Releases · microsoft/mssql-jdbc
[13.2.0] Stable Release ChangedFix trustStoreType JVM property consultation in SSL Connections #2724
What was fixed: Ensured fallback to JVM system property javax.net.ssl.trustStoreType if connection property is unset.
Who benefits: Users configuring SSL via system properties.
Impact: Enables proper SSL trust store resolution, improving compatibility with system configurations.
Addressed a data integrity and performance issue for bulk copy batch inserts with non-Unicode Strings #2735
What was fixed: Resolved performance degradation and incorrect string handling in bulk copy batch inserts when useBulkCopyForBatchInsert=true and sendStringParametersAsUnicode=false.
Who benefits: Applications using SQL Server JDBC Bulk Copy for high-volume inserts, especially with non-Unicode or accented character data.
Impact: Restores expected string storage, improves bulk insert performance for bulk copy API.
Handle SQL comments at start of query in getGenerateKeys() method #2731
What was fixed: Corrected detection of INSERT statements when preceded by SQL comments.
Who benefits: Developers using getGeneratedKeys() after commented SQL inserts.
Impact: Fixes previously failing key retrieval when comments are present before the query.
Fix update count handling for multi-statement queries executed via PreparedStatement execute method. #2737
What was fixed: Corrected update count handling for multi-statement queries run using PreparedStatement execute method, ensuring accurate counts for INSERT and other DML operations.
Who benefits: Java applications using SQL Server JDBC with multi-statement PreparedStatement executions.
Impact: Eliminates inconsistent update counts between Statement and PreparedStatement, improving reliability of result processing logic.
CodeQL suppression fixes #2728
What was fixed: Suppressed CodeQL rule [SM05141] with justification for broader usage scenarios.
Who benefits: Developers maintaining code quality tools like CodeQL.
Impact: Prevents unnecessary alerts for valid usage in multi-environment libraries.
JSON datatype support #2558
What was added: Support for reading and writing JSON columns in SQL Server.
Who benefits: Developers working with semi-structured data in SQL Server.
Impact: Enhances application flexibility by natively handling JSON content, reducing need for manual parsing.
Add order hints for Bulk Copy operations #2701
What was added: Support for specifying order hints during Bulk Copy.
Who benefits: Data engineers and DBAs managing large data migrations or ETL jobs.
Impact: Improves bulk data load performance.
Coding best practices and review process #2666
What was added: Introduced contributor guidelines, coding best practices, and review processes.
Who benefits: Open-source contributors and maintainers of the mssql-jdbc project.
Impact: Improves code quality, consistency, and onboarding experience for new contributors.
Add new trusted AKV URLs for FR and DE #2708
What was added: Registered four new Azure Key Vault and Managed HSM endpoints for France and Germany.
Who benefits: Customers in regulated regions (France, Germany) using AKV for encryption.
Impact: Enables secure key operations via region-specific trusted endpoints.
Fix for null handling in temporal types with bulk copy #2702
What was fixed: Properly handle null values for temporal types when sendTemporalDataTypesAsStringForBulkCopy=false.
Who benefits: Developers using batch insert with native temporal types in bulk copy.
Impact: Prevents failures during bulk inserts, improving reliability of time-sensitive data ingestion.
Fix string insertion with bulk copy API when sendStringParametersAsUnicode=false #2704
What was fixed: Resolved issue where strings were inserted as byte arrays in batch bulk copy mode when sendStringParametersAsUnicode is set to false.
Who benefits: Developers using non-Unicode string inserts in performance-sensitive batch operations.
Impact: Ensures string integrity during batch inserts, eliminating silent data corruption.
Vector datatype support #2634
What was added: Native support for SQL Server’s new VECTOR data type with APIs for inserts, selects, stored procedures, and bulk copy.
Who benefits: Developers building AI/ML workloads or semantic search applications using vector data in SQL Server.
Impact: Enables native vector storage and access via JDBC, eliminating the performance overhead of managing vector data as strings.
New connection options, quotedIdentifier and concatNullYieldsNull #2618
What was added: New connection options, quotedIdentifier and concatNullYieldsNull, to control QUOTED_IDENTIFIER and CONCAT_NULL_YIELDS_NULL session settings for both new and pooled connections.
