-
Notifications
You must be signed in to change notification settings - Fork 78
Extract workspace crates #167
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
…d prompt processing
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pull request overview
This PR restructures the repository into a Cargo workspace with separate crates for shared types (paddler_types) and a Rust client library (paddler_client), and updates the app code (paddler) to consume the extracted types. It also introduces a new inference parameter to cap embedding batch parallelism.
Changes:
- Converted the repo into a workspace (
Cargo.toml) with member cratespaddler,paddler_client, andpaddler_types. - Moved/duplicated many previously in-crate types into
paddler_typesand updated imports acrosspaddlerto use the shared crate. - Added
embedding_n_seq_maxto inference parameters and surfaced it in the TS schema + admin UI.
Reviewed changes
Copilot reviewed 96 out of 226 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Description |
|---|---|
| Cargo.toml | Defines the workspace, members, and shared dependency versions. |
| paddler/Cargo.toml | New crate manifest for the server/balancer/agent binary & services. |
| paddler_types/Cargo.toml | New shared-types crate with optional validation feature. |
| paddler_client/Cargo.toml | New Rust client crate for HTTP + WebSocket inference/management APIs. |
| paddler_types/src/inference_parameters.rs | Adds embedding_n_seq_max and updates defaults. |
| resources/ts/schemas/InferenceParameters.ts | Adds embedding_n_seq_max to the Zod schema. |
| resources/ts/components/ChangeModelForm.tsx | Adds UI input for embedding_n_seq_max. |
| paddler_client/src/inference_socket_pool.rs | Implements a pooled WebSocket request/response dispatcher. |
| paddler/src/balancer/http_route/get_health.rs | Adds a health endpoint module for services. |
| paddler/src/balancer/management_service/http_route/get_metrics.rs | Adds a Prometheus-style /metrics endpoint. |
Comments suppressed due to low confidence (1)
paddler_types/src/inference_parameters.rs:12
InferenceParametersis#[serde(deny_unknown_fields)]and the newly addedembedding_n_seq_maxfield is required. Deserializing older persisted configs/state that don’t include this field will now fail. Add a serde default for this field (or#[serde(default)]on the struct leveraging the existingDefaultimpl) so older JSON continues to load with a sensible default (e.g. 16).
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pull request overview
Copilot reviewed 96 out of 226 changed files in this pull request and generated 2 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pull request overview
Copilot reviewed 96 out of 226 changed files in this pull request and generated no new comments.
Comments suppressed due to low confidence (1)
paddler_types/src/inference_parameters.rs:36
- The default
batch_n_tokenschanged to 2048. This is a behavior change that can significantly increase memory usage and latency by default (especially on smaller agents). If this wasn’t intentional as part of the workspace extraction, consider keeping the previous default or documenting why the new default is safe.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
No description provided.