Enhance 'pg_createsubscriber' to fetch and append all databases

This patch enhances the 'pg_createsubscriber' utility by adding the '--all'
option. When '--all' is specified, the tool queries the source server
(publisher) for all databases and creates subscriptions on the target server
(subscriber) for databases with matching names.
This simplifies the process of converting a physical standby to a logical
subscriber, particularly during upgrades.

The options '--database', '--publication', '--subscription', and
'--replication-slot' cannot be used when '--all' is specified.

Author: KhannaShubham1197

doc/src/sgml/ref/pg_createsubscriber.sgml

@@ -87,17 +87,37 @@ PostgreSQL documentation
    command-line arguments:
 
    <variablelist>
+    <varlistentry>
+     <term><option>-a</option></term>
+     <term><option>--all</option></term>
+     <listitem>
+      <para>
+       Create one subscription per database on the target server. Exceptions
+       are template databases and databases that don't allow connections.
+       If the database name is not specified in publisher-server, the postgres
+       database will be used, or if that does not exist, template1 will be used.
+       Automatically generated names for subscriptions, publications, and
+       replication slots are used when this option is specified.
+       This option cannot be used along with <option>--database</option>,
+       <option>--publication</option>, <option>--replication-slot</option>, or
+       <option>--subscription</option>.
+      </para>
+     </listitem>
+    </varlistentry>
+
     <varlistentry>
      <term><option>-d <replaceable class="parameter">dbname</replaceable></option></term>
      <term><option>--database=<replaceable class="parameter">dbname</replaceable></option></term>
      <listitem>
       <para>
        The name of the database in which to create a subscription.  Multiple
        databases can be selected by writing multiple <option>-d</option>
-       switches. If <option>-d</option> option is not provided, the database
-       name will be obtained from <option>-P</option> option. If the database
-       name is not specified in either the <option>-d</option> option or
-       <option>-P</option> option, an error will be reported.
+       switches. This option cannot be used together with <option>-a</option>.
+       If <option>-d</option> option is not provided, the database name will be
+       obtained from <option>-P</option> option. If the database name is not
+       specified in either the <option>-d</option> option, or the
+       <option>-P</option> option, and <option>-a</option> option is not
+       specified, an error will be reported.
       </para>
      </listitem>
     </varlistentry>
@@ -253,7 +273,8 @@ PostgreSQL documentation
        names must match the number of specified databases, otherwise an error
        is reported.  The order of the multiple publication name switches must
        match the order of database switches.  If this option is not specified,
-       a generated name is assigned to the publication name.
+       a generated name is assigned to the publication name. This option cannot
+       be used together with <option>--all</option>.
       </para>
      </listitem>
     </varlistentry>
@@ -269,7 +290,8 @@ PostgreSQL documentation
        otherwise an error is reported.  The order of the multiple replication
        slot name switches must match the order of database switches.  If this
        option is not specified, the subscription name is assigned to the
-       replication slot name.
+       replication slot name. This option cannot be used together with
+       <option>--all</option>.
       </para>
      </listitem>
     </varlistentry>
@@ -284,7 +306,8 @@ PostgreSQL documentation
        names must match the number of specified databases, otherwise an error
        is reported.  The order of the multiple subscription name switches must
        match the order of database switches.  If this option is not specified,
-       a generated name is assigned to the subscription name.
+       a generated name is assigned to the subscription name. This option cannot
+       be used together with <option>--all</option>.
       </para>
      </listitem>
     </varlistentry>

src/bin/pg_basebackup/pg_createsubscriber.c

@@ -45,6 +45,7 @@ struct CreateSubscriberOptions
 	SimpleStringList sub_names; /* list of subscription names */
 	SimpleStringList replslot_names;	/* list of replication slot names */
 	int			recovery_timeout;	/* stop recovery after this time */
+	bool		all_dbs;		/* all option */
 	SimpleStringList objecttypes_to_remove; /* list of object types to remove */
 };
 
@@ -124,6 +125,8 @@ static void check_and_drop_existing_subscriptions(PGconn *conn,
 												  const struct LogicalRepInfo *dbinfo);
 static void drop_existing_subscriptions(PGconn *conn, const char *subname,
 										const char *dbname);
+static void get_publisher_databases(struct CreateSubscriberOptions *opt,
+									bool dbnamespecified);
 
