Skip site navigation (1)
Skip section navigation (2)
Todo List
Known Bugs
- [bugs] Deallocating large numbers of server side statements can break the
connection by filling network buffers. This is a very, very low probability
bug, but it is still possible. ref
→
Compliance
- [JDBC1] Implement Statement.setQueryTimeout. →
- [JDBC2] Sort DatabaseMetaData.getTypeInfo properly (by closest match). →
- [JDBC2] Implement SQLInput and SQLOutput to allow composite types to be used. →
- [JDBC3] Implement Statement.getGeneratedKeys. ref2 →
- [JDBC3] The JDBC 3 DatabaseMetaData methods sometimes return additional information.
Currently we only return JDBC 2 data for these methods. ref
→
- [JDBC3] Implement Clob write/position methods. →
Performance
- [] Add statement pooling to take advantage of server prepared statements. →
- [] Allow scrollable ResultSets to not fetch all results in one batch. →
- [] Allow refcursor ResultSets to not fetch all results in one batch. →
- [] Allow binary data transfers for all datatypes not just bytea. →
PG Extensions
- [] Allow configuration of GUC parameters via the Connection URL or Datasource.
The most obvious example of usefulness is search_path. ref
→
Other
- [test] Pass the JDBC CTS (Sun's test suite). →
- [code] Allow SSL to use client certificates. This can probably be done with
our existing SSLSocketFactory customization code, but it would be good to
provide an example or other wrapper so a non-expert can set it up.
ref1,
ref2 →
- [code] Currently the internal type cache is not schema aware. →
- [code] Need a much better lexer/parser than the ad hoc stuff in the driver.
ref2 →
Ideas
- [] Allow Blob/Clob to operate on bytea/text data. ref
→
- [] Allow getByte/getInt/... to work on boolean values ref
→
- [] Add a URL parameter to make the driver not force a rollback on error for
compatibility with other dbs. The driver can wrap each statement in a Savepoint.
ref →
- [] Combine DatabaseMetaData efforts with pl/java. ref
→
- [] ResultSetMetaData calls that run queries are cached on a per column basis, but
it seems likely that they're going to be called for all columns, so try to issue
one query per ResultSet, not per column. →
- [] Make PGConnection, PGStatement, ... extend java.sql.XXX ref
→
Documentation
- [] The PGResultSetMetaData interface is not mentioned. →
- [] Timestamp +/- Infinity values are not mentioned. →
- [] Async notifies are more async now. ref
→
Website
- [] Setup a cron job somewhere to build and deploy the sight on a daily
basis to keep API changes and translations up to date.
→
- [] Add a daily git snapshot build to make the latest updates available.
→