From: Tom Lane <tgl@sss.pgh.pa.us>
Date: Fri, 25 May 2012 18:35:53 +0000 (-0400)
Subject: Use binary search instead of brute-force scan in findNamespace().
X-Git-Tag: REL8_4_12~10
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=5e0f3c793a3767f38498d9cd54a3bfeeccdce6bf;p=postgresql

Use binary search instead of brute-force scan in findNamespace().

The previous coding presented a significant bottleneck when dumping
databases containing many thousands of schemas, since the total time
spent searching would increase roughly as O(N^2) in the number of objects.
Noted by Jeff Janes, though I rewrote his proposed patch to use the
existing findObjectByOid infrastructure.

Since this is a longstanding performance bug, backpatch to all supported
versions.
---

diff --git a/src/bin/pg_dump/common.c b/src/bin/pg_dump/common.c
index 04379e39c4..e159bc4478 100644
--- a/src/bin/pg_dump/common.c
+++ b/src/bin/pg_dump/common.c
@@ -50,14 +50,17 @@ static TableInfo *tblinfo;
 static TypeInfo *typinfo;
 static FuncInfo *funinfo;
 static OprInfo *oprinfo;
+static NamespaceInfo *nspinfo;
 static int	numTables;
 static int	numTypes;
 static int	numFuncs;
 static int	numOperators;
+static int	numNamespaces;
 static DumpableObject **tblinfoindex;
 static DumpableObject **typinfoindex;
 static DumpableObject **funinfoindex;
 static DumpableObject **oprinfoindex;
+static DumpableObject **nspinfoindex;
 
 
 static void flagInhTables(TableInfo *tbinfo, int numTables,
@@ -78,7 +81,6 @@ static int	strInArray(const char *pattern, char **arr, int arr_size);
 TableInfo *
 getSchemaData(int *numTablesPtr)
 {
-	NamespaceInfo *nsinfo;
 	AggInfo    *agginfo;
 	InhInfo    *inhinfo;
 	RuleInfo   *ruleinfo;
@@ -93,7 +95,6 @@ getSchemaData(int *numTablesPtr)
 	TSConfigInfo *cfginfo;
 	FdwInfo    *fdwinfo;
 	ForeignServerInfo *srvinfo;
-	int			numNamespaces;
 	int			numAggregates;
 	int			numInherits;
 	int			numRules;
@@ -111,7 +112,8 @@ getSchemaData(int *numTablesPtr)
 
 	if (g_verbose)
 		write_msg(NULL, "reading schemas\n");
-	nsinfo = getNamespaces(&numNamespaces);
+	nspinfo = getNamespaces(&numNamespaces);
+	nspinfoindex = buildIndexArray(nspinfo, numNamespaces, sizeof(NamespaceInfo));
 
 	/*
 	 * getTables should be done as soon as possible, so as to minimize the
@@ -708,6 +710,17 @@ findOprByOid(Oid oid)
 	return (OprInfo *) findObjectByOid(oid, oprinfoindex, numOperators);
 }
 
+/*
+ * findNamespaceByOid
+ *	  finds the entry (in nspinfo) of the namespace with the given oid
+ *	  returns NULL if not found
+ */
+NamespaceInfo *
+findNamespaceByOid(Oid oid)
+{
+	return (NamespaceInfo *) findObjectByOid(oid, nspinfoindex, numNamespaces);
+}
+
 
 /*
  * findParentsByOid
diff --git a/src/bin/pg_dump/pg_dump.c b/src/bin/pg_dump/pg_dump.c
index 0981fbd411..7c4730414e 100644
--- a/src/bin/pg_dump/pg_dump.c
+++ b/src/bin/pg_dump/pg_dump.c
@@ -102,10 +102,6 @@ char		g_comment_end[10];
 
 static const CatalogId nilCatalogId = {0, 0};
 
-/* these are to avoid passing around info for findNamespace() */
-static NamespaceInfo *g_namespaces;
-static int	g_numNamespaces;
-
 /* flags for various command-line long options */
 static int	binary_upgrade = 0;
 static int	disable_dollar_quoting = 0;
@@ -2184,8 +2180,7 @@ getNamespaces(int *numNamespaces)
 
 		selectDumpableNamespace(&nsinfo[1]);
 
-		g_namespaces = nsinfo;
-		g_numNamespaces = *numNamespaces = 2;
+		*numNamespaces = 2;
 
 		return nsinfo;
 	}
@@ -2238,8 +2233,7 @@ getNamespaces(int *numNamespaces)
 	PQclear(res);
 	destroyPQExpBuffer(query);
 
-	g_namespaces = nsinfo;
-	g_numNamespaces = *numNamespaces = ntups;
+	*numNamespaces = ntups;
 
 	return nsinfo;
 }
@@ -2250,36 +2244,37 @@ getNamespaces(int *numNamespaces)
  *		getNamespaces
  *
  * NB: for pre-7.3 source database, we use object OID to guess whether it's
- * a system object or not.	In 7.3 and later there is no guessing.
+ * a system object or not.	In 7.3 and later there is no guessing, and we
+ * don't use objoid at all.
  */
 static NamespaceInfo *
 findNamespace(Oid nsoid, Oid objoid)
 {
-	int			i;
+	NamespaceInfo *nsinfo;
 
 	if (g_fout->remoteVersion >= 70300)
 	{
-		for (i = 0; i < g_numNamespaces; i++)
-		{
-			NamespaceInfo *nsinfo = &g_namespaces[i];
-
-			if (nsoid == nsinfo->dobj.catId.oid)
-				return nsinfo;
-		}
-		write_msg(NULL, "schema with OID %u does not exist\n", nsoid);
-		exit_nicely();
+		nsinfo = findNamespaceByOid(nsoid);
 	}
 	else
 	{
-		/* This code depends on the layout set up by getNamespaces. */
+		/* This code depends on the dummy objects set up by getNamespaces. */
+		Oid		i;
+
 		if (objoid > g_last_builtin_oid)
 			i = 0;				/* user object */
 		else
 			i = 1;				/* system object */
-		return &g_namespaces[i];
+		nsinfo = findNamespaceByOid(i);
 	}
 
-	return NULL;				/* keep compiler quiet */
+	if (nsinfo == NULL)
+	{
+		write_msg(NULL, "schema with OID %u does not exist\n", nsoid);
+		exit_nicely();
+	}
+
+	return nsinfo;
 }
 
 /*
diff --git a/src/bin/pg_dump/pg_dump.h b/src/bin/pg_dump/pg_dump.h
index 017c04a48a..10ce6729a2 100644
--- a/src/bin/pg_dump/pg_dump.h
+++ b/src/bin/pg_dump/pg_dump.h
@@ -457,6 +457,7 @@ extern TableInfo *findTableByOid(Oid oid);
 extern TypeInfo *findTypeByOid(Oid oid);
 extern FuncInfo *findFuncByOid(Oid oid);
 extern OprInfo *findOprByOid(Oid oid);
+extern NamespaceInfo *findNamespaceByOid(Oid oid);
 
 extern void simple_oid_list_append(SimpleOidList *list, Oid val);
 extern void simple_string_list_append(SimpleStringList *list, const char *val);