fix(openai): support reasoning effort#101
Merged
Merged
Conversation
galzilber
changed the title
Contributor
📝 WalkthroughWalkthroughAdds an optional ChangesReasoning Effort Configuration
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Review rate limit: 4/5 reviews remaining, refill in 12 minutes. Comment |
Contributor
🔒 Container Vulnerability Scan (hub-migrations - amd64)Click to expand results |
Contributor
🔒 Container Vulnerability Scan (hub-migrations - arm64)Click to expand results |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@src/providers/azure/provider.rs`:
- Around line 27-30: The Azure path gives precedence to base.reasoning_effort
when constructing reasoning_effort (variable reasoning_effort using
base.reasoning_effort and base.reasoning), but the legacy validation still
checks payload.reasoning first and can reject a request even when
base.reasoning_effort is present; update the Azure validation logic so it
mirrors construction precedence: if base.reasoning_effort (or reasoning_effort
after construction) is present, skip/avoid validating legacy payload.reasoning,
otherwise validate payload.reasoning as before. Locate and change the validation
block that references payload.reasoning in the Azure provider to only run when
base.reasoning_effort is None (or when reasoning_effort is None), keeping the
variable names reasoning_effort, base.reasoning_effort, base.reasoning, and
payload.reasoning as references.
In `@src/providers/openai/provider.rs`:
- Around line 26-30: The validation path currently validates legacy
payload.reasoning before the conversion fallback, causing requests with a valid
base.reasoning_effort and an invalid base.reasoning to be rejected; update the
validation logic to mirror the conversion precedence used when building
reasoning_effort: first accept and validate base.reasoning_effort if present,
and only if it is None perform the legacy conversion/validation via
base.reasoning and to_openai_effort(); adjust the validator that checks
payload.reasoning to only run when base.reasoning_effort is absent so the
precedence is consistent with the reasoning_effort initialization.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 6ef84bca-aacb-4080-9754-7618e58dbd2e
📒 Files selected for processing (3)
src/models/chat.rssrc/providers/azure/provider.rssrc/providers/openai/provider.rs
Contributor
🔒 Container Vulnerability Scan (hub - amd64)Click to expand results |
Contributor
🔒 Container Vulnerability Scan (hub - arm64)Click to expand results |
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (1)
src/providers/openai/provider.rs (1)
88-97:⚠️ Potential issue | 🟠 Major | ⚡ Quick winAdd direct validation for top-level
reasoning_effort.When
reasoning_effortis present, invalid values (including whitespace-only) currently bypass validation and are sent downstream. This keeps validation behavior inconsistent with legacyreasoningand defers a user input error to the provider call path.Suggested fix
- if payload.reasoning_effort.is_none() { + if let Some(effort) = payload.reasoning_effort.as_deref() { + let effort = effort.trim(); + if !["low", "medium", "high"].contains(&effort) { + tracing::error!("Invalid reasoning_effort: {}", effort); + return Err(StatusCode::BAD_REQUEST); + } + } else { if let Some(reasoning) = &payload.reasoning { if let Err(e) = reasoning.validate() { tracing::error!("Invalid reasoning config: {}", e); return Err(StatusCode::BAD_REQUEST); } } }🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/providers/openai/provider.rs` around lines 88 - 97, The top-level payload.reasoning_effort is not validated when present, so invalid/whitespace-only values get sent downstream; update the same validation flow used for legacy payload.reasoning by validating payload.reasoning_effort when Some(...) and returning Err(StatusCode::BAD_REQUEST) on failure: call the same validation routine (or replicate its checks: trim/empty and semantic validation) and log errors via tracing::error! (consistent with reasoning.validate() handling), ensuring behavior matches OpenAIChatCompletionRequest::from precedence.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Duplicate comments:
In `@src/providers/openai/provider.rs`:
- Around line 88-97: The top-level payload.reasoning_effort is not validated
when present, so invalid/whitespace-only values get sent downstream; update the
same validation flow used for legacy payload.reasoning by validating
payload.reasoning_effort when Some(...) and returning
Err(StatusCode::BAD_REQUEST) on failure: call the same validation routine (or
replicate its checks: trim/empty and semantic validation) and log errors via
tracing::error! (consistent with reasoning.validate() handling), ensuring
behavior matches OpenAIChatCompletionRequest::from precedence.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 7eda2a90-130a-4ffa-b852-c4a36ef9843e
📒 Files selected for processing (2)
src/providers/azure/provider.rssrc/providers/openai/provider.rs
🚧 Files skipped from review as they are similar to previous changes (1)
- src/providers/azure/provider.rs
doronkopit5
approved these changes
May 3, 2026
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (1)
src/providers/azure/provider.rs (1)
286-330: 🏗️ Heavy liftAdd one provider-level test for the
chat_completionsguard path.These tests cover conversion well, but they don’t directly lock behavior in
AzureProvider::chat_completions(Line 87). A regression there could reintroduce the old rejection path even with a valid top-level effort. Consider adding one async/provider-level test where top-levelreasoning_effortis set and legacyreasoningis invalid, and assert it is not rejected at validation time.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/providers/azure/provider.rs` around lines 286 - 330, Add an async provider-level test that ensures AzureProvider::chat_completions accepts a request with a top-level reasoning_effort when a legacy nested Reasoning is invalid: create a #[tokio::test] async fn (e.g., accepts_top_level_effort_over_legacy_rejection) that builds the same base_request, sets req.reasoning_effort = Some("minimal".to_string()) and sets req.reasoning = Some(ReasoningConfig { effort: None, ..Default::default() }) or otherwise invalid legacy reasoning, constructs an AzureProvider instance (matching existing test setup), calls provider.chat_completions(req).await and asserts the call succeeds (is Ok) or does not return the old validation/rejection error; reference AzureProvider::chat_completions and AzureChatCompletionRequest conversion to locate where to wire the request.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In `@src/providers/azure/provider.rs`:
- Around line 286-330: Add an async provider-level test that ensures
AzureProvider::chat_completions accepts a request with a top-level
reasoning_effort when a legacy nested Reasoning is invalid: create a
#[tokio::test] async fn (e.g., accepts_top_level_effort_over_legacy_rejection)
that builds the same base_request, sets req.reasoning_effort =
Some("minimal".to_string()) and sets req.reasoning = Some(ReasoningConfig {
effort: None, ..Default::default() }) or otherwise invalid legacy reasoning,
constructs an AzureProvider instance (matching existing test setup), calls
provider.chat_completions(req).await and asserts the call succeeds (is Ok) or
does not return the old validation/rejection error; reference
AzureProvider::chat_completions and AzureChatCompletionRequest conversion to
locate where to wire the request.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 346a015b-79a4-4745-9c79-af6ac986f8c8
📒 Files selected for processing (2)
src/providers/azure/provider.rssrc/providers/openai/provider.rs
🚧 Files skipped from review as they are similar to previous changes (1)
- src/providers/openai/provider.rs
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary by CodeRabbit
New Features
Behavior Change
Tests