Files
office_translator/database/models.py
sepehr b4e873ad2c
Some checks failed
Deploy to Production / Build and Deploy (push) Failing after 2m14s
feat(review,teams): review foundation — segments, side-by-side editor, rebuild, XLIFF, team workspaces
Foundations:
- TranslationSegment model + migration f7e8d9c0b1a2 (segments, workspaces,
  workspace_members, glossaries.workspace_id)
- SegmentRecorder injected into all 4 translators: unique (source,
  translation) pairs captured per job and persisted (best-effort)
- set_segment_overrides: human-reviewed translations applied verbatim on
  rebuild — top priority over TM and provider, zero API calls

Review API (routes/review_routes.py):
- GET /translations/{id}/segments (owner or job token)
- PATCH /segments/{id} edit/approve — feeds the per-user TM so approved
  translations are reused in later jobs
- POST /translations/{id}/rebuild — rebuild document with reviewed text
- GET/POST /translations/{id}/xliff — XLIFF 1.2 export/import (edited
  segments export their reviewed text)

Review editor (frontend /dashboard/reviews/[jobId]):
- side-by-side source/translation table, inline edit, approve (single or
  all), rebuild & download (auth blob), XLIFF export/import, 13 locales
- 'Relire et corriger' link on the translation-complete screen

Team workspaces (routes/workspace_routes.py + /dashboard/teams):
- Workspace/WorkspaceMember models, roles owner/admin/member
- create (Business plan), list with seat usage, invite by email with
  seat-limit enforcement (Business=5, Enterprise unlimited), removal
- shared glossaries: workspace members can use a glossary shared to their
  workspace (access check extended)

Tests: 1184 passed / 0 failed (11 new: recorder, overrides, docx
capture->rebuild e2e, XLIFF structure/escaping, seats, workspace CRUD,
shared glossary access)
2026-08-29 19:04:32 +02:00

543 lines
18 KiB
Python

"""
SQLAlchemy models for the Document Translation API
"""
import uuid
from datetime import datetime, timezone
import warnings
from sqlalchemy import (
Column,
String,
Integer,
Float,
Boolean,
DateTime,
Text,
ForeignKey,
Enum,
Index,
JSON,
BigInteger,
CheckConstraint,
)
from sqlalchemy.orm import relationship, declarative_base
import enum
def _utcnow() -> datetime:
return datetime.now(timezone.utc)
Base = declarative_base()
def generate_uuid():
"""Generate a new UUID string"""
return str(uuid.uuid4())
def generate_uuid_value():
"""Generate a new UUID value for PostgreSQL UUID column"""
return uuid.uuid4()
class PlanType(str, enum.Enum):
FREE = "free"
STARTER = "starter"
PRO = "pro"
BUSINESS = "business"
ENTERPRISE = "enterprise"
class SubscriptionStatus(str, enum.Enum):
ACTIVE = "active"
CANCELED = "canceled"
PAST_DUE = "past_due"
TRIALING = "trialing"
PAUSED = "paused"
class User(Base):
"""User model for authentication and billing"""
__tablename__ = "users"
id = Column(String(36), primary_key=True, default=generate_uuid)
email = Column(String(255), unique=True, nullable=False, index=True)
name = Column(String(255), nullable=False)
hashed_password = Column(String(255), nullable=False)
tier = Column(String(10), default="free", nullable=False)
daily_translation_count = Column(Integer, default=0, nullable=False)
email_verified = Column(Boolean, default=False)
is_active = Column(Boolean, default=True)
avatar_url = Column(String(500), nullable=True)
plan = Column(Enum(PlanType), default=PlanType.FREE)
subscription_status = Column(
Enum(SubscriptionStatus), default=SubscriptionStatus.ACTIVE
)
stripe_customer_id = Column(String(255), nullable=True, index=True)
stripe_subscription_id = Column(String(255), nullable=True)
subscription_ends_at = Column(DateTime, nullable=True)
cancel_at_period_end = Column(Boolean, default=False, nullable=False)
reset_token = Column(String(255), nullable=True)
reset_token_expires = Column(DateTime, nullable=True)
docs_translated_this_month = Column(Integer, default=0)
pages_translated_this_month = Column(Integer, default=0)
api_calls_this_month = Column(Integer, default=0)
extra_credits = Column(Integer, default=0)
usage_reset_date = Column(DateTime, default=_utcnow)
created_at = Column(DateTime, default=_utcnow)
updated_at = Column(DateTime, default=_utcnow, onupdate=_utcnow)
last_login_at = Column(DateTime, nullable=True)
translations = relationship("Translation", back_populates="user", lazy="select")
