feat(ai): OpenAI retry parity — typed error + tests (closes #69) by Kavirubc · Pull Request #99 · similigh/simili-bot

Kavirubc · 2026-03-05T07:39:24Z

Summary

Bug fix: isRetryableError previously never retried OpenAI calls because errors from callOpenAIJSON were plain fmt.Errorf strings with no typed carrier. Introduced *openAIStatusError (carries HTTP status code) so errors.As can detect 429/5xx and retry them identically to Gemini.
Testability: Added baseURL string field to both Embedder and LLMClient (empty = production) and a matching parameter to callOpenAIJSON, allowing tests to inject an httptest.Server URL without any mocking framework.
Tests: 8 new tests in openai_retry_test.go covering retry-on-429, retry-on-500, no-retry-on-400, and retry exhaustion for both embedOpenAI and generateOpenAIText.

Files changed

File	Change
`internal/integrations/ai/retry.go`	Add `openAIStatusError`; update `isRetryableError`
`internal/integrations/ai/provider.go`	Return `*openAIStatusError`; add `baseURL` param to `callOpenAIJSON`
`internal/integrations/ai/embedder.go`	Add `baseURL` field; pass to `callOpenAIJSON`
`internal/integrations/ai/llm.go`	Add `baseURL` field; pass to `callOpenAIJSON`
`internal/integrations/ai/openai_retry_test.go`	New — all 8 retry tests

Test plan

go test ./internal/integrations/ai/... -run TestEmbedOpenAI -v — all 4 pass
go test ./internal/integrations/ai/... -run TestGenerateOpenAIText -v — all 4 pass
go test ./... — full suite passes (12/12 packages)
go build ./... — no compilation errors
go vet ./... — no issues

coderabbitai · 2026-03-05T07:39:41Z

Warning

Rate limit exceeded

@Kavirubc has exceeded the limit for the number of commits that can be reviewed per hour. Please wait 5 minutes and 23 seconds before requesting another review.

⌛ How to resolve this issue?

After the wait time has elapsed, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout.

Please see our FAQ for further information.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 18090923-a445-4c11-b593-4c4971892b06

📥 Commits

Reviewing files that changed from the base of the PR and between 3fb471a and 8ec0698.

📒 Files selected for processing (5)

internal/integrations/ai/embedder.go
internal/integrations/ai/llm.go
internal/integrations/ai/openai_retry_test.go
internal/integrations/ai/provider.go
internal/integrations/ai/retry.go

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch feature/openai-retry-parity-issue-69

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

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

…retry parity - Introduce *openAIStatusError in retry.go so isRetryableError recognises OpenAI 429/5xx responses and retries them (previously unretried) - Add baseURL param to callOpenAIJSON (empty = production openAIBaseURL) - Add unexported baseURL field to Embedder and LLMClient for test injection Signed-off-by: Kavirubc <hapuarachchikaviru@gmail.com>

8 httptest-based tests covering 429/5xx retry, non-retryable 400, and retry exhaustion for both embedOpenAI and generateOpenAIText. Signed-off-by: Kavirubc <hapuarachchikaviru@gmail.com>

gh-simili-bot · 2026-03-05T07:40:48Z

Simili Triage Report

Note

Quality Score: 9.5/10 (Excellent)
The issue could be improved. See suggestions below.

Classification

Category	Value
Labels

Quality Improvements

Reproduction steps are implicit (as it's a code-level fix, not a user-facing bug report)
No direct link to the original issue feat: add retry logic for OpenAI API calls (parity with Gemini) #69 within the body (though title links it)
Consider adding a direct link to the original issue feat: add retry logic for OpenAI API calls (parity with Gemini) #69 in the body for easier navigation.

Similar Threads

Similarity	Type	Thread	Status
86%	🔀	#66 fix: harden retry logic - typed error checking ...	Open
83%	📝	#69 feat: add retry logic for OpenAI API calls (par...	Open
79%	📝	#50 [Reliability]: Implement Exponential Backoff fo...	—

Warning

Possible Duplicate (Confidence: 98%)
This pull request might be a duplicate of #69.
Reason: The current issue's title explicitly states 'closes #69'. Issue #69's title 'feat: add retry logic for OpenAI API calls (parity with Gemini)' perfectly matches the summary of the current issue, which describes implementing typed error handling and tests to enable OpenAI retry parity with Gemini. They describe the exact same feature/bug fix, and fixing the current issue fully resolves #69.

⏳ This pull request will be automatically closed in 72 hours if no objections are raised. If you believe this is not a duplicate, please leave a comment explaining why.

_{Generated by Simili Bot}

gh-simili-bot · 2026-03-05T07:48:03Z

🧪 E2E Test

✅ Bot responded: yes

| Auto-closer (dry-run) | processed: 0 closed: 0 grace: 0 human: 0 |

Test repo → gh-simili-bot/simili-e2e-22707564014
Run → logs

_{Auto-generated by E2E pipeline}

Kavirubc added 2 commits March 5, 2026 13:09

test(ai): add OpenAI retry parity tests (issue #69)

8ec0698

8 httptest-based tests covering 429/5xx retry, non-retryable 400, and retry exhaustion for both embedOpenAI and generateOpenAIText. Signed-off-by: Kavirubc <hapuarachchikaviru@gmail.com>

Kavirubc force-pushed the feature/openai-retry-parity-issue-69 branch from 3156ea6 to 8ec0698 Compare March 5, 2026 07:40

gh-simili-bot added bug Something isn't working enhancement New feature or request tests labels Mar 5, 2026

github-project-automation Bot added this to simili-bot-v1-release Mar 5, 2026

github-project-automation Bot moved this to Todo in simili-bot-v1-release Mar 5, 2026

gh-simili-bot added the e2e label Mar 5, 2026

gh-simili-bot approved these changes Mar 5, 2026

View reviewed changes

gh-simili-bot merged commit 93e8d90 into main Mar 5, 2026
9 checks passed

gh-simili-bot deleted the feature/openai-retry-parity-issue-69 branch March 5, 2026 07:48

github-project-automation Bot moved this from Todo to Done in simili-bot-v1-release Mar 5, 2026

gh-simili-bot mentioned this pull request Mar 11, 2026

perf: parallelize EmbedBatch using goroutines and errgroup #107

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(ai): OpenAI retry parity — typed error + tests (closes #69)#99

feat(ai): OpenAI retry parity — typed error + tests (closes #69)#99
gh-simili-bot merged 2 commits intomainfrom
feature/openai-retry-parity-issue-69

Kavirubc commented Mar 5, 2026

Uh oh!

coderabbitai Bot commented Mar 5, 2026 •

edited

Loading

Rate limit exceeded

Uh oh!

gh-simili-bot commented Mar 5, 2026

Uh oh!

gh-simili-bot commented Mar 5, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

Kavirubc commented Mar 5, 2026

Summary

Files changed

Test plan

Uh oh!

coderabbitai Bot commented Mar 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Rate limit exceeded

Uh oh!

gh-simili-bot commented Mar 5, 2026

Simili Triage Report

Classification

Uh oh!

gh-simili-bot commented Mar 5, 2026

🧪 E2E Test

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

coderabbitai Bot commented Mar 5, 2026 •

edited

Loading