Include diagnostics in failed SARIF if they were created before a database was initialised

[mbg]

Diagnostics are typically written to the database cluster and are then included in the SARIF that is generated by the CLI. However, in the CodeQL Action we may produce diagnostics before the database cluster is initialised. To work around this, we have so far relied on tracking diagnostics in memory that were created before the database cluster is ready. After the cluster is initialised, those diagnostics are then flushed to disk. However, this represents a problem if the database cluster is never successfully initialised.

This PR changes this so that diagnostics are written directly to a temporary directory on disk if the database cluster is not yet ready. They are then moved to the database cluster once it is initialised. If the cluster is never initialised, then we point the CLI at the temporary directory on disk using a --diagnostic-dir argument. The CLI will then pick up any diagnostics that are still in that location and include them in the failed SARIF.

Risk assessment

For internal use only. Please select the risk level of this change:

• Low risk: Changes are fully under feature flags, or have been fully tested and validated in pre-production environments and are highly observable, or are documentation or test only.

Which use cases does this change impact?

Workflow types:

• Advanced setup - Impacts users who have custom CodeQL workflows.
• Managed - Impacts users with dynamic workflows (Default Setup, Code Quality, ...).

Products:

• Code Scanning - The changes impact analyses when analysis-kinds: code-scanning.
• Code Quality - The changes impact analyses when analysis-kinds: code-quality.
• Other first-party - The changes impact other first-party analyses.

Environments:

• Dotcom - Impacts CodeQL workflows on github.com and/or GitHub Enterprise Cloud with Data Residency.
• GHES - Impacts CodeQL workflows on GitHub Enterprise Server.

How did/will you validate this change?

• Unit tests - I am depending on unit test coverage (i.e. tests in .test.ts files).
• End-to-end tests - I am depending on PR checks (i.e. tests in pr-checks).

If something goes wrong after this change is released, what are the mitigation and rollback strategies?

• Rollback - Change can only be disabled by rolling back the release or releasing a new version with a fix.

How will you know if something goes wrong after this change is released?

• Telemetry - I rely on existing telemetry or have made changes to the telemetry.
  • Dashboards - I will watch relevant dashboards for issues after the release. Consider whether this requires this change to be released at a particular time rather than as part of a regular release.
  • Alerts - New or existing monitors will trip if something goes wrong with this change.

Are there any special considerations for merging or releasing this change?

• No special considerations - This change can be merged at any time.

Merge / deployment checklist

• Confirm this change is backwards compatible with existing workflows.
• Consider adding a changelog entry for this change.
• Confirm the readme and docs have been updated if necessary.

[Copilot]

Warning

• Copilot's review of this pull request may be incomplete because some of the changed files are excluded by your Copilot content exclusion settings. See Excluding content from Copilot for details.

🟡 Changes recommended

Temporary-write failures can abort initialization, cross-filesystem moves can lose diagnostics, and the new check does not validate failed SARIF export.

5 open findings

• Test does not verify failed-SARIF diagnostic export · New
• Best-effort diagnostics can make init fail on filesystem errors · New
• Cross-filesystem diagnostic moves lack an EXDEV fallback · New
• Spelling correction · New
• Documentation typo · New

What changed in this PR

Persists pre-database diagnostics on disk so they can be included in failed SARIF output.

Changes:

• Replaces in-memory buffering with temporary diagnostic files.
• Passes temporary diagnostics to CodeQL export.
• Adds unit and PR-check coverage.

File	Description
src/​diagnostics.ts	Implements temporary persistence and database relocation.
src/​init-action.ts	Flushes and logs temporary diagnostics.
src/​init-action-post-helper.ts	Supplies temporary diagnostics during failed SARIF generation.
src/​codeql.ts	Adds --diagnostic-dir support.
src/​diagnostics.test.ts	Tests temporary persistence and flushing.
src/​setup-codeql.test.ts	Provides temporary-directory test setup.
src/​config/​db-config.test.ts	Provides temporary-directory test setup.
pr-checks/​diagnostics-without-db.ts	Checks for the expected diagnostic file.
pr-checks/​checks/​diagnostics-without-db.yml	Defines the new PR check.
lib/​entry-points.js	Generated bundle; excluded from review.
.github/​workflows/​__diagnostics-without-db.yml	Generated workflow; excluded from review.

Files excluded by content exclusion policy (2)

• .github/workflows/__diagnostics-without-db.yml
• lib/entry-points.js

🧠 Review effort: Balanced

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

Warning

• Copilot's review of this pull request may be incomplete because some of the changed files are excluded by your Copilot content exclusion settings. See Excluding content from Copilot for details.

🟡 Changes recommended

Eager configuration loading can reject independently pinned upload steps even when no CLI merging is needed.

1 open finding

• Keep configuration lookup lazy for single-file SARIF uploads · New

5 resolved since last review

• Cross-filesystem diagnostic moves lack an EXDEV fallback
• Best-effort diagnostics can make init fail on filesystem errors
• Test does not verify failed-SARIF diagnostic export
• Documentation typo
• Spelling correction

Files excluded by content exclusion policy (3)

• .github/workflows/__diagnostics-export.yml
• .github/workflows/__diagnostics-without-db.yml
• lib/entry-points.js

🧠 Review effort: Balanced

[mbg]

getConfig rejects configurations written by a different Action version

This is true. There are two high-level scenarios:

• The rarer one is that there is a config from a previous step. It will be even rarer that the config was produced by a different version of the CodeQL Action, but if it is:
  • If there are multiple SARIF files, we now produce the error earlier than we would have previously. We therefore avoid work that would have been pointless if the error is raised later.
  • If there is only one, then combineSarifFilesUsingCLI never gets called. We produce an error when we wouldn't have previously.
• The more common one is that we don't have a config because upload-sarif is the only CodeQL Action step in a workflow job. No error is produced as before.

I have no strong feelings on the one case that matters here, because in principle we do not support mixed versions of the CodeQL Action in any case. If upload-sarif was working in a workflow job like that until now, then that was a happy accident. Equally, upload-sarif currently has no need for the config except when combineSarifFilesUsingCLI is called.

[Copilot]

Warning

• Copilot's review of this pull request may be incomplete because some of the changed files are excluded by your Copilot content exclusion settings. See Excluding content from Copilot for details.

🟢 Approval recommended

No blocking issues remain in the reviewable changes, which include targeted unit and end-to-end coverage.

1 open finding

• Keep configuration lookup lazy for single-file SARIF uploads

Files excluded by content exclusion policy (2)

• .github/workflows/__diagnostics-without-db.yml
• lib/entry-points.js

🧠 Review effort: Balanced