clarified the POD in Pg storage driver regarding multi-schema support

[dbsrgits/DBIx-Class.git] / lib / DBIx / Class / Storage / DBI / Pg.pm

diff --git a/lib/DBIx/Class/Storage/DBI/Pg.pm b/lib/DBIx/Class/Storage/DBI/Pg.pm
index b3d88f2..1a2bd8a 100644 (file)
--- a/lib/DBIx/Class/Storage/DBI/Pg.pm
+++ b/lib/DBIx/Class/Storage/DBI/Pg.pm
@@ -39,6 +39,7 @@ sub _dbh_last_insert_id {
 }
 
 
+# get the postgres search path, and cache it
 sub _get_pg_search_path {
     my ($self,$dbh) = @_;
     # cache the search path as ['schema','schema',...] in the storage
@@ -172,13 +173,17 @@ This class implements autoincrements for PostgreSQL.
 
 =head1 POSTGRESQL SCHEMA SUPPORT
 
-This supports multiple PostgreSQL schemas, with one caveat: for
-performance reasons, the schema search path is queried the first time it is
-needed and CACHED for subsequent uses.
+This driver supports multiple PostgreSQL schemas, with one caveat: for
+performance reasons, data about the search path, sequence names, and
+so forth is queried as needed and CACHED for subsequent uses.
 
-For this reason, you should do any necessary manipulation of the
-PostgreSQL search path BEFORE instantiating your schema object, or as
-part of the on_connect_do option to connect(), for example:
+For this reason, once your schema is instantiated, you should not
+change the PostgreSQL schema search path for that schema's database
+connection. If you do, Bad Things may happen.
+
+You should do any necessary manipulation of the search path BEFORE
+instantiating your schema object, or as part of the on_connect_do
+option to connect(), for example:
 
    my $schema = My::Schema->connect
                   ( $dsn,$user,$pass,