pg_replication: add a typical example of initial load (#648)

* add a typical example of initial load

* fixing lint errors

* Example load with no manual source creation

* No ambiguous source and destination pipelines

---------

Co-authored-by: anuunchin <88698977+anuunchin@users.noreply.github.com>

Author: molkazhani2001

sources/pg_replication/README.md

@@ -6,6 +6,7 @@ Resources that can be loaded using this verified source are:
 | Name                 | Description                                     |
 |----------------------|-------------------------------------------------|
 | replication_resource | Load published messages from a replication slot |
+| init_replication     | Initialize replication and optionally return snapshot resources for initial data load  |
 
 ## Initialize the pipeline
 
@@ -29,6 +30,13 @@ It also needs `CREATE` privilege on the database:
 GRANT CREATE ON DATABASE dlt_data TO replication_user;
 ```
 
+If not a superuser, the user must have ownership of the tables that need to be replicated:
+
+```sql
+ALTER TABLE your_table OWNER TO replication_user;  
+```
+
+
 ### Set up RDS
 1. You must enable replication for RDS Postgres instance via **Parameter Group**: https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/USER_PostgreSQL.Replication.ReadReplicas.html
 2. `WITH LOGIN REPLICATION;` does not work on RDS, instead do:

sources/pg_replication/helpers.py

@@ -42,7 +42,7 @@
 from dlt.sources.credentials import ConnectionStringCredentials
 from dlt.sources.sql_database import (
     sql_table as core_sql_table,
-    sql_database as core_sql_datbase,
+    sql_database as core_sql_database,
 )
 
 from .schema_types import _to_dlt_column_schema, _to_dlt_val
@@ -114,7 +114,8 @@ def init_replication(
         table_names (Optional[Union[str, Sequence[str]]]):  Name(s) of the table(s)
           to include in the publication. If not provided, the whole schema with `schema_name` will be replicated
           (also tables added to the schema after the publication was created). You need superuser privileges
-          for the schema replication.
+          for the whole schema replication. When specifying individual table names, the database role must
+          own the tables if it's not a superuser.
         credentials (ConnectionStringCredentials): Postgres database credentials.
         publish (str): Comma-separated string of DML operations. Can be used to
           control which changes are included in the publication. Allowed operations
@@ -184,7 +185,7 @@ def init_replication(
                     # do not include dlt tables
                     table_names = [
                         table_name
-                        for table_name in core_sql_datbase(
+                        for table_name in core_sql_database(
                             credentials, schema=schema_name, reflection_level="minimal"
                         ).resources.keys()
                         if not table_name.lower().startswith(DLT_NAME_PREFIX)