Обсуждение: public synonym

You can infer the context by first setting the search_path variable.  You can set it initially in your connection or do it for a database context or even a role context
SET search_path = MASTER_USER, public, pg_catalog;
ALTER DATABASE whatever  SET search_path = MASTER_USER, public, pg_catalog;
ALTER ROLE whoever SET search_path = MASTER_USER, public, pg_catalog;

Then you can continue to let the tables be non-qualified.   bye bye synonyms!

Regards
Michael

Eugene Yin

Thursday, January 7, 2016 3:17 PM

PostgreSQL ver 9.4.5.  Linux OS.

Application:  Web Based

Platform:  

    App Server (java) --> jdbc call --> Database Server (PostgreSQL)

I do know that PostgreSQL does not support the public synonym.  Now, for a user schema (let's call it MASTER_USER), if I coded in the stored function/procedure like the following, without using a public synonym to identify the table name: 

select user_name from user_info_table

then access it from the Java (web app) side via the JDBC call to the database, will that work?  

OR, 

I must use the identifier inside the sql, such as:

 select user_name from  MASTER_USER.user_info_table?

I come from the Oracle world, there I first create the public synonym for the table, then in the stored procedure I just directly reference the table with no need to identify the table with a schema name.  Like to know how it work under the PostgreSQL.

  

Thanks

Eugene

On 01/07/2016 12:27 PM, michael@sqlexec.com wrote:
> You can infer the context by first setting the search_path variable.
> You can set it initially in your connection or do it for a database
> context or even a role context
> SET search_path = MASTER_USER, public, pg_catalog;
> ALTER DATABASE whatever  SET search_path = MASTER_USER, public, pg_catalog;
> ALTER ROLE whoever SET search_path = MASTER_USER, public, pg_catalog;

Just be aware that Postgres folds identifiers to lower case:

test=# create schema MASTER_USER;
CREATE SCHEMA
test=# \dn    List of schemas    Name     |  Owner
-------------+---------- master_user | postgres public      | postgres

test=# create table MASTER_USER.test_tbl(id int);
CREATE TABLE

test=# \d master_user.test_tbl Table "master_user.test_tbl" Column |  Type   | Modifiers
--------+---------+----------- id     | integer |

UNLESS the tool you are using to create objects quotes the object name:

test=# create schema "MASTER_USER";
CREATE SCHEMA

test=# create table MASTER_USER.test_tbl(id int);
ERROR:  schema "master_user" does not exist

test=# create table "MASTER_USER".test_tbl(id int);
CREATE TABLE

test=# \d "MASTER_USER".test_tbl Table "MASTER_USER.test_tbl" Column |  Type   | Modifiers
--------+---------+----------- id     | integer |


More details at:

http://www.postgresql.org/docs/9.4/static/sql-syntax-lexical.html

4.1.1. Identifiers and Key Words





>
> Then you can continue to let the tables be non-qualified.   bye bye
> synonyms!
>
> Regards
> Michael
>> Eugene Yin <mailto:eugeneymail@ymail.com>
>> Thursday, January 7, 2016 3:17 PM
>> PostgreSQL ver 9.4.5.  Linux OS.
>> Application:  Web Based
>>
>> Platform:
>>     App Server (java) --> jdbc call --> Database Server (PostgreSQL)
>>
>>
>> I do know that PostgreSQL does not support the public synonym.Now, for
>> a user schema (let's call it MASTER_USER), if I coded in the stored
>> function/procedure like the following, without using a public synonym
>> to identify the table name:
>>
>> select user_name from user_info_table
>>
>> then access it from the Java (web app) sidevia the JDBC call to the
>> database, will that work?
>>
>> OR,
>>
>> I must use the identifier inside the sql, such as:
>>
>>  select user_name fromMASTER_USER.user_info_table?
>>
>>
>> I come from the Oracle world, there I first create the public synonym
>> for the table, then in the stored procedure I just directly reference
>> the table with no need to identify the table with a schema name.  Like
>> to know how it work under the PostgreSQL.
>>
>>
>> Thanks
>>
>> Eugene
>


-- 
Adrian Klaver
adrian.klaver@aklaver.com