DataManagement

This page will be used to store codes, protocols, and procedures for use by the Data Managers.

Babase Technical Information

These links are primarly of interest to the Babase support staff.

• [:RankerProgram:Ranker Program] -- The ranker program is used to develop and input Baboon social ranks.

• [:BabaseVPN:The Babase VPN] -- The Babase Virtual Private Network is available to those people who have a Unix Login on Papio. It is primarly used to run the ranker program.

• [:PsionFormat:The Psion Data Format] -- The Psions are handheld data collection devices. They output their data in a custom format.

• [:BabaseChangelog:Babase 2.0 Changelog]

Programs

• [:ProgramLinks:Generic Programs]

• [:ProgramLinks:Babase's Bespoke Programs]

Protocols

• [:DukeDMP:Duke's Data Management Protocol] -- Data Maintenance and Update Protocols

• [:DukeIDEP:Duke's Interaction Data Entry Protocol] -- Specific to Grooming, MCE, and Agonisms

Wiki Administration

• BabaseWikiAdminGroup

• BabaseWikiEditorsGroup

• DatasetStatus

• DatasetStatusTable

Misc Links

• [:TabbyTest:Tabby's test page]

• SandboxPage -- Sandbox page for sharing code remotely

• MultiPartyInteractionDevelopment

• RankerDiffs

Adding a User to Babase

• Users can be added to babase, as well as a schema created, and permissions set, by completing the following steps using a SSH client.

PGPASSWORD='*********' babase-user-add USERNAME babase_readers

• The ********* here would be the babase_admin password
• The Username selected will become not only the babase username and temporary password but also the schema name.
• It is very important that the user immediately change their password using PPA's password change option.

Copying the Babase Schema

The babase-copy-babase-schema Unix program copies the entire babase schema, including data, table definitions, validation, and everything else, from one database to another. All existing data, table definitions, validation, etc is deleted from the babase schema in the target database. You must be logged in to papio and at the Unix prompt to run the program.

The following example copies the babase schema from the babase database to the babase_test database. The first database (babase) is the database to copy from, the second (babase_test) the database to copy into.

PGPASSWORD='*********' babase-copy-babase-schema babase babase_test

• The ********* here would be the babase_admin password

The next example copies the babase scheme from the babase_test database to the babase_database. The program will not copy into the babase database unless the -f (force) flag is given.

PGPASSWORD='*********' babase-copy-babase-schema -f babase_test babase

Backing up the Database or Schemas

• When Backing up the DB, the 'help' directory can offer some direction

[YOU@papio ~]$ pg_dump --help

Restoring the Database or Schemas

• If you are going to want to restore tables within the schema:

[YOU@papio ~]$ pg_dump --file YOURBACKUP.sql --host=localhost --format=p schema=fecal --user YOU babase

• after the backup file is created, it is useful to be able to look at the file to ensure you have the correct search path (the tables are getting put in the correct place ie. your schema) one way to do this is to view what the command is doing by having the output displayed in your unix window

[YOU@papio ~]$ pg_restore --host=localhost --file=/dev/stdout --user YOU YOURBACKUP.sql

• you can also save this information into a file, rather than having it just display in your unix window

[YOU@papio ~]$ pg_restore --host=localhost --file=YOURRESTORE.sql --user YOU YOURBACKUP.sql

• you can then just read the output within the window as well as having saved it for restore use

[YOU@papio ~]$ less YOURRESTORE.SQL

• the output on your screen will then resemble the text below:

{{{CREATE SCHEMA schemaname1;

SET search_path = schemaname1, pg_catalog;

SET default_tablespace = ;

SET default_with_oids = false;}}}

• If you desire for the data to be restored to a different location, you can use the emacs application to edit the text of YOURRESTORE.sql.

[YOU@papio ~]$ emacs -nw YOURRESTORE.sql

SET default_with_oids = false;}}}

• the PSQL terminal should then be used for the restore, at which point the tables from your schema will be created in the schema specified in the search path

[YOU@papio ~]$ psql --dbname=babase --user=YOU --host=localhost < YOURRESTORE.sql

• If you enter the previous command without the '< YOURRESTORE.sql', you will be prompted with "babase>" This is just another front end for babase, and any queries could also be performed here, just for future reference.