<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>67.3. Extensibility</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets V1.79.1" /><link rel="prev" href="brin-builtin-opclasses.html" title="67.2. Built-in Operator Classes" /><link rel="next" href="hash-index.html" title="Chapter 68. Hash Indexes" /></head><body id="docContent" class="container-fluid col-10"><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">67.3. Extensibility</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="brin-builtin-opclasses.html" title="67.2. Built-in Operator Classes">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="brin.html" title="Chapter 67. BRIN Indexes">Up</a></td><th width="60%" align="center">Chapter 67. BRIN Indexes</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 13.4 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="hash-index.html" title="Chapter 68. Hash Indexes">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="BRIN-EXTENSIBILITY"><div class="titlepage"><div><div><h2 class="title" style="clear: both">67.3. Extensibility</h2></div></div></div><p>
  The <acronym class="acronym">BRIN</acronym> interface has a high level of abstraction,
  requiring the access method implementer only to implement the semantics
  of the data type being accessed.  The <acronym class="acronym">BRIN</acronym> layer
  itself takes care of concurrency, logging and searching the index structure.
 </p><p>
  All it takes to get a <acronym class="acronym">BRIN</acronym> access method working is to
  implement a few user-defined methods, which define the behavior of
  summary values stored in the index and the way they interact with
  scan keys.
  In short, <acronym class="acronym">BRIN</acronym> combines
  extensibility with generality, code reuse, and a clean interface.
 </p><p>
  There are four methods that an operator class for <acronym class="acronym">BRIN</acronym>
  must provide:

  </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="function">BrinOpcInfo *opcInfo(Oid type_oid)</code></span></dt><dd><p>
      Returns internal information about the indexed columns' summary data.
      The return value must point to a palloc'd <code class="structname">BrinOpcInfo</code>,
      which has this definition:
</p><pre class="programlisting">
typedef struct BrinOpcInfo
{
    /* Number of columns stored in an index column of this opclass */
    uint16      oi_nstored;

    /* Opaque pointer for the opclass' private use */
    void       *oi_opaque;

    /* Type cache entries of the stored columns */
    TypeCacheEntry *oi_typcache[FLEXIBLE_ARRAY_MEMBER];
} BrinOpcInfo;
</pre><p>
      <code class="structname">BrinOpcInfo</code>.<code class="structfield">oi_opaque</code> can be used by the
      operator class routines to pass information between support functions
      during an index scan.
     </p></dd><dt><span class="term"><code class="function">bool consistent(BrinDesc *bdesc, BrinValues *column,
       ScanKey key)</code></span></dt><dd><p>
      Returns whether the ScanKey is consistent with the given indexed
      values for a range.
      The attribute number to use is passed as part of the scan key.
     </p></dd><dt><span class="term"><code class="function">bool addValue(BrinDesc *bdesc, BrinValues *column,
       Datum newval, bool isnull)</code></span></dt><dd><p>
      Given an index tuple and an indexed value, modifies the indicated
      attribute of the tuple so that it additionally represents the new value.
      If any modification was done to the tuple, <code class="literal">true</code> is
      returned.
     </p></dd><dt><span class="term"><code class="function">bool unionTuples(BrinDesc *bdesc, BrinValues *a,
       BrinValues *b)</code></span></dt><dd><p>
      Consolidates two index tuples. Given two index tuples, modifies the
      indicated attribute of the first of them so that it represents both tuples.
      The second tuple is not modified.
     </p></dd></dl></div><p>

  An operator class for <acronym class="acronym">BRIN</acronym> can optionally specify the
  following method:

  </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="function">void options(local_relopts *relopts)</code></span></dt><dd><p>
       Defines a set of user-visible parameters that control operator class
       behavior.
      </p><p>
       The <code class="function">options</code> function is passed a pointer to a
       <em class="replaceable"><code>local_relopts</code></em> struct, which needs to be
       filled with a set of operator class specific options.  The options
       can be accessed from other support functions using the
       <code class="literal">PG_HAS_OPCLASS_OPTIONS()</code> and
       <code class="literal">PG_GET_OPCLASS_OPTIONS()</code> macros.
      </p><p>
       Since both key extraction of indexed values and representation of the
       key in <acronym class="acronym">BRIN</acronym> are flexible, they may depend on
       user-specified parameters.
      </p></dd></dl></div><p>

