feat(api): update API spec from langfuse/langfuse fd3dc9e#1748
feat(api): update API spec from langfuse/langfuse fd3dc9e#1748langfuse-bot wants to merge 1 commit into
Conversation
|
@claude review |
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
There was a problem hiding this comment.
Auto-generated Fern API spec update — adds PARQUET file type and booleanObject filter support.
Extended reasoning...
Overview
This PR is an auto-generated update from the Fern API spec generator (all four files carry the # This file was auto-generated by Fern from our API Definition. header). It makes two additions to the SDK's API surface:
- Adds
PARQUETto theBlobStorageIntegrationFileTypeenum, updates its docstring, updates thevisitmethod signature accordingly, and removes an outdated caveat fromBlobStorageIntegrationFileTypeResponse. - Extends the trace list filter documentation to include a new
booleanObjectfilter type and ascore_booleanscolumn.
Security risks
None. These are purely additive changes to enum values and docstring content in auto-generated bindings — no auth, crypto, or input-handling logic is touched.
Level of scrutiny
Low. These are mechanical, additive changes to Fern-generated code that mirror an upstream API spec. The visit method addition is internally consistent (new enum value → new callback parameter → new branch), and the trace filter changes are docstring-only.
Other factors
The changes are self-consistent and follow existing patterns in the file. Note that adding a required parquet parameter to visit() is technically a breaking change for any caller of that method, but this is standard for Fern-generated visitor patterns when a new enum variant is added upstream, and is expected behavior for API SDK updates.
1 participant
Greptile Summary
This PR syncs the Python SDK with an updated Langfuse API spec, promoting PARQUET from a response-only file type to a fully requestable export format, and extending the trace-list filter API with a new
booleanObjectfilter type for boolean score filtering.BlobStorageIntegrationFileTypegains aPARQUETenum value and a matchingparquetarm invisit(), whileBlobStorageIntegrationFileTypeResponsehas its docstring simplified now that PARQUET is available in both request and response types.trace/client.pyandtrace/raw_client.py(sync and async variants) have their inline filter-spec documentation updated to documentbooleanObjectoperator support and the newscore_booleanscolumn, keeping the four client classes in sync.Confidence Score: 5/5
All changes are auto-generated from the API spec and are additive — new enum value, new visit() arm, and new docstring entries. No existing behaviour is removed or altered.
The PR adds PARQUET to the request-side file type enum (it was already in the response enum), updates a visit() method to handle it, simplifies a now-outdated docstring, and extends trace-filter documentation with the new booleanObject type across all four client classes. All four files are stamped as auto-generated by Fern, the changes are consistent and additive, and no pre-existing logic is touched.
No files require special attention.
Flowchart
%%{init: {'theme': 'neutral'}}%% flowchart TD A[API Export Request] --> B{fileType?} B -->|JSON| C[JSON export] B -->|CSV| D[CSV export] B -->|JSONL| E[JSONL export] B -->|PARQUET NEW| F[Parquet export\ncolumnar binary format\ngzip does not apply\nno model-price columns in observations] G[Trace List Filter] --> H{filter type?} H -->|stringObject| I[string key-value filter] H -->|numberObject| J[number key-value filter] H -->|categoryOptions| K[categorical score filter] H -->|booleanObject NEW| L[boolean key-value filter\ne.g. score_booleans\noperators: = and neq] H -->|boolean| M[plain boolean filter] H -->|datetime / string / number / etc.| N[other filter types]%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%% flowchart TD A[API Export Request] --> B{fileType?} B -->|JSON| C[JSON export] B -->|CSV| D[CSV export] B -->|JSONL| E[JSONL export] B -->|PARQUET NEW| F[Parquet export\ncolumnar binary format\ngzip does not apply\nno model-price columns in observations] G[Trace List Filter] --> H{filter type?} H -->|stringObject| I[string key-value filter] H -->|numberObject| J[number key-value filter] H -->|categoryOptions| K[categorical score filter] H -->|booleanObject NEW| L[boolean key-value filter\ne.g. score_booleans\noperators: = and neq] H -->|boolean| M[plain boolean filter] H -->|datetime / string / number / etc.| N[other filter types]Reviews (1): Last reviewed commit: "feat(api): update API spec from langfuse..." | Re-trigger Greptile