<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> <HTML ><HEAD ><TITLE >Implementation</TITLE ><META NAME="GENERATOR" CONTENT="Modular DocBook HTML Stylesheet Version 1.79"><LINK REV="MADE" HREF="mailto:pgsql-docs@postgresql.org"><LINK REL="HOME" TITLE="PostgreSQL 8.4.8 Documentation" HREF="index.html"><LINK REL="UP" TITLE="GiST Indexes" HREF="gist.html"><LINK REL="PREVIOUS" TITLE="Extensibility" HREF="gist-extensibility.html"><LINK REL="NEXT" TITLE="Examples" HREF="gist-examples.html"><LINK REL="STYLESHEET" TYPE="text/css" HREF="stylesheet.css"><META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=ISO-8859-1"><META NAME="creation" CONTENT="2011-04-15T03:27:09"></HEAD ><BODY CLASS="SECT1" ><DIV CLASS="NAVHEADER" ><TABLE SUMMARY="Header navigation table" WIDTH="100%" BORDER="0" CELLPADDING="0" CELLSPACING="0" ><TR ><TH COLSPAN="5" ALIGN="center" VALIGN="bottom" >PostgreSQL 8.4.8 Documentation</TH ></TR ><TR ><TD WIDTH="10%" ALIGN="left" VALIGN="top" ><A HREF="gist-extensibility.html" ACCESSKEY="P" >Prev</A ></TD ><TD WIDTH="10%" ALIGN="left" VALIGN="top" ><A HREF="gist.html" >Fast Backward</A ></TD ><TD WIDTH="60%" ALIGN="center" VALIGN="bottom" >Chapter 51. GiST Indexes</TD ><TD WIDTH="10%" ALIGN="right" VALIGN="top" ><A HREF="gist.html" >Fast Forward</A ></TD ><TD WIDTH="10%" ALIGN="right" VALIGN="top" ><A HREF="gist-examples.html" ACCESSKEY="N" >Next</A ></TD ></TR ></TABLE ><HR ALIGN="LEFT" WIDTH="100%"></DIV ><DIV CLASS="SECT1" ><H1 CLASS="SECT1" ><A NAME="GIST-IMPLEMENTATION" >51.3. Implementation</A ></H1 ><P > There are seven methods that an index operator class for <ACRONYM CLASS="ACRONYM" >GiST</ACRONYM > must provide. Correctness of the index is ensured by proper implementation of the <CODE CLASS="FUNCTION" >same</CODE >, <CODE CLASS="FUNCTION" >consistent</CODE > and <CODE CLASS="FUNCTION" >union</CODE > methods, while efficiency (size and speed) of the index will depend on the <CODE CLASS="FUNCTION" >penalty</CODE > and <CODE CLASS="FUNCTION" >picksplit</CODE > methods. The remaining two methods are <CODE CLASS="FUNCTION" >compress</CODE > and <CODE CLASS="FUNCTION" >decompress</CODE >, which allow an index to have internal tree data of a different type than the data it indexes. The leaves are to be of the indexed data type, while the other tree nodes can be of any C struct (but you still have to follow <SPAN CLASS="PRODUCTNAME" >PostgreSQL</SPAN > datatype rules here, see about <TT CLASS="LITERAL" >varlena</TT > for variable sized data). If the tree's internal data type exists at the SQL level, the <TT CLASS="LITERAL" >STORAGE</TT > option of the <TT CLASS="COMMAND" >CREATE OPERATOR CLASS</TT > command can be used. </P ><P ></P ><DIV CLASS="VARIABLELIST" ><DL ><DT ><CODE CLASS="FUNCTION" >consistent</CODE ></DT ><DD ><P > Given an index entry <TT CLASS="LITERAL" >p</TT > and a query value <TT CLASS="LITERAL" >q</TT >, this function determines whether the index entry is <SPAN CLASS="QUOTE" >"consistent"</SPAN > with the query; that is, could the predicate <SPAN CLASS="QUOTE" >"<TT CLASS="REPLACEABLE" ><I >indexed_column</I ></TT > <TT CLASS="REPLACEABLE" ><I >indexable_operator</I ></TT > <TT CLASS="LITERAL" >q</TT >"</SPAN > be true for any row represented by the index entry? For a leaf index entry this is equivalent to testing the indexable condition, while for an internal tree node this determines whether it is necessary to scan the subtree of the index represented by the tree node. When the result is <TT CLASS="LITERAL" >true</TT >, a <TT CLASS="LITERAL" >recheck</TT > flag must also be returned. This indicates whether the predicate is certainly true or only possibly true. If <TT CLASS="LITERAL" >recheck</TT > = <TT CLASS="LITERAL" >false</TT > then the index has tested the predicate condition exactly, whereas if <TT CLASS="LITERAL" >recheck</TT > = <TT CLASS="LITERAL" >true</TT > the row is only a candidate match. In that case the system will automatically evaluate the <TT CLASS="REPLACEABLE" ><I >indexable_operator</I ></TT > against the actual row value to see if it is really a match. This convention allows <ACRONYM CLASS="ACRONYM" >GiST</ACRONYM > to support both lossless and lossy index structures. </P ><P > The <ACRONYM CLASS="ACRONYM" >SQL</ACRONYM > declaration of the function must look like this: </P><PRE CLASS="PROGRAMLISTING" >CREATE OR REPLACE FUNCTION my_consistent(internal, data_type, smallint, oid, internal) RETURNS bool AS 'MODULE_PATHNAME' LANGUAGE C STRICT;</PRE ><P> And the matching code in the C module could then follow this skeleton: </P><PRE CLASS="PROGRAMLISTING" >Datum my_consistent(PG_FUNCTION_ARGS); PG_FUNCTION_INFO_V1(my_consistent); Datum my_consistent(PG_FUNCTION_ARGS) { GISTENTRY *entry = (GISTENTRY *) PG_GETARG_POINTER(0); data_type *query = PG_GETARG_DATA_TYPE_P(1); StrategyNumber strategy = (StrategyNumber) PG_GETARG_UINT16(2); /* Oid subtype = PG_GETARG_OID(3); */ bool *recheck = (bool *) PG_GETARG_POINTER(4); data_type *key = DatumGetDataType(entry->key); bool retval; /* * determine return value as a function of strategy, key and query. * * Use GIST_LEAF(entry) to know where you're called in the index tree, * which comes handy when supporting the = operator for example (you could * check for non empty union() in non-leaf nodes and equality in leaf * nodes). */ *recheck = true; /* or false if check is exact */ PG_RETURN_BOOL(retval); }</PRE ><P> Here, <TT CLASS="VARNAME" >key</TT > is an element in the index and <TT CLASS="VARNAME" >query</TT > the value being looked up in the index. The <TT CLASS="LITERAL" >StrategyNumber</TT > parameter indicates which operator of your operator class is being applied — it matches one of the operator numbers in the <TT CLASS="COMMAND" >CREATE OPERATOR CLASS</TT > command. Depending on what operators you have included in the class, the data type of <TT CLASS="VARNAME" >query</TT > could vary with the operator, but the above skeleton assumes it doesn't. </P ></DD ><DT ><CODE CLASS="FUNCTION" >union</CODE ></DT ><DD ><P > This method consolidates information in the tree. Given a set of entries, this function generates a new index entry that represents all the given entries. </P ><P > The <ACRONYM CLASS="ACRONYM" >SQL</ACRONYM > declaration of the function must look like this: </P><PRE CLASS="PROGRAMLISTING" >CREATE OR REPLACE FUNCTION my_union(internal, internal) RETURNS internal AS 