  The core distribution includes support for two types of operator classes:
  minmax and inclusion.  Operator class definitions using them are shipped for
  in-core data types as appropriate.  Additional operator classes can be
  defined by the user for other data types using equivalent definitions,
  without having to write any source code; appropriate catalog entries being
  declared is enough.  Note that assumptions about the semantics of operator
  strategies are embedded in the support functions' source code.
 </p><p>
  Operator classes that implement completely different semantics are also
  possible, provided implementations of the four main support functions
  described above are written.  Note that backwards compatibility across major
  releases is not guaranteed: for example, additional support functions might
  be required in later releases.
 </p><p>
  To write an operator class for a data type that implements a totally
  ordered set, it is possible to use the minmax support functions
  alongside the corresponding operators, as shown in
  <a class="xref" href="brin-extensibility.html#BRIN-EXTENSIBILITY-MINMAX-TABLE" title="Table 67.2. Function and Support Numbers for Minmax Operator Classes">Table 67.2</a>.
  All operator class members (functions and operators) are mandatory.
 </p><div class="table" id="BRIN-EXTENSIBILITY-MINMAX-TABLE"><p class="title"><strong>Table 67.2. Function and Support Numbers for Minmax Operator Classes</strong></p><div class="table-contents"><table class="table" summary="Function and Support Numbers for Minmax Operator Classes" border="1"><colgroup><col class="col1" /><col class="col2" /></colgroup><thead><tr><th>Operator class member</th><th>Object</th></tr></thead><tbody><tr><td>Support Function 1</td><td>internal function <code class="function">brin_minmax_opcinfo()</code></td></tr><tr><td>Support Function 2</td><td>internal function <code class="function">brin_minmax_add_value()</code></td></tr><tr><td>Support Function 3</td><td>internal function <code class="function">brin_minmax_consistent()</code></td></tr><tr><td>Support Function 4</td><td>internal function <code class="function">brin_minmax_union()</code></td></tr><tr><td>Operator Strategy 1</td><td>operator less-than</td></tr><tr><td>Operator Strategy 2</td><td>operator less-than-or-equal-to</td></tr><tr><td>Operator Strategy 3</td><td>operator equal-to</td></tr><tr><td>Operator Strategy 4</td><td>operator greater-than-or-equal-to</td></tr><tr><td>Operator Strategy 5</td><td>operator greater-than</td></tr></tbody></table></div></div><br class="table-break" /><p>
  To write an operator class for a complex data type which has values
  included within another type, it's possible to use the inclusion support
  functions alongside the corresponding operators, as shown
  in <a class="xref" href="brin-extensibility.html#BRIN-EXTENSIBILITY-INCLUSION-TABLE" title="Table 67.3. Function and Support Numbers for Inclusion Operator Classes">Table 67.3</a>.  It requires
  only a single additional function, which can be written in any language.
  More functions can be defined for additional functionality.  All operators
  are optional.  Some operators require other operators, as shown as
  dependencies on the table.
 </p><div class="table" id="BRIN-EXTENSIBILITY-INCLUSION-TABLE"><p class="title"><strong>Table 67.3. Function and Support Numbers for Inclusion Operator Classes</strong></p><div class="table-contents"><table class="table" summary="Function and Support Numbers for Inclusion Operator Classes" border="1"><colgroup><col class="col1" /><col class="col2" /><col class="col3" /></colgroup><thead><tr><th>Operator class member</th><th>Object</th><th>Dependency</th></tr></thead><tbody><tr><td>Support Function 1</td><td>internal function <code class="function">brin_inclusion_opcinfo()</code></td><td> </td></tr><tr><td>Support Function 2</td><td>internal function <code class="function">brin_inclusion_add_value()</code></td><td> </td></tr><tr><td>Support Function 3</td><td>internal function <code class="function">brin_inclusion_consistent()</code></td><td> </td></tr><tr><td>Support Function 4</td><td>internal function <code class="function">brin_inclusion_union()</code></td><td> </td></tr><tr><td>Support Function 11</td><td>function to merge two elements</td><td> </td></tr><tr><td>Support Function 12</td><td>optional function to check whether two elements are mergeable</td><td> </td></tr><tr><td>Support Function 13</td><td>optional function to check if an element is contained within another</td><td> </td></tr><tr><td>Support Function 14</td><td>optional function to check whether an element is empty</td><td> </td></tr><tr><td>Operator