Browser and evals

[skull8888888]

Important

This pull request enhances evaluation and browser session handling by adding new API endpoints, updating database schemas, and improving frontend components.

• Backend Changes:
  • Added evals.rs to handle evaluation-related API endpoints, including init_eval and save_eval_datapoints.
  • Modified browser_sessions.rs to fix SQL syntax.
  • Updated evaluations.rs to include indices in set_evaluation_results().
  • Adjusted payload size thresholds in main.rs and spans.rs.
• Database:
  • Added index column to evaluation_results in 0018_vengeful_violations.sql.
  • Updated schema in 0018_snapshot.json and _journal.json.
• Frontend Changes:
  • Updated route.ts files for browser sessions and evaluations to handle new API logic.
  • Enhanced evaluation.tsx to support score columns and Supabase integration.
  • Modified session-player.tsx for improved event handling and UI updates.
  • Adjusted layout.tsx and page.tsx for project and evaluation pages.
  • Updated package.json to use rrweb-player version 2.0.0-alpha.18.

^{This description was created by for fc0cfe3. It will automatically update as commits are pushed.}

[ellipsis-dev]

👍 Looks good to me! Reviewed everything up to 6dcb0ff in 1 minute and 5 seconds

More details

• Looked at 414 lines of code in 6 files
• Skipped 0 files when reviewing.
• Skipped posting 4 drafted comments based on config settings.

1. app-server/src/api/v1/evals.rs:31

• Draft comment:
  Consider using unwrap_or("default") instead of unwrap_or_else(|| "default".to_string()) for better performance.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The use of unwrap_or_else with a string allocation can be optimized by using unwrap_or with a string slice.

2. app-server/src/api/v1/evals.rs:65

• Draft comment:
  Consider using unwrap_or("default") instead of unwrap_or_else(|| "default".to_string()) for better performance.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The use of unwrap_or_else with a string allocation can be optimized by using unwrap_or with a string slice.

3. app-server/src/main.rs:434

• Draft comment:
  Consider using the actix_cors middleware to handle CORS instead of manually setting headers.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The options_handler function is manually setting CORS headers. It's better to use the actix_cors middleware for handling CORS in a more robust and configurable way.

4. frontend/components/traces/session-player.tsx:44

• Draft comment:
  Use number instead of NodeJS.Timeout for debounceTimerRef to avoid type issues in a browser environment.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The debounceTimerRef is initialized with NodeJS.Timeout, which is not correct for a browser environment. It should be number instead.

Workflow ID: wflow_aCRaB4mlg5ZMO6pQ

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[dinmukhamedm]

we already have that in global Clickhouse settings in main.rs

[dinmukhamedm]

Is this an intermediate state while we migrate from the previous evaluations? If not, why separate file?

[ellipsis-dev]

👍 Looks good to me! Incremental review on 802728e in 33 seconds

More details

• Looked at 172 lines of code in 2 files
• Skipped 0 files when reviewing.
• Skipped posting 3 drafted comments based on config settings.

1. app-server/src/main.rs:82

• Draft comment:
  Consider documenting the reason for changing the GRPC_PAYLOAD_DECODING_LIMIT from 100MB to 10MB for future reference and clarity.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The change in payload limits is significant and should be documented for clarity and future reference.

2. app-server/src/main.rs:81

• Draft comment:
  Consider documenting the reason for changing the HTTP_PAYLOAD_LIMIT from 100MB to 5MB for future reference and clarity.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The change in payload limits is significant and should be documented for clarity and future reference.

3. frontend/components/traces/session-player.tsx:154

• Draft comment:
  Ensure that removing cleanupDanglingNodes does not lead to memory leaks or other issues with dangling nodes.
• Reason this comment was not posted:
  Comment did not seem useful.

Workflow ID: wflow_Mnz1iVxogIuhPsBb

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[ellipsis-dev]

👍 Looks good to me! Incremental review on 6a92202 in 13 seconds

More details

• Looked at 20 lines of code in 1 files
• Skipped 0 files when reviewing.
• Skipped posting 1 drafted comments based on config settings.

1. frontend/components/traces/session-player.tsx:6

• Draft comment:
  Consider organizing the import statements by grouping external libraries first, followed by internal modules. This improves readability.
• Reason this comment was not posted:
  Confidence changes required: 10%
  The import statements should be organized for better readability.

Workflow ID: wflow_006WIK4PSZRytZnW

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[ellipsis-dev]

👍 Looks good to me! Incremental review on 3f4f858 in 32 seconds

More details

• Looked at 3157 lines of code in 9 files
• Skipped 0 files when reviewing.
• Skipped posting 3 drafted comments based on config settings.

1. app-server/src/db/evaluations.rs:93

• Draft comment:
  The 'index' column is introduced, but the existing 'index_in_batch' column is not removed or deprecated. Ensure that both columns are necessary and clarify their distinct purposes to avoid redundancy.
• Reason this comment was not posted:
  Comment did not seem useful.

2. frontend/app/api/projects/[projectId]/evaluations/[evaluationId]/route.ts:55

• Draft comment:
  Ensure that the change in order of 'orderBy' clause to prioritize 'index' over 'createdAt' is consistent across all related queries to maintain the intended ordering logic.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The PR changes the order of the 'orderBy' clause in SQL queries to prioritize 'index' over 'createdAt'. This change should be consistent across all related queries to ensure the intended ordering logic is applied everywhere.

3. frontend/app/project/[projectId]/evaluations/[evaluationId]/page.tsx:87

• Draft comment:
  Ensure that the change in order of 'orderBy' clause to prioritize 'index' over 'createdAt' is consistent across all related queries to maintain the intended ordering logic.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The PR changes the order of the 'orderBy' clause in SQL queries to prioritize 'index' over 'createdAt'. This change should be consistent across all related queries to ensure the intended ordering logic is applied everywhere.

Workflow ID: wflow_MhdvCDNcA4d8tILd

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[ellipsis-dev]

👍 Looks good to me! Incremental review on 015e8c5 in 59 seconds

More details

• Looked at 346 lines of code in 8 files
• Skipped 0 files when reviewing.
• Skipped posting 0 drafted comments based on config settings.

Workflow ID: wflow_UTubcbRAdTYapOHg

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[ellipsis-dev]

👍 Looks good to me! Incremental review on fc0cfe3 in 16 seconds

More details

• Looked at 13 lines of code in 1 files
• Skipped 1 files when reviewing.
• Skipped posting 1 drafted comments based on config settings.

1. frontend/package.json:106

• Draft comment:
  Ensure that the update to rrweb-player is compatible with other dependencies and does not introduce any breaking changes.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The update to rrweb-player should be checked for compatibility with other dependencies.

Workflow ID: wflow_psTlsiW2UcYGURMy

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.