This can be done even with XSLT 1.0:
http://fxsl.sourceforge.net/articles/Random/Casting%20the%20Dice%20with%20FXS
L-htm.htm
More specifically, use the FXSL 1 stylesheet module randomList.xsl
Here is an existing test (demo):
When this transformation (testRandomList.xsl):
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:ext="http://exslt.org/common"
xmlns:f="http://fxsl.sf.net/"
xmlns:mySquare="f:mySquare"
xmlns:myDouble="f:myDouble"
exclude-result-prefixes="xsl f ext mySquare myDouble"
>
<xsl:import href="randomList.xsl"/>
<xsl:output omit-xml-declaration="yes" indent="yes"/>
<!-- This transformation must be applied to:
numList.xml
-->
<mySquare:mySquare/>
<myDouble:myDouble/>
<xsl:template match="/">
<xsl:variable name="vrtfRands">
<xsl:call-template name="randomSequence">
<xsl:with-param name="pLength" select="100"/>
</xsl:call-template>
</xsl:variable>
Random Recursive Index (dScale (randomSequence 100)):
<xsl:call-template name="_dScale">
<xsl:with-param name="pRandSeq"
select="ext:node-set($vrtfRands)/*"/>
</xsl:call-template>
Random Recursive Index 10:
<xsl:variable name="vrtfRecIndex">
<xsl:call-template name="_randomRecursiveIndex">
<xsl:with-param name="pList"
select="/*/*"/>
</xsl:call-template>
</xsl:variable>
<xsl:variable name="vRecIndex"
select="ext:node-set($vrtfRecIndex)/*"/>
<xsl:for-each select="$vRecIndex">
<xsl:copy-of select="."/>

</xsl:for-each>
Randomized 10-elements list:
<xsl:call-template name="_permutationFromRecursiveIndex">
<xsl:with-param name="pList" select="/*/*"/>
<xsl:with-param name="pRecIndex" select="$vRecIndex"/>
</xsl:call-template>
RandomizeList:
<xsl:call-template name="randomizeList">
<xsl:with-param name="pList" select="/*/*"/>
</xsl:call-template>
<xsl:variable name="vFunSquare"
select="document('')/*/mySquare:*[1]"/>
_mapFromRandIndex (^2) [1..10] seed:
<xsl:call-template name="_mapFromRandIndex">
<xsl:with-param name="pFun" select="$vFunSquare"/>
<xsl:with-param name="pList" select="/*/*"/>
<xsl:with-param name="pRecIndex" select="$vRecIndex"/>
</xsl:call-template>
<xsl:variable name="vFunDouble"
select="document('')/*/myDouble:*[1]"/>
randomMap (*2) [1..10] seed:
<xsl:call-template name="randomMap">
<xsl:with-param name="pFun" select="$vFunDouble"/>
<xsl:with-param name="pList" select="/*/*"/>
</xsl:call-template>
randListIndex [1..10] seed:
<xsl:call-template name="randListIndex">
<xsl:with-param name="pList" select="/*/*"/>
</xsl:call-template>
</xsl:template>
<xsl:template match="mySquare:*" mode="f:FXSL">
<xsl:param name="arg1"/>
<xsl:value-of select="$arg1 * $arg1"/>
</xsl:template>
<xsl:template match="myDouble:*" mode="f:FXSL">
<xsl:param name="arg1"/>
<xsl:value-of select="$arg1 + $arg1"/>
</xsl:template>
</xsl:stylesheet>
is applied on this source XML document:
<nums>
<num>01</num>
<num>02</num>
<num>03</num>
<num>04</num>
<num>05</num>
<num>06</num>
<num>07</num>
<num>08</num>
<num>09</num>
<num>10</num>
</nums>
the result is:
Random Recursive Index (dScale (randomSequence 100)):
<el>27</el>
<el>90</el>
<el>14</el>
<el>78</el>
<el>65</el>
<el>13</el>
<el>27</el>
<el>85</el>
<el>75</el>
<el>33</el>
<el>31</el>
<el>26</el>
<el>9</el>
<el>40</el>