api_keys = relationship("ApiKey", back_populates="user", lazy="select")
__table_args__ = (
CheckConstraint("tier IN ('free', 'starter', 'pro', 'business', 'enterprise')", name="ck_users_tier"),
Index("ix_users_email_active", "email", "is_active"),
Index("ix_users_stripe_customer", "stripe_customer_id"),
)
@property
def password_hash(self) -> str:
warnings.warn(
"password_hash is deprecated, use hashed_password instead",
DeprecationWarning,
stacklevel=2,
)
return self.hashed_password
@password_hash.setter
def password_hash(self, value: str) -> None:
warnings.warn(
"password_hash is deprecated, use hashed_password instead",
DeprecationWarning,
stacklevel=2,
)
self.hashed_password = value
def to_dict(self) -> dict:
return {
"id": self.id,
"email": self.email,
"name": self.name,
"avatar_url": self.avatar_url,
"tier": self.tier,
"plan": self.plan.value if self.plan else "free",
"subscription_status": self.subscription_status.value
if self.subscription_status
else "active",
"daily_translation_count": self.daily_translation_count,
"docs_translated_this_month": self.docs_translated_this_month,
"pages_translated_this_month": self.pages_translated_this_month,
"api_calls_this_month": self.api_calls_this_month,
"extra_credits": self.extra_credits,
"email_verified": self.email_verified,
"created_at": self.created_at.isoformat() if self.created_at else None,
}
class Translation(Base):
"""Translation history for analytics and billing"""
__tablename__ = "translations"
id = Column(String(36), primary_key=True, default=generate_uuid)
user_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
# File info
original_filename = Column(String(255), nullable=False)
file_type = Column(String(20), nullable=False) # xlsx, docx, pptx
file_size_bytes = Column(BigInteger, default=0)
page_count = Column(Integer, default=0)
# Translation details
source_language = Column(String(10), default="auto")
target_language = Column(String(10), nullable=False)
provider = Column(String(50), nullable=False) # google, deepl, ollama, etc.
# Processing info
status = Column(
String(20), default="pending"
) # pending, processing, completed, failed
error_message = Column(Text, nullable=True)
processing_time_ms = Column(Integer, nullable=True)
# Cost tracking (for paid providers)
characters_translated = Column(Integer, default=0)
estimated_cost_usd = Column(Float, default=0.0)
# Timestamps
created_at = Column(DateTime, default=_utcnow)
completed_at = Column(DateTime, nullable=True)
# Relationship
user = relationship("User", back_populates="translations")
# Indexes
__table_args__ = (
Index("ix_translations_user_date", "user_id", "created_at"),
Index("ix_translations_status", "status"),
)
def to_dict(self) -> dict:
return {
"id": self.id,
"original_filename": self.original_filename,
"file_type": self.file_type,
"file_size_bytes": self.file_size_bytes,
"page_count": self.page_count,
"source_language": self.source_language,
"target_language": self.target_language,
"provider": self.provider,
"status": self.status,
"processing_time_ms": self.processing_time_ms,
"characters_translated": self.characters_translated,
"created_at": self.created_at.isoformat() if self.created_at else None,
"completed_at": self.completed_at.isoformat()
if self.completed_at
else None,
}
class ApiKey(Base):
"""API keys for programmatic access"""
__tablename__ = "api_keys"
id = Column(String(36), primary_key=True, default=generate_uuid)
user_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
# Key info
name = Column(String(100), nullable=False) # User-friendly name
key_hash = Column(String(255), nullable=False) # SHA256 of the key
key_prefix = Column(String(10), nullable=False) # First 8 chars for identification
# Permissions
is_active = Column(Boolean, default=True)
scopes = Column(JSON, default=list) # ["translate", "read", "write"]
# Usage tracking
last_used_at = Column(DateTime, nullable=True)
usage_count = Column(Integer, default=0)
# Timestamps
created_at = Column(DateTime, default=_utcnow)
expires_at = Column(DateTime, nullable=True)
revoked_at = Column(DateTime, nullable=True) # Set when is_active=False
# Relationship
user = relationship("User", back_populates="api_keys")
# Indexes
__table_args__ = (
Index("ix_api_keys_prefix", "key_prefix"),
Index("ix_api_keys_hash", "key_hash"),
)
def to_dict(self) -> dict:
