|
| 1 | +import uuid |
| 2 | +from datetime import datetime |
| 3 | +from enum import Enum |
| 4 | +from typing import TYPE_CHECKING, Any, Dict, List, Optional |
| 5 | + |
| 6 | +from pydantic import BaseModel, Field, field_validator |
| 7 | +from sqlalchemy import ( |
| 8 | + JSON, |
| 9 | + Boolean, |
| 10 | + Column, |
| 11 | + DateTime, |
| 12 | +) |
| 13 | +from sqlalchemy import Enum as SQLEnum |
| 14 | +from sqlalchemy import ( |
| 15 | + ForeignKey, |
| 16 | + Integer, |
| 17 | + String, |
| 18 | + Text, |
| 19 | + TypeDecorator, |
| 20 | + func, |
| 21 | +) |
| 22 | +from sqlalchemy.dialects.postgresql import JSONB |
| 23 | +from sqlalchemy.dialects.postgresql import UUID as PG_UUID |
| 24 | +from sqlalchemy.orm import Mapped, mapped_column, relationship |
| 25 | + |
| 26 | +from models.base import Base |
| 27 | + |
| 28 | +if TYPE_CHECKING: |
| 29 | + from models.user import UserTable |
| 30 | + |
| 31 | + |
| 32 | +class UUID(TypeDecorator): |
| 33 | + """ |
| 34 | + Platform-independent UUID type. |
| 35 | +
|
| 36 | + Uses PostgreSQL's UUID type, otherwise uses |
| 37 | + CHAR(32), storing as string. |
| 38 | + """ |
| 39 | + |
| 40 | + impl = PG_UUID |
| 41 | + cache_ok = True |
| 42 | + |
| 43 | + def load_dialect_impl(self, dialect): |
| 44 | + if dialect.name == "postgresql": |
| 45 | + return dialect.type_descriptor(PG_UUID()) |
| 46 | + else: |
| 47 | + return dialect.type_descriptor(String(32)) |
| 48 | + |
| 49 | + def process_bind_param(self, value, dialect): |
| 50 | + if value is None: |
| 51 | + return value |
| 52 | + elif dialect.name == "postgresql": |
| 53 | + return str(value) |
| 54 | + else: |
| 55 | + if not isinstance(value, uuid.UUID): |
| 56 | + return "%.32x" % uuid.UUID(value).int |
| 57 | + else: |
| 58 | + # hexstring |
| 59 | + return "%.32x" % value.int |
| 60 | + |
| 61 | + def process_result_value(self, value, dialect): |
| 62 | + if value is None: |
| 63 | + return value |
| 64 | + else: |
| 65 | + if not isinstance(value, uuid.UUID): |
| 66 | + value = uuid.UUID(value) |
| 67 | + return value |
| 68 | + |
| 69 | + |
| 70 | +class CrossDatabaseJSON(TypeDecorator): |
| 71 | + """ |
| 72 | + Platform-independent JSON type. |
| 73 | +
|
| 74 | + Uses PostgreSQL's JSONB type for better performance, |
| 75 | + otherwise uses standard JSON type. |
| 76 | + """ |
| 77 | + |
| 78 | + impl = JSON |
| 79 | + cache_ok = True |
| 80 | + |
| 81 | + def load_dialect_impl(self, dialect): |
| 82 | + if dialect.name == "postgresql": |
| 83 | + return dialect.type_descriptor(JSONB()) |
| 84 | + else: |
| 85 | + return dialect.type_descriptor(JSON()) |
| 86 | + |
| 87 | + |
| 88 | +class ProjectStatusEnum(str, Enum): |
| 89 | + """Project status enumeration""" |
| 90 | + |
| 91 | + UPLOADING = "uploading" |
| 92 | + PROCESSING = "processing" |
| 93 | + READY = "ready" |
| 94 | + ERROR = "error" |
| 95 | + |
| 96 | + |
| 97 | +class ProjectTable(Base): |
| 98 | + """SQLAlchemy Project table model for PostgreSQL""" |
| 99 | + |
| 100 | + __tablename__ = "projects" |
| 101 | + |
| 102 | + id: Mapped[uuid.UUID] = mapped_column(UUID, primary_key=True, default=uuid.uuid4) |
| 103 | + user_id: Mapped[uuid.UUID] = mapped_column( |
| 104 | + UUID, ForeignKey("users.id", ondelete="CASCADE"), nullable=False, index=True |
| 105 | + ) |
| 106 | + name: Mapped[str] = mapped_column(String(255), nullable=False) |
| 107 | + description = Column(Text, nullable=True) |
| 108 | + csv_filename: Mapped[str] = mapped_column(String(255), nullable=False) |
| 109 | + csv_path: Mapped[str] = mapped_column(Text, nullable=False) |
| 110 | + row_count: Mapped[int] = mapped_column(Integer, nullable=False, default=0) |
| 111 | + column_count: Mapped[int] = mapped_column(Integer, nullable=False, default=0) |
| 112 | + columns_metadata = Column(CrossDatabaseJSON, nullable=True) |
| 113 | + status: Mapped[ProjectStatusEnum] = mapped_column( |
| 114 | + SQLEnum(ProjectStatusEnum), nullable=False, default=ProjectStatusEnum.UPLOADING |
| 115 | + ) |
| 116 | + |
| 117 | + # Timestamps |
| 118 | + created_at: Mapped[datetime] = mapped_column( |
| 119 | + DateTime(timezone=True), server_default=func.now() |
| 120 | + ) |
| 121 | + updated_at: Mapped[datetime] = mapped_column( |
| 122 | + DateTime(timezone=True), server_default=func.now(), onupdate=func.now() |
| 123 | + ) |
| 124 | + |
| 125 | + # Relationships |
| 126 | + user: Mapped["UserTable"] = relationship(back_populates="projects") |
| 127 | + # chat_messages: Mapped[List["ChatMessageTable"]] = relationship( |
