Release overview

This release brings a new algorithm for the SQL COPY statement [Reference], which is specifically designed to enable fast and easy CSV Import for large and unsorted data files. This feature comes in handy when there's a need to migrate a large amount of data from other databases to QuestDB [Guide: CSV import]. In addition, there are several improvements and fixes for the Influx line protocol, Core, and SQL engine.

Gotchas

No gotchas when you are upgrading from the start or 6.3, as it is just a software update. If you're upgrading from 6.2.1 or earlier, please refer to the gotchas of 6.3

What is new

CORE

• deterministically deposit hs_err_pid crash logs into db directory with Docker package by @bluestreak01 in https://github.com/questdb/questdb/pull/2346
• use all available cores by default @puzpuzpuz https://github.com/questdb/questdb/pull/2392

SQL

• high-performance implementation of the COPY statement for unordered CSV files by @eugenels, @bziobrowski in https://github.com/questdb/questdb/pull/2155
• implement the function substring to extract some characters from a string by @HeZean in https://github.com/questdb/questdb/pull/2251
• add write mode parameter for CREATE TABLE SQL by @ideoma in https://github.com/questdb/questdb/pull/2335
• improve PostgreSQL compatibility as a base for extending integrations later by @bluestreak01 https://github.com/questdb/questdb/pull/2210

What we fixed

CORE

• fix swallowing of all error and critical log messages by @ideoma in https://github.com/questdb/questdb/pull/2345
• fix docker entry point exiting prematurely on chown errors by @mariusgheorghies in https://github.com/questdb/questdb/pull/2354
• fix potential crash on adding column index by @ideoma in https://github.com/questdb/questdb/pull/2362

SQL

• incorrect result for negative LIMIT query with indexed symbol filter by @puzpuzpuz in https://github.com/questdb/questdb/pull/2336
• make aggregate check case-insensitive by @bziobrowski in https://github.com/questdb/questdb/pull/2331
• fix ORDER BY priority in case of duplicate column names by @jerrinot in https://github.com/questdb/questdb/pull/2339
• incorrect result in some aggregation functions by @eugenels in https://github.com/questdb/questdb/pull/2342
• fix incorrect timezone arithmetic in INSERT SQL by @bluestreak01 in https://github.com/questdb/questdb/pull/2366
• fix keyed group by crashing on int column with high cardinality @bziobrowski https://github.com/questdb/questdb/pull/2353
• fix error from queries that involve time interval & JIT search @bluestreak01 https://github.com/questdb/questdb/pull/2393
• fix boolean numeric representation @Zahlii https://github.com/questdb/questdb/pull/2386
• inconsistency in worker count configuration for asynchronous filters @eugenels https://github.com/questdb/questdb/pull/2399

ILP

• change default Sender port by @jerrinot in https://github.com/questdb/questdb/pull/2324

PGWire

• fix spurious error when executing "create table" SQL from Rust @bluestreak01 https://github.com/questdb/questdb/pull/2385

Full Changelog: https://github.com/questdb/questdb/compare/6.4.3...6.5