return {
"id": self.id,
"name": self.name,
"key_prefix": self.key_prefix,
"is_active": self.is_active,
"scopes": self.scopes,
"last_used_at": self.last_used_at.isoformat()
if self.last_used_at
else None,
"usage_count": self.usage_count,
"created_at": self.created_at.isoformat() if self.created_at else None,
"expires_at": self.expires_at.isoformat() if self.expires_at else None,
}
class UsageLog(Base):
"""Daily usage aggregation for billing and analytics"""
__tablename__ = "usage_logs"
id = Column(String(36), primary_key=True, default=generate_uuid)
user_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
# Date (for daily aggregation)
date = Column(DateTime, nullable=False, index=True)
# Aggregated counts
documents_count = Column(Integer, default=0)
pages_count = Column(Integer, default=0)
characters_count = Column(BigInteger, default=0)
api_calls_count = Column(Integer, default=0)
# By provider breakdown (JSON)
provider_breakdown = Column(JSON, default=dict)
# Indexes
__table_args__ = (Index("ix_usage_logs_user_date", "user_id", "date", unique=True),)
class PaymentHistory(Base):
"""Payment and invoice history"""
__tablename__ = "payment_history"
id = Column(String(36), primary_key=True, default=generate_uuid)
user_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
# Stripe info
stripe_payment_intent_id = Column(String(255), nullable=True)
stripe_invoice_id = Column(String(255), nullable=True)
# Payment details
amount_cents = Column(Integer, nullable=False)
currency = Column(String(3), default="usd")
payment_type = Column(String(50), nullable=False) # subscription, credits, one_time
status = Column(String(20), nullable=False) # succeeded, failed, pending, refunded
# Description
description = Column(String(255), nullable=True)
# Timestamps
created_at = Column(DateTime, default=_utcnow)
# Indexes
__table_args__ = (Index("ix_payment_history_user", "user_id", "created_at"),)
class Glossary(Base):
"""User's glossary containing source->target term pairs.
Story 3.9: Glossaires - Endpoint CRUD
"""
__tablename__ = "glossaries"
id = Column(String(36), primary_key=True, default=generate_uuid)
user_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
# Optional: workspace this glossary is shared with (owner still administers it)
workspace_id = Column(String(36), nullable=True)
name = Column(String(255), nullable=False)
source_language = Column(String(10), nullable=False, default="fr")
target_language = Column(String(10), nullable=True, default="en")
template_id = Column(String(50), nullable=True)
created_at = Column(DateTime, default=_utcnow)
updated_at = Column(DateTime, default=_utcnow, onupdate=_utcnow)
# Relationship
terms = relationship(
"GlossaryTerm", back_populates="glossary", cascade="all, delete-orphan"
)
# Indexes
__table_args__ = (
Index("ix_glossaries_user_id", "user_id"),
Index("ix_glossaries_template_id", "template_id"),
Index("ix_glossaries_workspace", "workspace_id"),
)
def to_dict(self) -> dict:
return {
"id": self.id,
"user_id": self.user_id,
"workspace_id": self.workspace_id,
"name": self.name,
"source_language": self.source_language,
"target_language": self.target_language,
"template_id": self.template_id,
"terms": [term.to_dict() for term in self.terms] if self.terms else [],
"created_at": self.created_at.isoformat() if self.created_at else None,
"updated_at": self.updated_at.isoformat() if self.updated_at else None,
}
class GlossaryTerm(Base):
"""Single term pair in a glossary.
Story 3.9: Glossaires - Endpoint CRUD
"""
__tablename__ = "glossary_terms"
id = Column(String(36), primary_key=True, default=generate_uuid)
glossary_id = Column(
String(36), ForeignKey("glossaries.id", ondelete="CASCADE"), nullable=False
)
source = Column(String(500), nullable=False)
target = Column(String(500), nullable=False)
translations = Column(JSON, nullable=True, default=dict)
created_at = Column(DateTime, default=_utcnow)
# Relationship
glossary = relationship("Glossary", back_populates="terms")
# Indexes
__table_args__ = (Index("ix_glossary_terms_glossary_id", "glossary_id"),)
def to_dict(self) -> dict:
return {
"id": self.id,
"source": self.source,
"target": self.target,
"translations": self.translations or {},
"created_at": self.created_at.isoformat() if self.created_at else None,
}
class CustomPrompt(Base):
"""User's custom prompts for LLM translation context.
