v.22.11
Backward Incompatible Changes
1Bug Fix
32- Updated QueryNormalizer to Clone Alias AST for Issue #42452 Resolution
- Fix Race Condition for Table Backup in Lazy Databases
- Fix skip_unavailable_shards for s3Cluster table function
- Fix schema inference in s3Cluster and improve hdfsCluster
- Fix URL table engine retries and error handling
- Segmentation Fault Fix for DNS and c-ares Issues
- Fix LOGICAL_ERROR in PK Analysis for Monotonic Binary Functions
- Fix Key Analysis for Nullable Key Types
- Fix typo in setting name affecting schema inference cache usage
- Fix Set Creation with LowCardinality Data Type
- (U)Int128 and (U)Int256 Checks Added in PREWHERE
- Fix segmentation fault bug in functions parser
- Fix locking in truncate table operation
- Fix crash in web disks when file is missing
- Fix auth_type mapping in system.session_log to include SSL_CERTIFICATE
- Fix stack-use-after-return in Create User query parser under ASAN build
- Fix lowerUTF8/upperUTF8 for Symbols Across 16-Byte Boundaries
- Additional bound check added to LZ4 decompression for malformed input fix
- Fix rare hang issue on query cancellation
- Fix Hash Join Issues with Multiple Disjuncts
- Null Pointer Issue in Three Table Join SQL Query
- Fix Memory Sanitizer Report in Cluster Discovery
- Enhance DateTime Schema Inference for Empty Strings
- Fix NOT_FOUND_COLUMN_IN_BLOCK Error in ClickHouse Projection
- Fix ATTACH TABLE for DATETIME in PostgreSQL Engine
- Fix lambda parsing, closes #41848
- Fix Key Analysis for Nullable Keys in Hyperrectangles
- Fix buffer over-reads in aggregate function state deserialization
- Fix NULL Handling in 'if' Function for Nullable Arguments
- Fix Decimal Math Overflow in DateTime Parsing with Best Effort Algorithm
- Indent Field Miscalculation in Git-Import Tool
- Fixed Interval Type Behavior with Subquery and Casting
Build/Testing/Packaging Improvement
6Experimental Feature
2Improvement
24- Trivial queries with small LIMIT correctly estimate row counts to check thresholds
- Add support for interactive parameters in INSERT VALUES queries
- Added allow_readonly Field to system.table_functions for Readonly Mode Support
- system.asynchronous_metrics documentation added and cache disk metrics error fixed
- Throttling algorithm updated to token bucket
- Mask sensitive data in query logs and error messages
- Remove Covered Parts to Prevent Replication Delay
- Progress Display Enhancement in ClickHouse Client
- Add FixedString Input Support to Base64 Coding Functions
- Add bytes_on_disk and path columns to system.detached_parts
- Improve Automatic Structure Detection in Table Functions for ClickHouse
- Fix Progress Indication for INSERT FROM INFILE
- Refactor tokens function for adjustable max token return (default disabled)
- Allow Date32 Arguments for formatDateTime and FROM_UNIXTIME Functions
- Update tzdata: Mexico abolishes DST, Chihuahua adopts year-round UTC-6
- Add FailedAsyncInsertQuery Event Metric for Async Inserts
- Implement Default `read-in-order` Optimization in Query Plan
- Increase S3 Multipart Upload Part Size to Avoid 10,000 Parts Limit Errors
- Insufficient Disk Space Due to Expired Parts in Merge Task
- Add oss function and OSS table engine compatible with s3
- Improve Error Reporting for OS-Related Info in system.asynchronous_metrics Table
- Modify INFORMATION_SCHEMA for MySQL Compatibility in ClickHouse
- Add PowerBI Compatibility Functions for MySQL Protocol
- Better Usability for Dashboard Changes
New Feature
15- Adds Retry Support for INSERTs in ReplicatedMergeTree During ClickHouse Keeper Session Loss
- Add Hudi and DeltaLake Read-Only Table Engines for S3
- Add hudi and deltaLake table functions
- Support for Composite Time Intervals in ClickHouse
- Added ** glob support for recursive directory traversal in filesystem and S3
- Introduce s3_plain Disk Type for Write-Once-Read-Many Operations in MergeTree Table
- Added Row-Level Policies to system.query_log
- Add Commands for Creating Snapshots and Retrieving Raft Information in ClickHouse Keeper
- Add ascii function similar to Apache Spark in ClickHouse
- Add pmod function for non-negative modulo results
- Add function formatReadableDecimalSize
- Add randCanonical function for pseudo-random value generation in [0, 1)
- Add displayName function, closes #36770
- Add min_age_to_force_merge_on_partition_only setting for partition optimization
- Add Generic Implementation for Named Collections in ClickHouse