'MODULE_PATHNAME' LANGUAGE C STRICT;</PRE ><P> And the matching code in the C module could then follow this skeleton: </P><PRE CLASS="PROGRAMLISTING" >Datum my_union(PG_FUNCTION_ARGS); PG_FUNCTION_INFO_V1(my_union); Datum my_union(PG_FUNCTION_ARGS) { GistEntryVector *entryvec = (GistEntryVector *) PG_GETARG_POINTER(0); GISTENTRY *ent = entryvec->vector; data_type *out, *tmp, *old; int numranges, i = 0; numranges = entryvec->n; tmp = DatumGetDataType(ent[0].key); out = tmp; if (numranges == 1) { out = data_type_deep_copy(tmp); PG_RETURN_DATA_TYPE_P(out); } for (i = 1; i < numranges; i++) { old = out; tmp = DatumGetDataType(ent[i].key); out = my_union_implementation(out, tmp); } PG_RETURN_DATA_TYPE_P(out); }</PRE ><P> </P ><P > As you can see, in this skeleton we're dealing with a data type where <TT CLASS="LITERAL" >union(X, Y, Z) = union(union(X, Y), Z)</TT >. It's easy enough to support data types where this is not the case, by implementing the proper union algorithm in this <ACRONYM CLASS="ACRONYM" >GiST</ACRONYM > support method. </P ><P > The <CODE CLASS="FUNCTION" >union</CODE > implementation function should return a pointer to newly <CODE CLASS="FUNCTION" >palloc()</CODE >ed memory. You can't just return whatever the input is. </P ></DD ><DT ><CODE CLASS="FUNCTION" >compress</CODE ></DT ><DD ><P > Converts the data item into a format suitable for physical storage in an index page. </P ><P > The <ACRONYM CLASS="ACRONYM" >SQL</ACRONYM > declaration of the function must look like this: </P><PRE CLASS="PROGRAMLISTING" >CREATE OR REPLACE FUNCTION my_compress(internal) RETURNS internal AS 'MODULE_PATHNAME' LANGUAGE C STRICT;</PRE ><P> And the matching code in the C module could then follow this skeleton: </P><PRE CLASS="PROGRAMLISTING" >Datum my_compress(PG_FUNCTION_ARGS); PG_FUNCTION_INFO_V1(my_compress); Datum my_compress(PG_FUNCTION_ARGS) { GISTENTRY *entry = (GISTENTRY *) PG_GETARG_POINTER(0); GISTENTRY *retval; if (entry->leafkey) { /* replace entry->key with a compressed version */ compressed_data_type *compressed_data = palloc(sizeof(compressed_data_type)); /* fill *compressed_data from entry->key ... */ retval = palloc(sizeof(GISTENTRY)); gistentryinit(*retval, PointerGetDatum(compressed_data), entry->rel, entry->page, entry->offset, FALSE); } else { /* typically we needn't do anything with non-leaf entries */ retval = entry; } PG_RETURN_POINTER(retval); }</PRE ><P> </P ><P > You have to adapt <TT CLASS="REPLACEABLE" ><I >compressed_data_type</I ></TT > to the specific type you're converting to in order to compress your leaf nodes, of course. </P ><P > Depending on your needs, you could also need to care about compressing <TT CLASS="LITERAL" >NULL</TT > values in there, storing for example <TT CLASS="LITERAL" >(Datum) 0</TT > like <TT CLASS="LITERAL" >gist_circle_compress</TT > does. </P ></DD ><DT ><CODE CLASS="FUNCTION" >decompress</CODE ></DT ><DD ><P > The reverse of the <CODE CLASS="FUNCTION" >compress</CODE > method. Converts the index representation of the data item into a format that can be manipulated by the database. </P ><P > The <ACRONYM CLASS="ACRONYM" >SQL</ACRONYM > declaration of the function must look like this: </P><PRE