Story 3.11: Custom Prompts - Endpoint CRUD
"""
__tablename__ = "custom_prompts"
id = Column(String(36), primary_key=True, default=generate_uuid)
user_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
name = Column(String(255), nullable=False)
content = Column(Text, nullable=False)
created_at = Column(DateTime, default=_utcnow)
updated_at = Column(DateTime, default=_utcnow, onupdate=_utcnow)
# Indexes
__table_args__ = (Index("ix_custom_prompts_user_id", "user_id"),)
def to_dict(self) -> dict:
return {
"id": self.id,
"user_id": self.user_id,
"name": self.name,
"content": self.content,
"created_at": self.created_at.isoformat() if self.created_at else None,
"updated_at": self.updated_at.isoformat() if self.updated_at else None,
}
class TranslationSegment(Base):
"""One translatable unit of a finished job (review foundation).
Segments persist (source, translation) pairs per job so reviewers can
edit/approve them side-by-side, rebuild the document with the reviewed
text, and export/import XLIFF.
"""
__tablename__ = "translation_segments"
id = Column(String(36), primary_key=True, default=generate_uuid)
job_id = Column(String(64), nullable=False)
user_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
segment_index = Column(Integer, nullable=False, default=0)
source_text = Column(Text, nullable=False)
translated_text = Column(Text, nullable=False, default="")
# pending | approved | edited
status = Column(String(20), nullable=False, default="pending")
# Reviewer's edit (used for rebuild when status == "edited")
reviewed_text = Column(Text, nullable=True)
is_heading = Column(Boolean, nullable=False, default=False)
created_at = Column(DateTime, default=_utcnow)
updated_at = Column(DateTime, default=_utcnow, onupdate=_utcnow)
__table_args__ = (
Index("ix_segments_job_index", "job_id", "segment_index"),
Index("ix_segments_user", "user_id"),
Index("ix_segments_status", "job_id", "status"),
)
def to_dict(self) -> dict:
return {
"id": self.id,
"job_id": self.job_id,
"segment_index": self.segment_index,
"source_text": self.source_text,
"translated_text": self.translated_text,
"status": self.status,
"reviewed_text": self.reviewed_text,
"is_heading": self.is_heading,
"created_at": self.created_at.isoformat() if self.created_at else None,
"updated_at": self.updated_at.isoformat() if self.updated_at else None,
}
class Workspace(Base):
"""Team workspace: shared glossaries + seat-based membership.
Seats are enforced from the owner's plan (Business = 5 seats).
"""
__tablename__ = "workspaces"
id = Column(String(36), primary_key=True, default=generate_uuid)
name = Column(String(255), nullable=False)
owner_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
created_at = Column(DateTime, default=_utcnow)
members = relationship(
"WorkspaceMember", back_populates="workspace", cascade="all, delete-orphan"
)
__table_args__ = (Index("ix_workspaces_owner", "owner_id"),)
def to_dict(self) -> dict:
return {
"id": self.id,
"name": self.name,
"owner_id": self.owner_id,
"member_count": len(self.members) if self.members else 0,
"members": [m.to_dict() for m in self.members] if self.members else [],
"created_at": self.created_at.isoformat() if self.created_at else None,
}
class WorkspaceMember(Base):
"""Membership row linking a user to a workspace with a role."""
__tablename__ = "workspace_members"
id = Column(String(36), primary_key=True, default=generate_uuid)
workspace_id = Column(
String(36), ForeignKey("workspaces.id", ondelete="CASCADE"), nullable=False
)
user_id = Column(
String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False
)
# owner | admin | member
role = Column(String(20), nullable=False, default="member")
created_at = Column(DateTime, default=_utcnow)
workspace = relationship("Workspace", back_populates="members")
__table_args__ = (
Index("ix_workspace_members_workspace", "workspace_id"),
Index("ix_workspace_members_user", "user_id"),
# UniqueConstraint imported lazily below if needed; enforced in app layer
)
def to_dict(self) -> dict:
return {
"id": self.id,
"workspace_id": self.workspace_id,
"user_id": self.user_id,
"role": self.role,
"created_at": self.created_at.isoformat() if self.created_at else None,
}