From: Jonathan C. <cra...@sn...> - 2002-07-26 03:36:42
|
I've started to work on improving the scripts for dumping and loading the GUSdev and GUS 3.0 schemas. Since I'll be out of the office until around mid-day Monday I've placed a snapshot of what I've done so far on the CBIL download site: http://www.cbil.upenn.edu/downloads/GUS/schema/ The relevant file is 'gusdev-sql-jul-25-2002.tar.gz' and it contains up-to-date load scripts for the sequences, tables, views, constraints, and indexes in GUSdev. The major improvement is that the scripts now all prompt for the 'schema', i.e. the name of the user under whom the objects are to be created, and also the names of the tablespaces where the tables and indexes should be placed. As noted in the README file, I have tested the scripts to create the sequences, tables, and views, but not those for the constraints and indexes. When I get back on Monday I'll work on adding the following: -script to create the owning schema (not as important for gusdev since there is only one) -scripts/utility to automate the process of populating the essential system tables (i.e. TableInfo, UserInfo, Algorithm, etc.) -scripts to automate granting permissions on the various objects to users other than the owning schema I'm also trying to make sure that all the changes I make will carry over to GUS 3.0, and in general I want to make sure that people implementing the database can choose whatever names they want for the schemas involved. Jonathan |