feat(api): (1/n) datasets api clean up (#1573)

## PR Stack
- https://github.com/meta-llama/llama-stack/pull/1573
- https://github.com/meta-llama/llama-stack/pull/1625
- https://github.com/meta-llama/llama-stack/pull/1656
- https://github.com/meta-llama/llama-stack/pull/1657
- https://github.com/meta-llama/llama-stack/pull/1658
- https://github.com/meta-llama/llama-stack/pull/1659
- https://github.com/meta-llama/llama-stack/pull/1660

**Client SDK**
- https://github.com/meta-llama/llama-stack-client-python/pull/203

**CI**
- 1391130488
<img width="1042" alt="image"
src="https://github.com/user-attachments/assets/69636067-376d-436b-9204-896e2dd490ca"
/>
-- the test_rag_agent_with_attachments is flaky and not related to this
PR

## Doc
<img width="789" alt="image"
src="https://github.com/user-attachments/assets/b88390f3-73d6-4483-b09a-a192064e32d9"
/>


## Client Usage
```python
client.datasets.register(
    source={
        "type": "uri",
        "uri": "lsfs://mydata.jsonl",
    },
    schema="jsonl_messages",
    # optional 
    dataset_id="my_first_train_data"
)

# quick prototype debugging
client.datasets.register(
    data_reference={
        "type": "rows",
        "rows": [
                "messages": [...],
        ],
    },
    schema="jsonl_messages",
)
```

## Test Plan
- CI:
1387805545

```
LLAMA_STACK_CONFIG=fireworks pytest -v tests/integration/datasets/test_datasets.py
```

```
LLAMA_STACK_CONFIG=fireworks pytest -v tests/integration/scoring/test_scoring.py
```

```
pytest -v -s --nbval-lax ./docs/notebooks/Llama_Stack_Benchmark_Evals.ipynb
```

llama_stack/distribution/routers/routers.py

@@ -12,7 +12,8 @@ from llama_stack.apis.common.content_types import (
     InterleavedContent,
     InterleavedContentItem,
 )
-from llama_stack.apis.datasetio import DatasetIO, PaginatedRowsResult
+from llama_stack.apis.datasetio import DatasetIO, IterrowsResponse
+from llama_stack.apis.datasets import DatasetPurpose, DataSource
 from llama_stack.apis.eval import (
     BenchmarkConfig,
     Eval,
@@ -160,7 +161,11 @@ class InferenceRouter(Inference):
         await self.routing_table.register_model(model_id, provider_model_id, provider_id, metadata, model_type)
 
     def _construct_metrics(
-        self, prompt_tokens: int, completion_tokens: int, total_tokens: int, model: Model
+        self,
+        prompt_tokens: int,
+        completion_tokens: int,
+        total_tokens: int,
+        model: Model,
     ) -> List[MetricEvent]:
         """Constructs a list of MetricEvent objects containing token usage metrics.
 
@@ -298,7 +303,12 @@ class InferenceRouter(Inference):
                             completion_text += chunk.event.delta.text
                     if chunk.event.event_type == ChatCompletionResponseEventType.complete:
                         completion_tokens = await self._count_tokens(
-                            [CompletionMessage(content=completion_text, stop_reason=StopReason.end_of_turn)],
+                            [
+                                CompletionMessage(
+                                    content=completion_text,
+                                    stop_reason=StopReason.end_of_turn,
+                                )
+                            ],
                             tool_config.tool_prompt_format,
                         )
                         total_tokens = (prompt_tokens or 0) + (completion_tokens or 0)
@@ -471,21 +481,36 @@ class DatasetIORouter(DatasetIO):
         logger.debug("DatasetIORouter.shutdown")
         pass
 
-    async def get_rows_paginated(
+    async def register_dataset(
+        self,
+        purpose: DatasetPurpose,
+        source: DataSource,
+        metadata: Optional[Dict[str, Any]] = None,
+        dataset_id: Optional[str] = None,
+    ) -> None:
+        logger.debug(
+            f"DatasetIORouter.register_dataset: {purpose=} {source=} {metadata=} {dataset_id=}",
+        )
+        await self.routing_table.register_dataset(
+            purpose=purpose,
+            source=source,
+            metadata=metadata,
+            dataset_id=dataset_id,
+        )
+
+    async def iterrows(
         self,
         dataset_id: str,
-        rows_in_page: int,
-        page_token: Optional[str] = None,
-        filter_condition: Optional[str] = None,
-    ) -> PaginatedRowsResult:
+        start_index: Optional[int] = None,
+        limit: Optional[int] = None,
+    ) -> IterrowsResponse:
         logger.debug(
-            f"DatasetIORouter.get_rows_paginated: {dataset_id}, rows_in_page={rows_in_page}",
+            f"DatasetIORouter.iterrows: {dataset_id}, {start_index=} {limit=}",
         )
-        return await self.routing_table.get_provider_impl(dataset_id).get_rows_paginated(
+        return await self.routing_table.get_provider_impl(dataset_id).iterrows(
             dataset_id=dataset_id,
-            rows_in_page=rows_in_page,
-            page_token=page_token,
-            filter_condition=filter_condition,
+            start_index=start_index,
+            limit=limit,
         )
 
