====================================
BioSQL
====================================

This is the BioSQL distribution. BioSQL is a generic unifying schema
for storing sequences from different sources, for instance Genbank or
Swissprot.

BioSQL is meant to be a common data storage layer supported by all the
different Bio* projects, Bioperl, Biojava, Biopython, and Bioruby.
Entries stored through an application written in, say, Bioperl could
be retrieved by another written in Biojava.

There are currently four different RDBMSs supported: MySQL,
PostgreSQL, Oracle, and most recently SQLite. The MySQL schema
DDL is in sql/biosqldb-mysql.sql, the PostgreSQL schema is in
sql/biosqldb-pg.sql, and the Oracle schema is in multiple files
in the directory sql/biosql-ora, and the SQLite schema is in
sql/biosqldb-sqlite.sql. In order to instantiate the schema, feed
the respective file or files to your SQL shell (e.g. mysql for
MySQL, and psql for PostgreSQL).

Note the SQLite schema is new, and is not yet supported by all the
Bio* projects (at the time of writing, just Biopython supports this).

======================================================================
INSTALLATION NOTES

Generally, the support for each RDBMS is maintained individually,
as there are too many differences between the RDBMSs for automatic
generation from a single DDL. Please see the INSTALL for details.

======================================================================
POST v1.0 TODO
======================================================================

The following changes are scheduled for the next release of
BioSQL. These changes incur incompatibilities with the 1.0 version and
will require a database migration.

  - Rename comment table to anncomment and rename term_synonym.synonym
  to term_synonym.name in mysql and pg schemas. This will make the
  names consistent with the Oracle schema, avoiding Oracle reserved
  words.