Doris Flink Connector

This document lists Doris Flink Connector release notes in reverse chronological order.

26.1.1

Source: Release Note 26.1.1

Bug Fixes

• Fixed an issue where batch sinks might freeze during prolonged operation when compression is enabled. #653

Thanks

• @addu390

26.1.0

Source: Release Note 26.1.0

Features and Improvements

• Enabled gzip compression by default for Stream Load. #648

Thanks

• @JNSimba

26.0.0

Source: Release Note 26.0.0

Features and Improvements

• Supported the Doris DECIMAL256 type. #628
• Supported nested type fields in ArrayData. #606
• Supported HTTP requests with UTF-8 charset encoding. #623
• Added configurable Thrift max message size. #593
• Optimized MongoDB CDC type inference to avoid unnecessary decimal conversion. #596
• Improved the error message when the FE host is invalid. #629
• Supported Flink 2.x. #636

Bug Fixes

• Fixed label handling during group commit retries. #635
• Fixed batch Stream Load handling for types that do not have line delimiters. #619
• Fixed incorrect DDL generation when a Doris field has an empty string as its default value. #600
• Fixed a typo in BackendClient that could affect Arrow Flight reads. #622

Thanks

• @QuakeWang
• @JNSimba
• @jqcc
• @potterhe
• @kwonder0926
• @gnehil
• @keytouch
• @AlexRiedler

25.1.0

Source: Release Note 25.1.0

Features and Improvements

• Automatically created new synchronization tables in sharding scenarios. #564
• Used Arrow Flight SQL as the default read path for Doris Source. #574
• Added integration and end-to-end test cases. #569 #577 #580
• Added compatibility for reading the IP type in Doris 2.1.3 and later. #576
• Added release scripts. #567
• Increased the retry interval in batch mode. #579
• Automatically created new tables for MongoDB CDC synchronization. #573
• Added usage examples. #589

Bug Fixes

• Fixed an issue where the Flink task could get stuck after the Stream Load thread exits abnormally. #578
• Fixed type conversion issues in PostgreSQL CDC synchronization. #582
• Fixed schema length mismatch for non-_id ObjectId fields in MongoDB CDC synchronization. #588

Thanks

• @Aalron
• @JNSimba
• @kwonder0926
• @vinlee19

25.0.0

Source: Release Note 25.0.0

Features and Improvements

• Supported reading catalog tables through Arrow Flight SQL. #530
• Supported INSERT OVERWRITE. #544
• Supported partial limit pushdown. #553
• Concatenated the doris.filter.query option during pushdown. #552
• Added compatibility for FE API changes by supporting configurable response fields to ignore. #549
• Optimized MongoDB CDC sampleSize calculation logic. #542
• Ignored failed MongoDB schema changes. #537
• Added prefixes for lookup queries and Arrow Flight queries. #531 #530

Bug Fixes

• Fixed an issue where write errors could bypass checkpoints in extreme cases. #560 #555
• Fixed an issue where writes could get stuck when an HTTP error occurs. #539
• Fixed parsing issues for MongoDB timestamp and array types. #547
• Fixed the ORA-12733 issue when many tables are synchronized. #532
• Fixed timestamp format pushdown errors. #528
• Fixed inaccurate parsing of SQL parser schema table comments and field types. #540
• Fixed the Transfer-Encoding header already present error. #525
• Fixed a null pointer exception in multi-database synchronization. #534

Thanks

• @awol2005ex
• @DongLiang-0
• @JNSimba
• @nativeCat
• @vinlee19
• @xuqinghuang