Initial public release

backend/app/schemas/assistant.py

@@ -0,0 +1,93 @@
+from __future__ import annotations
+
+from datetime import datetime
+from typing import Literal
+from uuid import UUID
+
+from pydantic import BaseModel, Field
+
+from app.schemas.operations import VectorSelectionBBox
+
+
+class AssistantChatMessage(BaseModel):
+    role: Literal["user", "assistant"]
+    content: str = Field(min_length=1, max_length=4_000)
+
+
+class AssistantQueryRequest(BaseModel):
+    question: str = Field(min_length=2, max_length=2_000)
+    model: str | None = Field(default=None, max_length=255)
+    bbox: VectorSelectionBBox | None = None
+    area_id: UUID | None = None
+    history: list[AssistantChatMessage] = Field(default_factory=list, max_length=8)
+
+
+class AssistantModelRead(BaseModel):
+    name: str
+    size_bytes: int | None = None
+    parameter_size: str | None = None
+    quantization_level: str | None = None
+    capabilities: list[str] = Field(default_factory=list)
+
+
+class AssistantModelList(BaseModel):
+    items: list[AssistantModelRead]
+    total: int
+    default_model: str | None = None
+
+
+class AssistantStatus(BaseModel):
+    enabled: bool
+    reachable: bool
+    status: str
+    base_url: str
+    default_model: str | None = None
+    model_count: int = 0
+    limitation_message: str
+
+
+class AssistantContextMetric(BaseModel):
+    theme: str
+    label: str
+    value: float
+    unit: str
+    source: str
+    dataset_id: UUID
+    observed_at: datetime | None = None
+    is_estimate: bool = False
+
+
+class AssistantTemporalSeries(BaseModel):
+    temporal_series_key: str
+    label: str
+    source: str
+    first_year: int
+    last_year: int
+    observation_count: int
+
+
+class AssistantEstimateDisclosure(BaseModel):
+    """A value in the answer that the source itself calls an estimate.
+
+    Derived from metric metadata rather than from the generated sentences, so
+    the disclosure is present whatever wording the model chose.
+    """
+
+    theme: str
+    label: str
+    unit: str
+    source: str
+    dataset_id: UUID
+    reason: str
+
+
+class AssistantQueryResponse(BaseModel):
+    answer: str
+    model: str
+    scope_label: str
+    context_metrics: list[AssistantContextMetric]
+    temporal_series: list[AssistantTemporalSeries]
+    estimate_disclosures: list[AssistantEstimateDisclosure] = Field(default_factory=list)
+    source_dataset_ids: list[UUID]
+    warnings: list[str]
+    generated_at: datetime