 #define	USEC_PER_SEC	1000000
 #define	WAIT_INTERVAL	1		/* 1 second */
@@ -243,6 +246,8 @@ usage(void)
 	printf(_("Usage:\n"));
 	printf(_("  %s [OPTION]...\n"), progname);
 	printf(_("\nOptions:\n"));
+	printf(_("  -a, --all                       create subscriptions for all databases except template\n"
+			 "                                  databases or databases with connection restrictions\n"));
 	printf(_("  -d, --database=DBNAME           database in which to create a subscription\n"));
 	printf(_("  -D, --pgdata=DATADIR            location for the subscriber data directory\n"));
 	printf(_("  -n, --dry-run                   dry run, just show what would be done\n"));
@@ -1959,11 +1964,67 @@ enable_subscription(PGconn *conn, const struct LogicalRepInfo *dbinfo)
 	destroyPQExpBuffer(str);
 }
 
+/*
+ * Fetch a list of all not-template databases from the source server.
+ * Internally, this is treated as if the user specified multiple --database
+ * options, one for each source database.
+ */
+static void
+get_publisher_databases(struct CreateSubscriberOptions *opt,
+						bool dbnamespecified)
+{
+	PGconn	   *conn;
+	PGresult   *res;
+
+	/* If a database name was specified, just connect to it. */
+	if (dbnamespecified)
+		conn = connect_database(opt->pub_conninfo_str, true);
+	else
+	{
+		/* Otherwise, try postgres first and then template1. */
+		char	   *conninfo;
+
+		conninfo = concat_conninfo_dbname(opt->pub_conninfo_str, "postgres");
+		conn = connect_database(conninfo, false);
+		pg_free(conninfo);
+		if (!conn)
+		{
+			conninfo = concat_conninfo_dbname(opt->pub_conninfo_str, "template1");
+			conn = connect_database(conninfo, true);
+			pg_free(conninfo);
+		}
+	}
+
+	res = PQexec(conn, "SELECT datname FROM pg_database WHERE datistemplate = false AND datallowconn AND datconnlimit <> -2 ORDER BY 1");
+
+	/* Check for errors during query execution */
+	if (PQresultStatus(res) != PGRES_TUPLES_OK)
+	{
+		pg_log_error("could not obtain a list of databases: %s", PQresultErrorMessage(res));
+		PQclear(res);
+		disconnect_database(conn, true);
+	}
+
+	for (int i = 0; i < PQntuples(res); i++)
+	{
+		const char *dbname = PQgetvalue(res, i, 0);
+
+		simple_string_list_append(&opt->database_names, dbname);
+
+		/* Increment num_dbs to reflect multiple --database options */
+		num_dbs++;
+	}
+
+	PQclear(res);
+	disconnect_database(conn, false);
+}
+
 int
 main(int argc, char **argv)
 {
 	static struct option long_options[] =
 	{
+		{"all", no_argument, NULL, 'a'},
 		{"database", required_argument, NULL, 'd'},
 		{"pgdata", required_argument, NULL, 'D'},
 		{"dry-run", no_argument, NULL, 'n'},
@@ -2034,6 +2095,7 @@ main(int argc, char **argv)
 		0
 	};
 	opt.recovery_timeout = 0;
+	opt.all_dbs = false;
 
 	/*
 	 * Don't allow it to be run as root. It uses pg_ctl which does not allow
@@ -2051,11 +2113,14 @@ main(int argc, char **argv)
 
 	get_restricted_token();
 
-	while ((c = getopt_long(argc, argv, "d:D:np:P:R:s:t:TU:v",
+	while ((c = getopt_long(argc, argv, "ad:D:np:P:R:s:t:TU:v",
 							long_options, &option_index)) != -1)
 	{
 		switch (c)
 		{
+			case 'a':
+				opt.all_dbs = true;
+				break;
 			case 'd':
 				if (!simple_string_list_member(&opt.database_names, optarg))
 				{
@@ -2149,6 +2214,28 @@ main(int argc, char **argv)
 		}
 	}
 
