7f72171572
Initialize git, add Apache-2.0 LICENSE, .gitattributes (LF line endings), AGENTS.md (entry points, stack, discovery order, baseline checks), RUNBOOK.md (dev boot, prod deploy with overlay, ingestion, failures, rollback, scaling notes), .env.prod.example with rotated credential placeholders, and dev-only warnings on .env.example. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
100 lines
2.2 KiB
Python
100 lines
2.2 KiB
Python
"""Pydantic request/response schemas for the LegacyHUB API."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import uuid
|
|
from datetime import datetime
|
|
from typing import Any, Literal
|
|
|
|
from pydantic import BaseModel, Field
|
|
|
|
|
|
# ---------------- Health ----------------
|
|
|
|
class ComponentHealth(BaseModel):
|
|
name: str
|
|
status: Literal["ok", "error", "degraded"]
|
|
detail: dict[str, Any] = Field(default_factory=dict)
|
|
|
|
|
|
class HealthResponse(BaseModel):
|
|
status: Literal["ok", "error", "degraded"]
|
|
version: str
|
|
components: list[ComponentHealth]
|
|
|
|
|
|
# ---------------- Ingestion ----------------
|
|
|
|
class IngestFolderRequest(BaseModel):
|
|
path: str = Field(..., description="Absolute path inside the API container")
|
|
recursive: bool = True
|
|
force: bool = False
|
|
|
|
|
|
class IngestFolderResponse(BaseModel):
|
|
run_id: uuid.UUID
|
|
discovered: int
|
|
queued: int
|
|
skipped_duplicates: int
|
|
invalid_files: int
|
|
|
|
|
|
class DocumentSummary(BaseModel):
|
|
id: uuid.UUID
|
|
original_file_name: str
|
|
source_path: str
|
|
sha256: str
|
|
status: str
|
|
file_size_bytes: int
|
|
created_at: datetime
|
|
|
|
|
|
# ---------------- Search ----------------
|
|
|
|
SearchMode = Literal["lexical", "semantic", "hybrid"]
|
|
|
|
|
|
class SearchFilters(BaseModel):
|
|
document_id: uuid.UUID | None = None
|
|
source_path: str | None = None
|
|
block_type: str | None = None
|
|
min_ocr_confidence: float | None = Field(None, ge=0.0, le=1.0)
|
|
|
|
|
|
class SearchRequest(BaseModel):
|
|
query: str = Field(..., min_length=1)
|
|
limit: int = Field(10, ge=1, le=100)
|
|
filters: SearchFilters = Field(default_factory=SearchFilters)
|
|
search_mode: SearchMode = "hybrid"
|
|
|
|
|
|
class Citation(BaseModel):
|
|
pdf: str
|
|
page: int
|
|
block_id: str | None = None
|
|
table_id: str | None = None
|
|
figure_id: str | None = None
|
|
|
|
|
|
class SearchHit(BaseModel):
|
|
rank: int
|
|
score: float
|
|
document_id: uuid.UUID
|
|
chunk_id: uuid.UUID
|
|
original_file_name: str
|
|
source_path: str
|
|
page_number: int
|
|
block_type: str
|
|
text: str
|
|
citation: Citation
|
|
quality_flags: dict[str, Any] = Field(default_factory=dict)
|
|
metadata: dict[str, Any] = Field(default_factory=dict)
|
|
|
|
|
|
class SearchResponse(BaseModel):
|
|
query: str
|
|
mode: SearchMode
|
|
total_candidates: int
|
|
reranked: bool
|
|
results: list[SearchHit]
|