Strategy 1</td><td>operator left-of</td><td>Operator Strategy 4</td></tr><tr><td>Operator Strategy 2</td><td>operator does-not-extend-to-the-right-of</td><td>Operator Strategy 5</td></tr><tr><td>Operator Strategy 3</td><td>operator overlaps</td><td> </td></tr><tr><td>Operator Strategy 4</td><td>operator does-not-extend-to-the-left-of</td><td>Operator Strategy 1</td></tr><tr><td>Operator Strategy 5</td><td>operator right-of</td><td>Operator Strategy 2</td></tr><tr><td>Operator Strategy 6, 18</td><td>operator same-as-or-equal-to</td><td>Operator Strategy 7</td></tr><tr><td>Operator Strategy 7, 13, 16, 24, 25</td><td>operator contains-or-equal-to</td><td> </td></tr><tr><td>Operator Strategy 8, 14, 26, 27</td><td>operator is-contained-by-or-equal-to</td><td>Operator Strategy 3</td></tr><tr><td>Operator Strategy 9</td><td>operator does-not-extend-above</td><td>Operator Strategy 11</td></tr><tr><td>Operator Strategy 10</td><td>operator is-below</td><td>Operator Strategy 12</td></tr><tr><td>Operator Strategy 11</td><td>operator is-above</td><td>Operator Strategy 9</td></tr><tr><td>Operator Strategy 12</td><td>operator does-not-extend-below</td><td>Operator Strategy 10</td></tr><tr><td>Operator Strategy 20</td><td>operator less-than</td><td>Operator Strategy 5</td></tr><tr><td>Operator Strategy 21</td><td>operator less-than-or-equal-to</td><td>Operator Strategy 5</td></tr><tr><td>Operator Strategy 22</td><td>operator greater-than</td><td>Operator Strategy 1</td></tr><tr><td>Operator Strategy 23</td><td>operator greater-than-or-equal-to</td><td>Operator Strategy 1</td></tr></tbody></table></div></div><br class="table-break" /><p>
    Support function numbers 1 through 10 are reserved for the BRIN internal
    functions, so the SQL level functions start with number 11.  Support
    function number 11 is the main function required to build the index.
    It should accept two arguments with the same data type as the operator class,
    and return the union of them.  The inclusion operator class can store union
    values with different data types if it is defined with the
    <code class="literal">STORAGE</code> parameter.  The return value of the union
    function should match the <code class="literal">STORAGE</code> data type.
 </p><p>
    Support function numbers 12 and 14 are provided to support
    irregularities of built-in data types.  Function number 12
    is used to support network addresses from different families which
    are not mergeable.  Function number 14 is used to support
    empty ranges.  Function number 13 is an optional but
    recommended one, which allows the new value to be checked before
    it is passed to the union function.  As the BRIN framework can shortcut
    some operations when the union is not changed, using this
    function can improve index performance.
 </p><p>
    Both minmax and inclusion operator classes support cross-data-type
    operators, though with these the dependencies become more complicated.
    The minmax operator class requires a full set of operators to be
    defined with both arguments having the same data type.  It allows
    additional data types to be supported by defining extra sets
    of operators.  Inclusion operator class operator strategies are dependent
    on another operator strategy as shown in
    <a class="xref" href="brin-extensibility.html#BRIN-EXTENSIBILITY-INCLUSION-TABLE" title="Table 67.3. Function and Support Numbers for Inclusion Operator Classes">Table 67.3</a>, or the same
    operator strategy as themselves.  They require the dependency
    operator to be defined with the <code class="literal">STORAGE</code> data type as the
    left-hand-side argument and the other supported data type to be the
    right-hand-side argument of the supported operator.  See
    <code class="literal">float4_minmax_ops</code> as an example of minmax, and
    <code class="literal">box_inclusion_ops</code> as an example of inclusion.
 </p></div><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navfooter"><hr></hr><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="brin-builtin-opclasses.html" title="67.2. Built-in Operator Classes">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="brin.html" title="Chapter 67. BRIN Indexes">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="hash-index.html" title="Chapter 68. Hash Indexes">Next</a></td></tr><tr><td width="40%" align="left" valign="top">67.2. Built-in Operator Classes </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 13.4 Documentation">Home</a></td><td width="40%" align="right" valign="top"> Chapter 68. Hash Indexes</td></tr></table></div></body></html>