duckdb 0.9.2

0.9.2 Bugfix Release

on Python PyPI

This is a bug fix release for various issues discovered after we released 0.9.1. There are no new features, just bug fixes. Database files created by DuckDB v0.9.0 or v0.9.1 can be read by DuckDB v0.9.2 (i.e. v0.9.2 is backwards compatible with both v0.9.0 and 0.9.1 and vice versa).

What's Changed

• [Dev] Fix Wasm.yml by @carlopi in #9303
• [Dev] Fix Wasm.yml, hardcoding extension config to latest duckdb-wasm by @carlopi in #9307
• Fix R CMD check script by @Tmonster in #9326
• [Dev] Remove unused tools/wasm folder by @carlopi in #9328
• [ODBC] SQLColAttribute fix by @maiadegraaf in #9316
• Fix typo by @shreeve in #9313
• [Dev] Trigger R.yml only only if there are changes to R.yml itself by @carlopi in #9304
• Add thread limit to test by @lnkuiper in #9305
• fix typing_extensions ImportError in experimental spark api by @Mause in #9346
• HTTPFS: Move from HTTPException to base class IOException by @carlopi in #9351
• fix for System.Data.ODBC GetSchema() by @bucweat in #9336
• Removes some static global variables by @TinyTinni in #9310
• Bump Julia to 0.9.1 by @Mytherin in #9354
• [Optimizer] Fix transitive filter issue by @Tishj in #9337
• Julia - v0.9 by @Mytherin in #9359
• fix time_bucket_tz(day): don't truncate input to a day boundary in #9320
• Increase the minimum cmake version to 3.5 by @taniabogatsch in #9364
• [PyArrow] Fix bug in timestamp pushdown by @Tishj in #9377
• Satisfy GCC's LTO checks by @krlmlr in #9386
• New generate_query_graph tool. by @Tmonster in #9212
• Fix LIST aggregate prepare statement exception by @taniabogatsch in #9370
• [Python] Throw explicit error for misuse of execute by @Tishj in #9394
• update R CMD workflow to apply patches in .github/patches/duckdb-r by @Tmonster in #9412
• [UnionVerify] Properly deal with SelectionVectors by @Tishj in #9409
• [ART] Improve error message for zero bytes in BLOBs by @taniabogatsch in #9415
• Enable serialization of LogicalCopyToFile by @Maxxen in #9418
• Nits for storage and Python API comments by @szarnyasg in #9414
• Issue #9416: Windowed Peer Functions by @hawkfish in #9425
• [Python] Adjust relation.df() to output microsecond precision for DATE types. by @Tishj in #9362
• [Python] Support PEP 563 by @Tishj in #9385
• Fix #8185 - avoid infinite recursion in AddTransitiveFilters by @Mytherin in #9440
• Fix 9447: SIGSEGV when executing read_csv() query for struct with empty VARCHAR by @ryderblue in #9448
• Fuzzer #1294: Non-Constant NULL Format by @hawkfish in #9450
• GitHub Actions bot: Prune search space with '--search' by @szarnyasg in #9445
• Issue #9396: AsOf Inequality Optimisation by @hawkfish in #9449
• Add bot for the 'Needs Documentation' label by @szarnyasg in #9444
• Fix field ids in LogicalCopyToFile:Deserialize by @Maxxen in #9424
• [Export Database] Produce up-to-date query for ViewCatalogEntry by @Tishj in #9375
• include csv_rejects_table headers for amalgamation by @Maxxen in #9454
• feat(jdbc): result streaming by @Mause in #9437
• [Arrow] Add support for dictionary's in child arrays (i.e list of ENUM) by @Tishj in #9331
• Detect AppleClang in cmake, add defines for DUCKDB_MAJOR/MINOR/PATCH_VERSION by @Maxxen in #9457
• [Dev] Fix build for extension-upload-wasm to fix (avoiding manual copy) by @carlopi in #9467
• Fix #9459 - remove unused qualified name parsing by @Mytherin in #9472
• Add license to Swift client by @szarnyasg in #9446
• Issue #8937: Window Order Collation by @hawkfish in #9477
• adding System.Data.ODBC testing to Windows.yml CI. by @bucweat in #9372
• fix(jdbc): support decimal arrays by @Mause in #9489
• Needs documentation workflow: Use correct event by @szarnyasg in #9464
• Needs documentation label: Add PR number in backlink by @szarnyasg in #9491
• Fix 9399 by @lnkuiper in #9411
• Enable option to skip building duckdb shell in extension distribution CI by @Maxxen in #9483
