
Box 266, Kars, Ontario CANADA K0A-2E0  +1(613)489-0999  (Fax:-0995)
Website: XSL/XML/DSSSL/SGML services outline,  XSL/DSSSL shareware,
         stylesheet resource library, conference training schedule,
         commercial stylesheet training materials, on-line XSL CBT.
Next instructor-led XSLT Training:                  WWW8:1999-05-11


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list