Админ-панель (доступ: is_admin + секретный код сессии в /admin/<code>): - разделы: дашборд (здоровье сервисов, статистика), клиенты, работы, база документов, хранилище MinIO, источники парсинга, отстойник работ - backend: модели ParseSource/StagedWork/AdminSession, миграция 003, core/admin (авторизация), роутер api/admin - frontend: AdminLayout + 7 вкладок, adminApi, вход из кабинета - Celery index.run_parser (фоновый парсинг), автосбор проверенных работ в отстойник с ручным одобрением → индексация в базу Исправления: - openalex parser: тип article (не journal-article), убран is_oa-фильтр - winnowing: приведение xxh64 к signed int64 (фикс bigint out of range) - bcrypt пин 4.0.1 (совместимость с passlib 1.7.4) - alembic: prepend_sys_path + asyncpg-драйвер - frontend: контракт Task (public_id вместо id), react-dropzone Инфраструктура: - docker-compose.test.yml: api+frontend+воркеры локально, БД/кэш/хранилище/ брокер/LLM — на удалённых серверах через .env - .gitignore: исправлено ошибочное игнорирование кода services/*/app/models/ - LICENSE: проприетарная Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
104 lines
4.9 KiB
Python
104 lines
4.9 KiB
Python
"""Минимальные модели для индексер-воркера."""
|
|
|
|
from datetime import datetime
|
|
|
|
from sqlalchemy import JSON, BigInteger, ForeignKey, Integer, String, Text, func
|
|
from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column
|
|
|
|
|
|
class Base(DeclarativeBase):
|
|
pass
|
|
|
|
|
|
class User(Base):
|
|
__tablename__ = "users"
|
|
id: Mapped[int] = mapped_column(Integer, primary_key=True)
|
|
email: Mapped[str] = mapped_column(String(255))
|
|
name: Mapped[str] = mapped_column(String(255))
|
|
plan: Mapped[str] = mapped_column(String(20), default="free")
|
|
|
|
|
|
class Task(Base):
|
|
__tablename__ = "tasks"
|
|
id: Mapped[str] = mapped_column(String(36), primary_key=True)
|
|
user_id: Mapped[int] = mapped_column(ForeignKey("users.id"))
|
|
type: Mapped[str] = mapped_column(String(20))
|
|
status: Mapped[str] = mapped_column(String(20), default="queued")
|
|
celery_task_id: Mapped[str | None] = mapped_column(String(255), nullable=True)
|
|
input_data: Mapped[dict] = mapped_column(JSON, default=dict)
|
|
result: Mapped[dict | None] = mapped_column(JSON, nullable=True)
|
|
error: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
queue_position: Mapped[int | None] = mapped_column(nullable=True)
|
|
created_at: Mapped[datetime] = mapped_column(server_default=func.now())
|
|
updated_at: Mapped[datetime | None] = mapped_column(nullable=True)
|
|
|
|
|
|
class Document(Base):
|
|
__tablename__ = "documents"
|
|
id: Mapped[int] = mapped_column(Integer, primary_key=True)
|
|
source: Mapped[str] = mapped_column(String(50))
|
|
ext_id: Mapped[str] = mapped_column(String(255), unique=True)
|
|
doi: Mapped[str | None] = mapped_column(String(255), nullable=True)
|
|
title: Mapped[str] = mapped_column(Text)
|
|
authors: Mapped[list] = mapped_column(JSON, default=list)
|
|
year: Mapped[int | None] = mapped_column(nullable=True)
|
|
lang: Mapped[str | None] = mapped_column(String(10), nullable=True)
|
|
journal: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
volume: Mapped[str | None] = mapped_column(String(50), nullable=True)
|
|
issue: Mapped[str | None] = mapped_column(String(50), nullable=True)
|
|
pages: Mapped[str | None] = mapped_column(String(50), nullable=True)
|
|
abstract: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
url: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
minio_key: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
faiss_id: Mapped[int | None] = mapped_column(nullable=True)
|
|
indexed_at: Mapped[datetime] = mapped_column(server_default=func.now())
|
|
|
|
|
|
class Fingerprint(Base):
|
|
__tablename__ = "fingerprints"
|
|
id: Mapped[int] = mapped_column(Integer, primary_key=True)
|
|
doc_id: Mapped[int] = mapped_column(ForeignKey("documents.id"))
|
|
hash_value: Mapped[int] = mapped_column(BigInteger)
|
|
position: Mapped[int] = mapped_column()
|
|
|
|
|
|
class ParseSource(Base):
|
|
__tablename__ = "parse_sources"
|
|
id: Mapped[int] = mapped_column(Integer, primary_key=True)
|
|
source_type: Mapped[str] = mapped_column(String(50))
|
|
name: Mapped[str] = mapped_column(String(255))
|
|
query: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
lang: Mapped[str | None] = mapped_column(String(10), nullable=True)
|
|
year_from: Mapped[int | None] = mapped_column(nullable=True)
|
|
year_to: Mapped[int | None] = mapped_column(nullable=True)
|
|
limit: Mapped[int] = mapped_column(default=1000)
|
|
enabled: Mapped[bool] = mapped_column(default=True)
|
|
last_status: Mapped[str] = mapped_column(String(20), default="idle")
|
|
last_error: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
last_run_at: Mapped[datetime | None] = mapped_column(nullable=True)
|
|
docs_added: Mapped[int] = mapped_column(default=0)
|
|
created_at: Mapped[datetime] = mapped_column(server_default=func.now())
|
|
|
|
|
|
class StagedWork(Base):
|
|
__tablename__ = "staged_works"
|
|
id: Mapped[int] = mapped_column(Integer, primary_key=True)
|
|
user_id: Mapped[int | None] = mapped_column(nullable=True)
|
|
task_id: Mapped[str | None] = mapped_column(String(36), nullable=True)
|
|
filename: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
minio_key: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
text_key: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
title: Mapped[str | None] = mapped_column(Text, nullable=True)
|
|
authors: Mapped[list] = mapped_column(JSON, default=list)
|
|
year: Mapped[int | None] = mapped_column(nullable=True)
|
|
lang: Mapped[str | None] = mapped_column(String(10), nullable=True)
|
|
word_count: Mapped[int | None] = mapped_column(nullable=True)
|
|
status: Mapped[str] = mapped_column(String(20), default="pending")
|
|
document_id: Mapped[int | None] = mapped_column(nullable=True)
|
|
reviewed_by: Mapped[int | None] = mapped_column(nullable=True)
|
|
reviewed_at: Mapped[datetime | None] = mapped_column(nullable=True)
|
|
created_at: Mapped[datetime] = mapped_column(server_default=func.now())
|
|
|
|
|
|
__all__ = ["Base", "User", "Task", "Document", "Fingerprint", "ParseSource", "StagedWork"]
|