     async def append_rows(self, dataset_id: str, rows: List[Dict[str, Any]]) -> None:

llama_stack/distribution/routers/routing_tables.py

@@ -5,6 +5,7 @@
 # the root directory of this source tree.
 
 import logging
+import uuid
 from typing import Any, Dict, List, Optional
 
 from pydantic import TypeAdapter
@@ -12,7 +13,14 @@ from pydantic import TypeAdapter
 from llama_stack.apis.benchmarks import Benchmark, Benchmarks, ListBenchmarksResponse
 from llama_stack.apis.common.content_types import URL
 from llama_stack.apis.common.type_system import ParamType
-from llama_stack.apis.datasets import Dataset, Datasets, ListDatasetsResponse
+from llama_stack.apis.datasets import (
+    Dataset,
+    DatasetPurpose,
+    Datasets,
+    DatasetType,
+    DataSource,
+    ListDatasetsResponse,
+)
 from llama_stack.apis.models import ListModelsResponse, Model, Models, ModelType
 from llama_stack.apis.resource import ResourceType
 from llama_stack.apis.scoring_functions import (
@@ -352,34 +360,42 @@ class DatasetsRoutingTable(CommonRoutingTableImpl, Datasets):
 
     async def register_dataset(
         self,
-        dataset_id: str,
-        dataset_schema: Dict[str, ParamType],
-        url: URL,
-        provider_dataset_id: Optional[str] = None,
-        provider_id: Optional[str] = None,
+        purpose: DatasetPurpose,
+        source: DataSource,
         metadata: Optional[Dict[str, Any]] = None,
-    ) -> None:
-        if provider_dataset_id is None:
-            provider_dataset_id = dataset_id
-        if provider_id is None:
-            # If provider_id not specified, use the only provider if it supports this dataset
-            if len(self.impls_by_provider_id) == 1:
-                provider_id = list(self.impls_by_provider_id.keys())[0]
+        dataset_id: Optional[str] = None,
+    ) -> Dataset:
+        if not dataset_id:
+            dataset_id = f"dataset-{str(uuid.uuid4())}"
+
+        provider_dataset_id = dataset_id
+
+        # infer provider from source
+        if source.type == DatasetType.rows.value:
+            provider_id = "localfs"
+        elif source.type == DatasetType.uri.value:
+            # infer provider from uri
+            if source.uri.startswith("huggingface"):
+                provider_id = "huggingface"
             else:
-                raise ValueError(
-                    f"No provider specified and multiple providers available. Please specify a provider_id. Available providers: {self.impls_by_provider_id.keys()}"
-                )
+                provider_id = "localfs"
+        else:
+            raise ValueError(f"Unknown data source type: {source.type}")
+
         if metadata is None:
             metadata = {}
+
         dataset = Dataset(
             identifier=dataset_id,
             provider_resource_id=provider_dataset_id,
             provider_id=provider_id,
-            dataset_schema=dataset_schema,
-            url=url,
+            purpose=purpose,
+            source=source,
             metadata=metadata,
         )
+
         await self.register_object(dataset)
+        return dataset
 
     async def unregister_dataset(self, dataset_id: str) -> None:
         dataset = await self.get_dataset(dataset_id)

llama_stack/distribution/ui/page/evaluations/native_eval.py

@@ -166,11 +166,10 @@ def run_evaluation_3():
     eval_candidate = st.session_state["eval_candidate"]
 
     dataset_id = benchmarks[selected_benchmark].dataset_id
-    rows = llama_stack_api.client.datasetio.get_rows_paginated(
+    rows = llama_stack_api.client.datasets.iterrows(
         dataset_id=dataset_id,
-        rows_in_page=-1,
     )
-    total_rows = len(rows.rows)
+    total_rows = len(rows.data)
     # Add number of examples control
     num_rows = st.number_input(
         "Number of Examples to Evaluate",
@@ -195,7 +194,7 @@ def run_evaluation_3():
     if st.button("Run Evaluation"):
         progress_text = "Running evaluation..."
         progress_bar = st.progress(0, text=progress_text)
-        rows = rows.rows
+        rows = rows.data
         if num_rows < total_rows:
             rows = rows[:num_rows]