Support for pg_dump to dump tables in cluster order if a clustered index is defined on the table, a little hacked in with how the data is passed around and how the order is pulled out of the db. The latter is the only semi-problematic part as you might be able to generate (very odd) table column names that would break the regex used there which would cause the sql query to be invalid and therefore not dump data for that table. But as long as you don't name an clustered column/function something like "foo ) WHERE" or the like should be ok.

    Pulled from REL9_1_2_P and fixed some merge conflicts

    Conflicts:
    	src/bin/pg_dump/pg_dump.c

Author: tgarnett

src/bin/pg_dump/pg_dump.c

@@ -133,6 +133,7 @@ static const CatalogId nilCatalogId = {0, 0};
 static int	binary_upgrade = 0;
 static int	disable_dollar_quoting = 0;
 static int	dump_inserts = 0;
+static int	cluster_order = 0;
 static int	column_inserts = 0;
 static int	no_security_labels = 0;
 static int	no_unlogged_table_data = 0;
@@ -336,6 +337,7 @@ main(int argc, char **argv)
 		 */
 		{"attribute-inserts", no_argument, &column_inserts, 1},
 		{"binary-upgrade", no_argument, &binary_upgrade, 1},
+		{"cluster-order", no_argument, &cluster_order, 1},
 		{"column-inserts", no_argument, &column_inserts, 1},
 		{"disable-dollar-quoting", no_argument, &disable_dollar_quoting, 1},
 		{"disable-triggers", no_argument, &disable_triggers, 1},
@@ -848,6 +850,7 @@ help(const char *progname)
 	printf(_("  -T, --exclude-table=TABLE    do NOT dump the named table(s)\n"));
 	printf(_("  -x, --no-privileges          do not dump privileges (grant/revoke)\n"));
 	printf(_("  --binary-upgrade             for use by upgrade utilities only\n"));
+	printf(_("  --cluster-order              dump table data in clustered index order (>= 8.2)\n"));
 	printf(_("  --column-inserts             dump data as INSERT commands with column names\n"));
 	printf(_("  --disable-dollar-quoting     disable dollar quoting, use SQL standard quoting\n"));
 	printf(_("  --disable-triggers           disable triggers during data-only restore\n"));
@@ -1296,7 +1299,7 @@ dumpTableData_copy(Archive *fout, void *dcontext)
 										 classname),
 						  column_list);
 	}
-	else if (tdinfo->filtercond)
+	else if (tdinfo->filtercond || tbinfo->ordercond)
 	{
 		/* Note: this syntax is only supported in 8.2 and up */
 		appendPQExpBufferStr(q, "COPY (SELECT ");
@@ -1308,11 +1311,15 @@ dumpTableData_copy(Archive *fout, void *dcontext)
 		}
 		else
 			appendPQExpBufferStr(q, "* ");
-		appendPQExpBuffer(q, "FROM %s %s) TO stdout;",
+		appendPQExpBuffer(q, "FROM %s ",
 						  fmtQualifiedId(fout,
 										 tbinfo->dobj.namespace->dobj.name,
-										 classname),
-						  tdinfo->filtercond);
+										 classname));
+		if (tdinfo->filtercond)
+		  appendPQExpBuffer(q, "%s ", tdinfo->filtercond);
+		if (tbinfo->ordercond)
+		  appendPQExpBuffer(q, "%s", tbinfo->ordercond);
+		appendPQExpBuffer(q, ") TO stdout;");
 	}
 	else
 	{
@@ -1456,6 +1463,8 @@ dumpTableData_insert(Archive *fout, void *dcontext)
 	}
 	if (tdinfo->filtercond)
 		appendPQExpBuffer(q, " %s", tdinfo->filtercond);
+	if (tbinfo->ordercond)
+		appendPQExpBuffer(q, " %s", tbinfo->ordercond);
 
 	ExecuteSqlStatement(fout, q->data);
 
