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 fb2889e9c082772489efd6453ad5597a25406bad..2a3a27a98a191767e56baa8ace46c45aa7ad6275 100644 (file)
--- a/src/bin/pg_dump/common.c
+++ b/src/bin/pg_dump/common.c
@@ -51,14 +51,17 @@ 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,
@@ -80,7 +83,6 @@ 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)
    TSTemplateInfo *tmplinfo;
    TSDictInfo *dictinfo;
    TSConfigInfo *cfginfo;
-   int         numNamespaces;
    int         numAggregates;
    int         numInherits;
    int         numRules;
@@ -109,7 +110,8 @@ 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
@@ -736,6 +738,17 @@ 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 d4c82bea8be99cc9f446f732a0bfe1a7527287c6..e289f9e2ba7eaa8e565a0f9dfcfd4d440101efca 100644 (file)
--- a/src/bin/pg_dump/pg_dump.c
+++ b/src/bin/pg_dump/pg_dump.c
@@ -111,10 +111,6 @@ 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;
-
 /* flag to turn on/off dollar quoting */
 static int disable_dollar_quoting = 0;
 
@@ -1887,8 +1883,7 @@ getNamespaces(int *numNamespaces)
 
        selectDumpableNamespace(&nsinfo[1]);
 
-       g_namespaces = nsinfo;
-       g_numNamespaces = *numNamespaces = 2;
+       *numNamespaces = 2;
 
        return nsinfo;
    }
@@ -1941,8 +1936,7 @@ getNamespaces(int *numNamespaces)
    PQclear(res);
    destroyPQExpBuffer(query);
 
-   g_namespaces = nsinfo;
-   g_numNamespaces = *numNamespaces = ntups;
+   *numNamespaces = ntups;
 
    return nsinfo;
 }
@@ -1953,36 +1947,37 @@ 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 3d213db0744c989cdc3793c159db19b6a1e998dc..bd1cd72a4ec9f917377eb67f53194e607c2ccccd 100644 (file)
--- a/src/bin/pg_dump/pg_dump.h
+++ b/src/bin/pg_dump/pg_dump.h
@@ -445,6 +445,7 @@ 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);