Who benefits: Developers needing Sybase-default behavior for their applications.
Impact: Reduces Sybase migration effort by allowing control of these session values.
Support for temporal and money datatypes when using bulk copy for batch insert operations #2670
What was added: Support for batch inserts of DATETIME, DATE, MONEY, etc., when using the useBulkCopyForBatchInsert option.
Who benefits: High-volume data insert users.
Impact: Improves performance for previously unsupported types (except in Azure Synapse Analytics (formerly Azure SQL DW).
Mockito integration into JDBC driver tests #2644
What was added: Mockito added as a test dependency.
Who benefits: Developers and contributors writing unit tests.
Impact: Enables better control and test coverage.
releaseSavepoint exception type standardized to SQLFeatureNotSupported #2583
What changed: SQLServerException replaced with SQLFeatureNotSupportedException to comply with JDBC specification.
Who benefits: Developers expecting JDBC standard exceptions.
Impact: Enables better exception handling in client code.
ActiveDirectoryPassword authentication deprecated #2624
What changed: Added deprecation warning for Microsoft Entra ID password authentication method.
Who benefits: Microsoft Entra ID authentication users.
Impact: Encourages migration to more secure methods.
Include Columnstore indexes in getIndexInfo() #2598
What changed: Replaced sp_statistics with a custom query to support all index types.
Who benefits: Developers using metadata APIs.
Impact: More accurate index metadata.
Corrected schema filtering in getSchemas() #2643
What changed: Ensured schemas are properly filtered when a catalog name is provided.
Who benefits: Users of metadata functions.
Impact: Complies with JDBC specification and avoids confusion.
Increased redirection limit #2659
What changed: Raised maximum redirection hops from 1 to 10.
Who benefits: Azure SQL Fabric users.
Impact: Enables successful multi-hop connection scenarios.
Session recovery with Entra ID authentication and redirect mode #2668
What was fixed: Redirect information is now followed during session recovery when using Entra ID authentication.
Who benefits: Azure SQL DB users using Entra ID and connection resiliency.
Impact: Prevents connection recovery failures.
Javadoc build warnings #2640
What was fixed: Cleaned up invalid Javadoc syntax.
Who benefits: Developers building the driver.
Impact: Cleaner builds, better docs.
OffsetDateTime formatting in SQLServerDataTable #2652
What was fixed: The OffsetDateTime toString() method was omitting seconds when seconds are zero, which resulted in the server rejecting those records. Used DateTimeFormatter to avoid this invalid format.
Who benefits: Users inserting OffsetDateTime values.
Impact: Prevents conversion errors.
String comparison in SQLServerDataTable.equals() #2653
What was fixed: Replaced == with .equals() for strings.
Who benefits: Anyone comparing data tables.
Impact: Accurate comparison behavior.
PreparedStatement metadata caching for encrypted columns #2663
What was fixed: Fixed loss of type metadata across PreparedStatement reuse.
Who benefits: Users of Always Encrypted with secure enclaves.
Impact: Prevents insert failures in batch encryption.
Make IBM security module optional #2636
What was fixed: IBM security dependency is now optional.
Who benefits: Applications not using IBM modules.
Impact: Avoids classloading issues.
Invalidate enclave session on reconnect #2638
What was fixed: Enclave session cache no longer reused across reconnects.
Who benefits: Users of Always Encrypted with secure enclaves in failover configurations.
Impact: Prevents internal enclave errors due to invalid enclave session references after failover.
File path error handling in ConfigurableRetryLogic #2650
What was fixed: Robust error handling for missing/unreadable retry config files.
Who benefits: All users.
Impact: Prevents driver errors due to application file path issues.
Suppressed CodeQL warnings for crypto usages #2677
What was fixed: Suppressed CodeQL security warnings for cryptographic use cases.
Who benefits: Reviewers using CodeQL.
Impact: Maintains compatibility and audit clarity.
Batch insert fix for case-sensitive column name mismatch #2695
What was fixed: executeBatch() now respects case sensitivity of collations when matching column names.
Who benefits: Applications using case-sensitive/case-insensitive schemas.
Impact: Avoids metadata retrieval failures during batch inserts.
mssql-jdbc.properties location in test environments #2579connectRetryCount is set to a value greater than 1 #2513You can’t perform that action at this time.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4