@@ -4473,6 +4482,7 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 				i_oid,
 				i_indexname,
 				i_indexdef,
+				i_indexdeforderclause,
 				i_indnkeys,
 				i_indkey,
 				i_indisclustered,
@@ -4528,6 +4538,14 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 							  "SELECT t.tableoid, t.oid, "
 							  "t.relname AS indexname, "
 					 "pg_catalog.pg_get_indexdef(i.indexrelid) AS indexdef, "
+					 /*
+					  * FIXME - regex is not the best way to to do this, but
+					  * indexdef is consistent enough that it should work
+					  * Any missing corner cases worth worrying about?
+					  * A column named "foo ) WITH" would be problematic...
+					  * but would need a parser to cover all possible cases
+					  */
+					 "CASE WHEN i.indisclustered THEN 'ORDER BY ' || regexp_replace(pg_catalog.pg_get_indexdef(i.indexrelid), E'^CREATE (UNIQUE )?INDEX (CONCURRENTLY )?.* ON .* USING [^\\\\(]*\\\\((.*?)\\\\)($| WITH .*| WHERE .*| TABLESPACE .*)', E'\\\\3') ELSE NULL END AS indexdeforderclause, "
 							  "t.relnatts AS indnkeys, "
 							  "i.indkey, i.indisclustered, "
 							  "c.contype, c.conname, "
@@ -4554,6 +4572,14 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 							  "SELECT t.tableoid, t.oid, "
 							  "t.relname AS indexname, "
 					 "pg_catalog.pg_get_indexdef(i.indexrelid) AS indexdef, "
+					 /*
+					  * FIXME - regex is not the best way to to do this, but
+					  * indexdef is consistent enough that it should work
+					  * Any missing corner cases worth worrying about?
+					  * A column named "foo ) WITH" would be problematic...
+					  * but would need a parser to cover all possible cases
+					  */
+					 "CASE WHEN i.indisclustered THEN 'ORDER BY ' || regexp_replace(pg_catalog.pg_get_indexdef(i.indexrelid), E'^CREATE (UNIQUE )?INDEX (CONCURRENTLY )?.* ON .* USING [^\\\\(]*\\\\((.*?)\\\\)($| WITH .*| WHERE .*| TABLESPACE .*)', E'\\\\3') ELSE NULL END AS indexdeforderclause, "
 							  "t.relnatts AS indnkeys, "
 							  "i.indkey, i.indisclustered, "
 							  "c.contype, c.conname, "
@@ -4583,6 +4609,7 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 							  "SELECT t.tableoid, t.oid, "
 							  "t.relname AS indexname, "
 					 "pg_catalog.pg_get_indexdef(i.indexrelid) AS indexdef, "
+					 "NULL AS indexdeforderclause, "
 							  "t.relnatts AS indnkeys, "
 							  "i.indkey, i.indisclustered, "
 							  "c.contype, c.conname, "
@@ -4611,6 +4638,7 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 							  "SELECT t.tableoid, t.oid, "
 							  "t.relname AS indexname, "
 					 "pg_catalog.pg_get_indexdef(i.indexrelid) AS indexdef, "
+					 "NULL AS indexdeforderclause, "
 							  "t.relnatts AS indnkeys, "
 							  "i.indkey, i.indisclustered, "
 							  "c.contype, c.conname, "
@@ -4639,6 +4667,7 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 							  "SELECT t.tableoid, t.oid, "
 							  "t.relname AS indexname, "
 							  "pg_get_indexdef(i.indexrelid) AS indexdef, "
+							  "NULL AS indexdeforderclause, "
 							  "t.relnatts AS indnkeys, "
 							  "i.indkey, false AS indisclustered, "
 							  "CASE WHEN i.indisprimary THEN 'p'::char "
@@ -4665,6 +4694,7 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 							  "t.oid, "
 							  "t.relname AS indexname, "
 							  "pg_get_indexdef(i.indexrelid) AS indexdef, "
+							  "NULL AS indexdeforderclause, "
 							  "t.relnatts AS indnkeys, "
 							  "i.indkey, false AS indisclustered, "
 							  "CASE WHEN i.indisprimary THEN 'p'::char "
@@ -4692,6 +4722,7 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 		i_oid = PQfnumber(res, "oid");
 		i_indexname = PQfnumber(res, "indexname");
 		i_indexdef = PQfnumber(res, "indexdef");
+		i_indexdeforderclause = PQfnumber(res, "indexdeforderclause");
 		i_indnkeys = PQfnumber(res, "indnkeys");
 		i_indkey = PQfnumber(res, "indkey");
 		i_indisclustered = PQfnumber(res, "indisclustered");
@@ -4779,6 +4810,15 @@ getIndexes(Archive *fout, TableInfo tblinfo[], int numTables)
 				/* Plain secondary index */
 				indxinfo[j].indexconstraint = 0;
 			}
+
+			/*
+			 * propogate table order clause if clustered index for this
+			 * table and --cluster-order is specified
+			 */
+			if (cluster_order && fout->remoteVersion >= 80200 &&
+				indxinfo[j].indisclustered && !tbinfo->ordercond &&
+				PQgetvalue(res, j, i_indexdeforderclause))
+				tbinfo->ordercond = strdup(PQgetvalue(res, j, i_indexdeforderclause));
 		}
 
 		PQclear(res);

src/bin/pg_dump/pg_dump.h

@@ -291,6 +291,7 @@ typedef struct _tableInfo
 	int			numParents;		/* number of (immediate) parent tables */
 	struct _tableInfo **parents;	/* TableInfos of immediate parents */
 	struct _tableDataInfo *dataObj;		/* TableDataInfo, if dumping its data */
+	char     *ordercond;    /* ORDER condition to order dumped rows, should prob. be in dataObj, but we find it out from the indexes before that's created (though arguably should do it's own query since that's the only way to support --data-only with --cluster-order) */
 } TableInfo;
 
 typedef struct _attrDefInfo