CLASS="PROGRAMLISTING" >CREATE OR REPLACE FUNCTION my_decompress(internal) RETURNS internal AS 'MODULE_PATHNAME' LANGUAGE C STRICT;</PRE ><P> And the matching code in the C module could then follow this skeleton: </P><PRE CLASS="PROGRAMLISTING" >Datum my_decompress(PG_FUNCTION_ARGS); PG_FUNCTION_INFO_V1(my_decompress); Datum my_decompress(PG_FUNCTION_ARGS) { PG_RETURN_POINTER(PG_GETARG_POINTER(0)); }</PRE ><P> The above skeleton is suitable for the case where no decompression is needed. </P ></DD ><DT ><CODE CLASS="FUNCTION" >penalty</CODE ></DT ><DD ><P > Returns a value indicating the <SPAN CLASS="QUOTE" >"cost"</SPAN > of inserting the new entry into a particular branch of the tree. Items will be inserted down the path of least <CODE CLASS="FUNCTION" >penalty</CODE > in the tree. </P ><P > The <ACRONYM CLASS="ACRONYM" >SQL</ACRONYM > declaration of the function must look like this: </P><PRE CLASS="PROGRAMLISTING" >CREATE OR REPLACE FUNCTION my_penalty(internal, internal, internal) RETURNS internal AS 'MODULE_PATHNAME' LANGUAGE C STRICT; -- in some cases penalty functions need not be strict</PRE ><P> And the matching code in the C module could then follow this skeleton: </P><PRE CLASS="PROGRAMLISTING" >Datum my_penalty(PG_FUNCTION_ARGS); PG_FUNCTION_INFO_V1(my_penalty); Datum my_penalty(PG_FUNCTION_ARGS) { GISTENTRY *origentry = (GISTENTRY *) PG_GETARG_POINTER(0); GISTENTRY *newentry = (GISTENTRY *) PG_GETARG_POINTER(1); float *penalty = (float *) PG_GETARG_POINTER(2); data_type *orig = DatumGetDataType(origentry->key); data_type *new = DatumGetDataType(newentry->key); *penalty = my_penalty_implementation(orig, new); PG_RETURN_POINTER(penalty); }</PRE ><P> </P ><P > The <CODE CLASS="FUNCTION" >penalty</CODE > function is crucial to good performance of the index. It'll get used at insertion time to determine which branch to follow when choosing where to add the new entry in the tree. At query time, the more balanced the index, the quicker the lookup. </P ></DD ><DT ><CODE CLASS="FUNCTION" >picksplit</CODE ></DT ><DD ><P > When an index page split is necessary, this function decides which entries on the page are to stay on the old page, and which are to move to the new page. </P ><P > The <ACRONYM CLASS="ACRONYM" >SQL</ACRONYM > declaration of the function must look like this: </P><PRE CLASS="PROGRAMLISTING" >CREATE OR REPLACE FUNCTION my_picksplit(internal, internal) RETURNS internal AS 'MODULE_PATHNAME' LANGUAGE C STRICT;</PRE ><P> And the matching code in the C module could then follow this skeleton: </P><PRE CLASS="PROGRAMLISTING" >Datum my_picksplit(PG_FUNCTION_ARGS); PG_FUNCTION_INFO_V1(my_picksplit); Datum my_picksplit(PG_FUNCTION_ARGS) { GistEntryVector *entryvec = (GistEntryVector *) PG_GETARG_POINTER(0); OffsetNumber maxoff = entryvec->n - 1; GISTENTRY *ent = entryvec->vector; GIST_SPLITVEC *v = (GIST_SPLITVEC *) PG_GETARG_POINTER(1); int i, nbytes; OffsetNumber *left, *right; data_type *tmp_union; data_type *unionL; data_type *unionR; GISTENTRY **raw_entryvec; maxoff = entryvec->n - 1; nbytes = (maxoff + 1) * sizeof(OffsetNumber); v->spl_left = (OffsetNumber *) palloc(nbytes); left = v->spl_left; v->spl_nleft = 0; v->spl_right = (OffsetNumber *) palloc(nbytes); right = v->spl_right; v->spl_nright = 0; unionL = NULL; unionR = NULL; /* Initialize the raw entry vector. */ raw_entryvec = (GISTENTRY **) malloc(entryvec->n * sizeof(void *)); for (i = FirstOffsetNumber; i <= maxoff; i = OffsetNumberNext(i)) raw_entryvec[i] = &(entryvec->vector[i]); for (i = FirstOffsetNumber; i <= maxoff; i = OffsetNumberNext(i)) { int real_index = raw_entryvec[i] - entryvec->vector; tmp_union = DatumGetDataType(entryvec->vector[real_index].key); Assert(tmp_union != NULL); /* * Choose where to put the index entries and update unionL and unionR * accordingly. Append the entries to either v_spl_left or * v_spl_right, and care about the counters. */ if (my_choice_is_left(unionL, curl, unionR, curr)) { if (unionL == NULL) unionL = tmp_union; else unionL = my_union_implementation(unionL, tmp_union); *left = real_index; ++left; ++(v->spl_nleft); } else { /* * Same on the right */ } } v->spl_ldatum = DataTypeGetDatum(unionL); v->spl_rdatum = DataTypeGetDatum(unionR); PG_RETURN_POINTER(v); }</PRE ><P> </P ><P > Like <CODE CLASS="FUNCTION" >penalty</CODE >, the <CODE CLASS="FUNCTION" >picksplit</CODE > function is crucial to good performance of the index. Designing suitable <CODE CLASS="FUNCTION" >penalty</CODE > and <CODE CLASS="FUNCTION" >picksplit</CODE > implementations is where the challenge of implementing well-performing <ACRONYM CLASS="ACRONYM" >GiST</ACRONYM > indexes lies. </P ></DD ><DT ><CODE CLASS="FUNCTION" >same</CODE ></DT ><DD ><P > Returns true if two index entries are identical, false otherwise. </P ><P > The <ACRONYM CLASS="ACRONYM" >SQL</ACRONYM > declaration of the function must look like this: </P><PRE CLASS="PROGRAMLISTING" >CREATE OR REPLACE FUNCTION my_same(internal, internal, internal) RETURNS internal AS 'MODULE_PATHNAME' LANGUAGE C STRICT;</PRE ><P> And the matching code in the C module could then follow this skeleton: </P><PRE CLASS="PROGRAMLISTING" >Datum my_same(PG_FUNCTION_ARGS); PG_FUNCTION_INFO_V1(my_same); Datum my_same(PG_FUNCTION_ARGS) { prefix_range *v1 = PG_GETARG_PREFIX_RANGE_P(0); prefix_range *v2 = PG_GETARG_PREFIX_RANGE_P(1); bool *result = (bool *) PG_GETARG_POINTER(2); *result = my_eq(v1, v2); PG_RETURN_POINTER(result); }</PRE ><P> For historical reasons, the <CODE CLASS="FUNCTION" >same</CODE > function doesn't just return a boolean result; instead it has to store the flag at the location indicated by the third argument. </P ></DD ></DL ></DIV ></DIV ><DIV CLASS="NAVFOOTER" ><HR ALIGN="LEFT" WIDTH="100%"><TABLE SUMMARY="Footer navigation table" WIDTH="100%" BORDER="0" CELLPADDING="0" CELLSPACING="0" ><TR ><TD WIDTH="33%" ALIGN="left" VALIGN="top" ><A HREF="gist-extensibility.html" ACCESSKEY="P" >Prev</A ></TD ><TD WIDTH="34%" ALIGN="center" VALIGN="top" ><A HREF="index.html" ACCESSKEY="H" >Home</A ></TD ><TD WIDTH="33%" ALIGN="right" VALIGN="top" ><A HREF="gist-examples.html" ACCESSKEY="N" >Next</A ></TD ></TR ><TR ><TD WIDTH="33%" ALIGN="left" VALIGN="top" >Extensibility</TD ><TD WIDTH="34%" ALIGN="center" VALIGN="top" ><A HREF="gist.html" ACCESSKEY="U" >Up</A ></TD ><TD WIDTH="33%" ALIGN="right" VALIGN="top" >Examples</TD ></TR ></TABLE ></DIV ></BODY ></HTML >