tame/core/vector/minmax.xml

<?xml version="1.0" encoding="ISO-8859-1"?>
<!--
  Copyright (C) 2014-2019 Ryan Specialty Group, LLC.

  This file is part of tame-core.

  tame-core is free software: you can redistribute it and/or modify it
  under the terms of the GNU General Public License as
  published by the Free Software Foundation, either version 3 of the
  License, or (at your option) any later version.

  This program is distributed in the hope that it will be useful,
  but WITHOUT ANY WARRANTY; without even the implied warranty of
  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  GNU General Public License for more details.

  You should have received a copy of the GNU General Public License
  along with this program.  If not, see <http://www.gnu.org/licenses/>.
-->
<package xmlns="http://www.lovullo.com/rater"
  xmlns:c="http://www.lovullo.com/calc"
  xmlns:t="http://www.lovullo.com/rater/apply-template"
  core="true"
  title="Maximum and Minimum Elements">

  <import package="../base" />
  <import package="../numeric/common" />
  <import package="../numeric/minmax" />


  <section title="Vector Reduction">
    Core currently only offers a~maximum reduction on
    a~vector.  \todo{Add a~minimum reduction.}  \ref{_maxreduce_}
    provides a convenient template-based abstraction.

    <template name="_maxreduce_"
              desc="Reduce a set to its maximum">
      <param name="@values@"   desc="Values to reduce" />
      <param name="@isvector@" desc="Set to 'true' if the nodes should
                                     not be wrapped in c:vector" />

      <param name="@label@" desc="Application label">
        <!-- default empty -->
        <text></text>
      </param>


      <c:apply name="maxreduce" label="@label@">
        <c:arg name="maxreduce_set">
          <!-- if we were not provided with a vector (default), create
               one out of the given nodes -->
          <unless name="@isvector@" eq="true">
            <c:vector>
              <param-copy name="@values@" />
            </c:vector>
          </unless>

          <!-- if they told us that they have provided a vector, then
               do not create one -->
          <if name="@isvector@" eq="true">
            <param-copy name="@values@" />
          </if>
        </c:arg>
      </c:apply>
    </template>


    \ref{maxreduce} is the~function encapsulated by the template and
    may be applied directly.

    \todo{This function was originally written before TAME supported
    let~expressions and other convenience templates.  It has since
    been refactored slightly, but can be made to be more concise.}

    <function name="maxreduce" desc="Reduce a set to its maximum">
      <param name="maxreduce_set" type="float" set="vector"
             desc="Set to find max of" />

      <c:let>
        <c:values>
          <c:value name="n" type="integer"
                   desc="Length of set to reduce">
            <c:length-of>
              <c:value-of name="maxreduce_set" />
            </c:length-of>
          </c:value>
        </c:values>


        <c:cases>
          <!-- if we have no values to reduce, then simply return 0 -->
          <c:case>
            <c:when name="n">
              <c:eq>
                <c:const value="0"
                         desc="When there are no elements" />
              </c:eq>
            </c:when>

            <c:const value="0" type="integer" desc="No value" />
          </c:case>

          <!-- we have values; perform reduction -->
          <c:otherwise>
            <c:apply name="_maxreduce">
              <c:arg name="_maxreduce_set">
                <c:value-of name="maxreduce_set" />
              </c:arg>

              <c:arg name="_maxreduce_i">
                <t:dec>
                  <c:length-of>
                    <c:value-of name="maxreduce_set" />
                  </c:length-of>
                </t:dec>
              </c:arg>
            </c:apply>
          </c:otherwise>
        </c:cases>
      </c:let>
    </function>


    \ref{_maxreduce} is a~helper function that recursively determines
    the maximum value of the given set.  It should not be invoked
    directly; use \ref{maxreduce} or~\ref{_maxreduce_} instead.

    For example, given the vector~\vector{2,4,3,1}, here's the
    recursive application:

    \begingroup
      \def\max{\textrm{max}}

      \begin{equation}
                \max( 1, \max( 3, max( 4, 2 ) ) )
      \\ \yield \max( 1, \max( 3, 4 ) )
      \\ \yield \max( 1, 4 )
      \\ \yield 4
      \end{equation}
    \endgroup

    \todo{If we eventually support first-class functions, then this
    can simply be a~generic reduce function that accepts max/min/etc.}

    \todo{This function was written before local variables generated
    their own unique symbols; strip the prefixes.}

    <function name="_maxreduce"
              desc="Recursively reduce a set to its maximum (called by maxreduce)">
      <param name="_maxreduce_set" type="float" set="vector"
             desc="Set to find max of" />
      <param name="_maxreduce_i"   type="integer"
             desc="Index" />

      <c:cases>
        <!-- base case: if we're on the last index, do not recurse -->
        <c:case>
          <c:when name="_maxreduce_i">
            <c:eq>
              <c:const value="0"
                       desc="Return when only one element remains" />
            </c:eq>
          </c:when>

          <!-- return the first value -->
          <c:value-of name="_maxreduce_set">
            <c:index>
              <c:const value="0"
                       desc="First item in set" />
            </c:index>
          </c:value-of>
        </c:case>

        <!-- we have more elements in the set; recursively determine
             the maximum value -->
        <c:otherwise>
          <c:apply name="max">
            <!-- the first element to compare is our index -->
            <c:arg name="max1">
              <c:value-of name="_maxreduce_set">
                <c:index>
                  <c:value-of name="_maxreduce_i" />
                </c:index>
              </c:value-of>
            </c:arg>

            <!-- and we'll compare to the recursive application of the
                 same set on the previous index -->
            <c:arg name="max2">
              <c:apply name="_maxreduce">
                <c:arg name="_maxreduce_set">
                  <c:value-of name="_maxreduce_set" />
                </c:arg>

                <c:arg name="_maxreduce_i">
                  <c:sum>
                    <c:value-of name="_maxreduce_i" />
                    <c:const value="-1" type="integer" desc="Decrement index by 1" />
                  </c:sum>
                </c:arg>
              </c:apply>
            </c:arg>
          </c:apply>
        </c:otherwise>
      </c:cases>
    </function>
  </section>



  <!-- TODO: Everything below this line must be moved.  This
       technically operates on a matrix as well. -->

  <!-- simplifies retrieving the max of a set of values -->
  <template name="_maxOfEach_" desc="Take the max of the given set of values">
    <param name="@class@" desc="Class to match on" />
    <param name="@values@" desc="Individual values without set" />
    <param name="@generates@" desc="Value to generate into" />
    <param name="@index@" desc="Index to use for rate-each" />

    <param name="@yields@" desc="Yield variable">
      <text>_</text>
      <param-value name="@generates@" />
    </param>

    <rate-each class="@class@" accumulate="none" yields="@yields@" generates="@generates@" index="@index@">
      <c:apply name="maxreduce">
        <c:arg name="maxreduce_set">
          <c:vector>
            <param-copy name="@values@" />
          </c:vector>
        </c:arg>
      </c:apply>
    </rate-each>
  </template>
</package>