Code Health Trend report

As you manage the health of your code base, it is crucial to actively clean up existing issues and prevent new issues from making it to your default branch. This new report makes it easy to understand how many net new code health issues are being introduced in your code base.

Go to Reports → Code Health Trend in the repository view. You can also see the report across all active repositories from the team overview.

Issues Prevented report

We've added a new report to help you visualize the impact of having DeepSource as part of your software development workflow. The Issues Prevented report shows you the total number of code health issues you've prevented in pull requests, so they didn't make it into your code base.

This report is available for each repository as well as your entire team.

New in Analyzers

• 39 new checks:
• Java: JAVA-C1001, JAVA-P1007, JAVA-W1057, JAVA-W1058, JAVA-W1059, JAVA-E1081, JAVA-W1060
• C#: CS-R1093, CS-R1094, CS-R1095, CS-R1096, CS-R1097, CS-W1066, CS-W1067, CS-W1068, and CS-W1069
• Rust: RS-W1075, RS-W1076, RS-W1077, RS-W1101
• Go: GO-R1003, GO-R1004, GO-W1022, GO-W1023, GO-W1024
• Terraform / Azure: TF-S1024, TF-S1025, TF-S1026, TF-S1027, TF-S1028, TF-S1029, TF-S1030
• Terraform / GCP: TF-S2020, TF-S2021, TF-S2022, TF-S2023, TF-S2024, TF-S2025, TF-S2026
• Some Rails-specific checks in the Ruby Analyzer were temporarily disabled and are now available again: RB-A1001, RB-A1002, RB-A1003, RB-A1004, RB-A1005, RB-A1006, RB-A1007, RB-A1008, RB-A1009, RB-A1010, RB-A1011
• Rust: Support for new syntax introduced in stable version 1.65. Read more on Discuss.
• JavaScript: Fixed a bug where .eslintrc breaks if there is a syntax error in the package.json. This is now handled gracefully and will not break the analysis.

Fixes and improvements

• API: The Repository object now includes the list of active Analyzers. Learn more in the docs.
• A few users had reported that the DeepSource widget on Bitbucket pull requests is too prominent and noisy when the analysis isn't active. We've made it muted and non-obtrusive.
• We've fixed a rounding error leading to metric trends being shown as 0.0% on the analysis run's page.