<el>31</el>
<el>80</el>
<el>19</el>
<el>44</el>
<el>52</el>
<el>7</el>
<el>8</el>
<el>73</el>
<el>55</el>
<el>16</el>
<el>68</el>
<el>20</el>
<el>29</el>
<el>4</el>
<el>3</el>
<el>30</el>
<el>51</el>
<el>41</el>
<el>14</el>
<el>32</el>
<el>66</el>
<el>4</el>
<el>19</el>
<el>51</el>
<el>48</el>
<el>59</el>
<el>30</el>
<el>1</el>
<el>49</el>
<el>57</el>
<el>14</el>
<el>53</el>
<el>13</el>
<el>10</el>
<el>10</el>
<el>38</el>
<el>13</el>
<el>37</el>
<el>13</el>
<el>36</el>
<el>22</el>
<el>7</el>
<el>28</el>
<el>25</el>
<el>28</el>
<el>7</el>
<el>29</el>
<el>3</el>
<el>34</el>
<el>28</el>
<el>7</el>
<el>13</el>
<el>14</el>
<el>5</el>
<el>32</el>
<el>25</el>
<el>25</el>
<el>24</el>
<el>8</el>
<el>26</el>
<el>23</el>
<el>14</el>
<el>11</el>
<el>18</el>
<el>15</el>
<el>6</el>
<el>5</el>
<el>6</el>
<el>9</el>
<el>4</el>
<el>8</el>
<el>14</el>
<el>12</el>
<el>12</el>
<el>5</el>
<el>2</el>
<el>5</el>
<el>1</el>
<el>4</el>
<el>1</el>
<el>4</el>
<el>4</el>
<el>1</el>
<el>2</el>
<el>1</el>
<el>1</el>
Random Recursive Index 10:
<el>3</el>
<el>9</el>
<el>2</el>
<el>6</el>
<el>5</el>
<el>1</el>
<el>2</el>
<el>3</el>
<el>2</el>
<el>1</el>
Randomized 10-elements list:
<el>03</el>
<el>10</el>
<el>02</el>
<el>08</el>
<el>07</el>
<el>01</el>
<el>05</el>
<el>09</el>
<el>06</el>
<el>04</el>
RandomizeList:
<el>03</el>
<el>10</el>
<el>02</el>
<el>08</el>
<el>07</el>
<el>01</el>
<el>05</el>
<el>09</el>
<el>06</el>
<el>04</el>
_mapFromRandIndex (^2) [1..10] seed:
<el>9</el>
<el>100</el>
<el>4</el>
<el>64</el>
<el>49</el>
<el>1</el>
<el>25</el>
<el>81</el>
<el>36</el>
<el>16</el>
randomMap (*2) [1..10] seed:
<el>6</el>
<el>20</el>
<el>4</el>
<el>16</el>
<el>14</el>
<el>2</el>
<el>10</el>
<el>18</el>
<el>12</el>
<el>8</el>
randListIndex [1..10] seed:
<el>3</el>
<el>10</el>
<el>2</el>
<el>8</el>
<el>7</el>
<el>1</el>
<el>5</el>
<el>9</el>
<el>6</el>
<el>4</el>
The last result gives us a random index of the integers in [1, 10]
These indexes can be used to select in a random order 10 nodes.
In XSLT 2 one can use the standard XPath function current-time() for
the creation of the seed:
http://www.w3.org/TR/xpath-functions/#func-current-time
Cheers,
Dimitre
On Sat, Sep 13, 2014 at 7:29 AM, Eliot Kimber ekimber@xxxxxxxxxxxx
<xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:
> Using XSLT 2 I need to implement rendering of "match table" questions
> where you have two sets of items, the match item and the thing it matches
> to. I want to present this as a literal table, where the first column is
> the match-from items in source order and the second column is the match-to
> items, in random order.
>
> I think this is best characterized as a "shuffle" problem, where you want
> to reorder a list randomly but all items in the list must be accounted
> for.
>
> I can think of a recursive algorithm: given a list, generate a random
> integer between 1 and the list length, select that item and add it to the
> result list, then call this function on the original list minus the node
> you just selected.
>
> Is there an easier or more efficient way to do it?
>
> Thanks,
>
> Eliot
> bbbbb
> Eliot Kimber, Owner
> Contrext, LLC
> http://contrext.com
|