Not later than this week a bug regarding [pg_dump] (https://www.postgresql.org/docs/devel/static/app-pgdump.html) and compression with zlib when dumping data has been reported [here] (https://www.postgresql.org/message-id/20150217153446.2590.24945@wrigleys.postgresql.org).

As a preparation of an upcoming patch for full-page write compression in WAL, a patch has been pushed this week to make PGLZ, the in-core compression algorithm of PostgreSQL used for TOAST tables, more pluggable for plugins and frontend applications, particularly pg_xlogdump that needs to be able to decode a WAL record using the XLOG reader facility even if blocks are compressed to be able to reconstitute them. It makes as well sense to expose this algorithm as PGLZ compressed data would not be limited only to the internal backend usage of TOAST tables in a PostgreSQL instance, but as well to WAL data, be it simple WAL archive or a WAL streaming flow. So, here is the commit:

A couple of days ago the following commit has popped up in PostgreSQL tree for the upcoming 9.5, introducing a feature particularly interesting for developers of backend extensions and plugins:

A couple of days back a new mode has been added in [vacuumdb] (https://www.postgresql.org/docs/devel/static/app-vacuumdb.html) for the support of parallel jobs:

Many improvements are being done in the ODBC driver for PostgreSQL these days, one being for example the addition of more and more integrated regression tests insuring the code quality. One feature particularly interested has been committed these days and consists of the following commit:

Today, here is an explanation of one of the largest patches that has hit PostgreSQL 9.5 illustrated by this commit:

Foreign tables are getting step-by-step closer to the features that are present in normal relations with the addition of support for CHECK constraints.

PostgreSQL 9.5 has added a new mode for [REINDEX] (https://www.postgresql.org/docs/devel/static/sql-reindex.html) with this commit:

PostgreSQL offers many ways to define how a node reaches the end of recovery with many parameters related to [recovery target] (https://www.postgresql.org/docs/devel/static/recovery-target-settings.html), like a timestamp, a transaction ID, a target name and recovery timeline, using recovery_target_timeline. A parameter called pause_at_recovery_target that exists since 9.1 allows as well to put a standby in pause when the recovery target is reached. The upcoming release 9.5 has made some improvements in this area with the following commit:

The extension module [pageinspect] (https://www.postgresql.org/docs/devel/static/pageinspect.html) has been already dumped to version 1.3 in PostgreSQL 9.5 with the addition of functions for [BRIN indexes] (/postgresql-2/postgres-9-5-feature-highlight-brin-indexes/). A couple of days back a new set of functions has been added for GIN indexes with this commit.