fix(LaunchDarkly): Import processing not idempotent for completed requests

[zain-asif-dev]

• I have read the Contributing Guide.
• I have added information to docs/ if required so people know about the feature.
• I have filled in the "Changes" section below.
• I have filled in the "How did you test this code" section below.

Changes

Closes #7950

The Sentry trace attached to the issue shows process_import_request crashing with:

django.core.signing.BadSignature: No ":" found in value

while unsigning import_request.ld_token. This isn't actually about the create endpoint accepting an empty token: CreateLaunchDarklyImportRequestSerializer.token is a plain CharField, and allow_blank defaults to False, so the create endpoint already rejects a blank token with a 400 (verified locally, see test plan below).

The real issue is in process_import_request itself: _complete_import_request clears import_request.ld_token to "" once the request finishes (success or failure), by design, so the token isn't kept around longer than needed. If process_import_request runs again for that same, already-completed request (for example a stale/duplicate task-processor delivery, which the task's timeout and failure-retry behavior make possible), it tries to unsign that now-blank token and blows up with the exact BadSignature error from the Sentry trace, rather than being a harmless no-op.

This adds a guard at the top of process_import_request that returns early (with a log message) if import_request.completed_at is already set, making reprocessing idempotent instead of crashing.

How did you test this code?

Added test_process_import_request__already_completed__does_not_reprocess, which marks an import request as completed with a blank ld_token (mirroring what _complete_import_request leaves behind) and asserts that calling process_import_request on it again is a no-op (the LaunchDarkly client is never re-instantiated). I confirmed this test reproduces the exact BadSignature: No ":" found in value crash from the Sentry trace when the guard is removed, and passes with the fix.

Also ran the full tests/unit/integrations/launch_darkly/ suite (47 passed), ruff check/ruff format --check, and mypy on the changed files.

[vercel]

@zain-asif-dev is attempting to deploy a commit to the Flagsmith Team on Vercel.

A member of the Team first needs to authorize it.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 9d976da5-046c-4fa0-8c11-338e9e96a4c5

📥 Commits

Reviewing files that changed from the base of the PR and between 50986a6 and cfd0c07.

📒 Files selected for processing (2)

• api/integrations/launch_darkly/services.py
• api/tests/unit/integrations/launch_darkly/test_services.py

💤 Files with no reviewable changes (2)

• api/integrations/launch_darkly/services.py
• api/tests/unit/integrations/launch_darkly/test_services.py

---

📝 Walkthrough

Walkthrough

process_import_request now returns immediately when a LaunchDarklyImportRequest already has completed_at set, logging a warning instead of continuing. The test suite adds timezone-aware setup and a regression test that re-calls the function on an already completed request and checks that the LaunchDarkly client is not created and environments are not fetched.

Estimated code review effort: 2 (Simple) | ~10 minutes

Related PRs: None identified.

Suggested labels: bug, backend, tests

Suggested reviewers: None identified.

🐰 A guard now stands where imports once looped,
Completed requests are quietly scooped,
No token, no fetch, no client anew—
Just a warning logged, and the work is through.

---

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: b6cfc865-b39a-4f55-8716-58da1e75f887

📥 Commits

Reviewing files that changed from the base of the PR and between da194ca and 50986a6.

📒 Files selected for processing (2)

• api/integrations/launch_darkly/services.py
• api/tests/unit/integrations/launch_darkly/test_services.py

[khvn26]

just a couple edits away from merging this 👍

[zain-asif-dev]

Removed those comments.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.49%. Comparing base (da194ca) to head (cfd0c07).
⚠️ Report is 4 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #7954      +/-   ##
==========================================
- Coverage   98.63%   98.49%   -0.15%     
==========================================
  Files        1497     1497              
  Lines       59189    59198       +9     
==========================================
- Hits        58383    58305      -78     
- Misses        806      893      +87     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.