• fix confusing variable name by @SkyFan2002 in #9503
• Add MySQL Scanner Aliases + Enable Autoloading by @Mytherin in #9505
• Fix #9498: Amalgamation to properly define DUCKDB_VERSION by @carlopi in #9500
• Issue #582: Greenland TimeZone Change by @hawkfish in #9508
• [Union] Fix issue with keyword/quoted UNION member names. by @Tishj in #9432
• Dev: Avoid adding name_extension for extensions with DONT_LINK by @carlopi in #9496
• Implement array-based JDBC ResultSet by @elefeint in #8972
• Fix #9360, fix #9466: grab a lock before creating directories to fix race condition on Windows in partitioned write by @Mytherin in #9473
• Internal #582: Remove utc_offset Dependency by @hawkfish in #9521
• [ART] Don't allow index creation in readonly mode by @taniabogatsch in #9527
• Remove redundant class by @ccfelius in #9511
• Hive partitioning: Fix preprocessing of CreateDirectories by @carlopi in #9535
• fix: check for IsRowId before accessing column name by @stephaniewang526 in #9542
• Commit drop of index memory by @taniabogatsch in #9540
• [Catalog] Fix locking issues + identify problem in MappingValue by @Tishj in #9523
• JDBC - Add append method for LocalDateTime by @Jens-H in #9435
• Use run_tests_one_by_one to fix CI by @Mytherin in #9570
• pin run-vcpkg action version by @samansmink in #9579
• Add rowsort to test_window_order_collate.test by @carlopi in #9574
• Fix exploding Delim Joins by @Tmonster in #9564
• HivePartition: Add also lock on partition_state by @carlopi in #9576
• Fix typo by @lnkuiper in #9587
• [C-API] fix duckdb_parameter_name declaration by @suketa in #9566
• Fix incorrect template specialization by @jhammer in #9529
• fix(jdbc): allow use of nested result values after close by @Mause in #9592
• Allow pausing pipeline in NextBatch call by @ywelsch in #9562
• Optimize performance of jaccard function by @ucasfl in #9560
• [Python] Add missing default values to stubs of aggregate functions. by @Tishj in #9595
• Fix symbol leakage check by @Mytherin in #9604
• Issue: #8867 Fix: remove unused variables for logical_root. by @Light-City in #8866
• CI: Add job for 'needs maintainer approval' PR tag by @szarnyasg in #8853
• Ci: Fix trigger of 'needs maintainer approval' job by @szarnyasg in #9610
• fix(jdbc): combine jar publish steps by @Mause in #9484
• Increment postgres_scanner version by @Mytherin in #9504
• Improve linear probing performance of GroupedAggregateHashTable by @lnkuiper in #9575
• Expose InterruptState in NextBatch by @ywelsch in #9606
• [Arrow] Support LargeString and LargeList by @Tishj in #9471
• Issue #9589: Prefer strict TIMESTAMPs by @hawkfish in #9600
• Fix issue with streaming query results by @samansmink in #9619
• bump extension versions as prep for 0.9.2 by @samansmink in #9621
• feat: duckdb_api and custom_user_agent options by @elefeint in #9603
• Revert "fix(jdbc): combine jar publish steps" by @Mause in #9620
• Read JSON bugfix by @lnkuiper in #9623
• Fix #9548 - Throw a more clear error when using parameters inside of the source of a top-level PIVOT statement by @Mytherin in #9626
• Add SQL:2016 listagg by @renevdzee in #9613
• Bump spatial by @Maxxen in #9633
• [ODBC] Reorganize the SQLColAttribute Tests by @maiadegraaf in #9580
• Fix #9584: Correctly bail-out on committing appends of 0 rows, which can happen in certain edge cases by @Mytherin in #9640
• Add log(b,x) function by @hannes in #9637
• allow dbgen to be interrupted by user by @elefeint in #9624
• Fix httpfs CI by @samansmink in #9649
• fix 9232 by @samansmink in #9647
• Shell: Set globalDb to NULL after closing database by @Flogex in #9655
• Fix #9308 - avoid pruning plan for ANTI join with filter that is always true by @Mytherin in #9654
• Serialize decimal quantiles by @carlopi in #9653
• Partial revert of #9624: test seemed to cause some memory leaks in CI by @carlopi in #9660

Full Changelog: v0.9.1...v0.9.2