+	/* Validate that --all is not used with incompatible options */
+	if (opt.all_dbs)
+	{
+		char	   *bad_switch = NULL;
+
+		if (num_dbs > 0)
+			bad_switch = "--database";
+		else if (num_pubs > 0)
+			bad_switch = "--publication";
+		else if (num_replslots > 0)
+			bad_switch = "--replication-slot";
+		else if (num_subs > 0)
+			bad_switch = "--subscription";
+
+		if (bad_switch)
+		{
+			pg_log_error("%s cannot be used with --all", bad_switch);
+			pg_log_error_hint("Try \"%s --help\" for more information.", progname);
+			exit(1);
+		}
+	}
+
 	/* Any non-option arguments? */
 	if (optind < argc)
 	{
@@ -2202,14 +2289,25 @@ main(int argc, char **argv)
 	pg_log_info("validating subscriber connection string");
 	sub_base_conninfo = get_sub_conninfo(&opt);
 
+	/*
+	 * Fetch all databases from the source (publisher) if --all is specified.
+	 * This is treated as if the user specified multiple --database options,
+	 * one for each source database.
+	 */
+	if (opt.all_dbs)
+	{
+		bool		dbnamespecified = (dbname_conninfo != NULL);
+
+		get_publisher_databases(&opt, dbnamespecified);
+	}
+
 	if (opt.database_names.head == NULL)
 	{
 		pg_log_info("no database was specified");
 
 		/*
-		 * If --database option is not provided, try to obtain the dbname from
-		 * the publisher conninfo. If dbname parameter is not available, error
-		 * out.
+		 * Try to obtain the dbname from the publisher conninfo. If dbname
+		 * parameter is not available, error out.
 		 */
 		if (dbname_conninfo)
 		{

src/bin/pg_basebackup/t/040_pg_createsubscriber.pl

@@ -386,6 +386,63 @@ sub generate_db
 	],
 	'run pg_createsubscriber without --databases');
 
+# run pg_createsubscriber with '--database' and '--all' without '--dry-run'
+# and verify the failure
+command_fails_like(
+	[
+		'pg_createsubscriber',
+		'--verbose',
+		'--pgdata' => $node_s->data_dir,
+		'--publisher-server' => $node_p->connstr($db1),
+		'--socketdir' => $node_s->host,
+		'--subscriber-port' => $node_s->port,
+		'--database' => $db1,
+		'--all',
+	],
+	qr/--database cannot be used with --all/,
+	'fail if --database is used with --all');
+
+# run pg_createsubscriber with '--publication' and '--all' and verify
+# the failure
+command_fails_like(
+	[
+		'pg_createsubscriber',
+		'--verbose',
+		'--dry-run',
+		'--pgdata' => $node_s->data_dir,
+		'--publisher-server' => $node_p->connstr($db1),
+		'--socketdir' => $node_s->host,
+		'--subscriber-port' => $node_s->port,
+		'--all',
+		'--publication' => 'pub1',
+	],
+	qr/--publication cannot be used with --all/,
+	'fail if --publication is used with --all');
+
+# run pg_createsubscriber with '--all' option
+my ($stdout, $stderr) = run_command(
+	[
+		'pg_createsubscriber',
+		'--verbose',
+		'--dry-run',
+		'--recovery-timeout' => $PostgreSQL::Test::Utils::timeout_default,
+		'--pgdata' => $node_s->data_dir,
+		'--publisher-server' => $node_p->connstr,
+		'--socketdir' => $node_s->host,
+		'--subscriber-port' => $node_s->port,
+		'--all',
+	],
+	'run pg_createsubscriber with --all');
+
+# Verify that the required logical replication objects are output.
+# The expected count 3 refers to postgres, $db1 and $db2 databases.
+is(scalar(() = $stderr =~ /creating publication/g),
+	3, "verify publications are created for all databases");
+is(scalar(() = $stderr =~ /creating the replication slot/g),
+	3, "verify replication slots are created for all databases");
+is(scalar(() = $stderr =~ /creating subscription/g),
+	3, "verify subscriptions are created for all databases");
+
 # Run pg_createsubscriber on node S.  --verbose is used twice
 # to show more information.
 # In passing, also test the --enable-two-phase option and