| 128 | + # back_populates="project", cascade="all, delete-orphan" |
| 129 | + # ) |
| 130 | + |
| 131 | + def __repr__(self): |
| 132 | + return f"<Project(id={self.id}, name='{self.name}', user_id={self.user_id}, status='{self.status}')>" |
| 133 | + |
| 134 | + |
| 135 | +# Pydantic models for API validation and serialization |
| 136 | + |
| 137 | + |
| 138 | +class ColumnMetadata(BaseModel): |
| 139 | + """Column metadata model""" |
| 140 | + |
| 141 | + name: str |
| 142 | + type: str |
| 143 | + nullable: bool = True |
| 144 | + sample_values: List[Any] = Field(default_factory=list) |
| 145 | + unique_count: Optional[int] = None |
| 146 | + min_value: Optional[float] = None |
| 147 | + max_value: Optional[float] = None |
| 148 | + |
| 149 | + class Config: |
| 150 | + from_attributes = True |
| 151 | + |
| 152 | + |
| 153 | +class ProjectBase(BaseModel): |
| 154 | + """Base project model with common fields""" |
| 155 | + |
| 156 | + name: str |
| 157 | + description: Optional[str] = None |
| 158 | + csv_filename: str |
| 159 | + csv_path: str |
| 160 | + row_count: int = 0 |
| 161 | + column_count: int = 0 |
| 162 | + columns_metadata: List[ColumnMetadata] = Field(default_factory=list) |
| 163 | + status: ProjectStatusEnum = ProjectStatusEnum.UPLOADING |
| 164 | + |
| 165 | + class Config: |
| 166 | + from_attributes = True |
| 167 | + |
| 168 | + |
| 169 | +class ProjectCreate(BaseModel): |
| 170 | + """Project creation model""" |
| 171 | + |
| 172 | + name: str |
| 173 | + description: Optional[str] = None |
| 174 | + |
| 175 | + @field_validator("name") |
| 176 | + @classmethod |
| 177 | + def validate_name(cls, v): |
| 178 | + if not v or not v.strip(): |
| 179 | + raise ValueError("Project name cannot be empty") |
| 180 | + if len(v.strip()) > 255: |
| 181 | + raise ValueError("Project name cannot exceed 255 characters") |
| 182 | + return v.strip() |
| 183 | + |
| 184 | + @field_validator("description") |
| 185 | + @classmethod |
| 186 | + def validate_description(cls, v): |
| 187 | + if v is not None and len(v.strip()) > 1000: |
| 188 | + raise ValueError("Description cannot exceed 1000 characters") |
| 189 | + return v.strip() if v else None |
| 190 | + |
| 191 | + class Config: |
| 192 | + from_attributes = True |
| 193 | + |
| 194 | + |
| 195 | +class ProjectUpdate(BaseModel): |
| 196 | + """Project update model""" |
| 197 | + |
| 198 | + name: Optional[str] = None |
| 199 | + description: Optional[str] = None |
| 200 | + csv_filename: Optional[str] = None |
| 201 | + csv_path: Optional[str] = None |
| 202 | + row_count: Optional[int] = None |
| 203 | + column_count: Optional[int] = None |
| 204 | + columns_metadata: Optional[List[ColumnMetadata]] = None |
| 205 | + status: Optional[ProjectStatusEnum] = None |
| 206 | + |
| 207 | + @field_validator("name") |
| 208 | + @classmethod |
| 209 | + def validate_name(cls, v): |
| 210 | + if v is not None: |
| 211 | + if not v or not v.strip(): |
| 212 | + raise ValueError("Project name cannot be empty") |
| 213 | + if len(v.strip()) > 255: |
| 214 | + raise ValueError("Project name cannot exceed 255 characters") |
| 215 | + return v.strip() |
| 216 | + return v |
| 217 | + |
| 218 | + @field_validator("description") |
| 219 | + @classmethod |
| 220 | + def validate_description(cls, v): |
| 221 | + if v is not None and len(v.strip()) > 1000: |
| 222 | + raise ValueError("Description cannot exceed 1000 characters") |
| 223 | + return v.strip() if v else None |
| 224 | + |
| 225 | + @field_validator("row_count", "column_count") |
| 226 | + @classmethod |
| 227 | + def validate_counts(cls, v): |
| 228 | + if v is not None and v < 0: |
| 229 | + raise ValueError("Counts cannot be negative") |
| 230 | + return v |
| 231 | + |
| 232 | + class Config: |
| 233 | + from_attributes = True |
| 234 | + |
| 235 | + |
| 236 | +class ProjectInDB(ProjectBase): |
| 237 | + """Project model as stored in database""" |
| 238 | + |
| 239 | + id: uuid.UUID |
| 240 | + user_id: uuid.UUID |
| 241 | + created_at: datetime |
| 242 | + updated_at: datetime |
| 243 | + |
| 244 | + class Config: |
| 245 | + from_attributes = True |
| 246 | + |
| 247 | + |
| 248 | +class ProjectPublic(BaseModel): |
| 249 | + """Public project model for API responses""" |
| 250 | + |
| 251 | + id: str |
| 252 | + user_id: str |
| 253 | + name: str |
| 254 | + description: Optional[str] = None |
| 255 | + csv_filename: str |
| 256 | + csv_path: str |
| 257 | + row_count: int |
| 258 | + column_count: int |
| 259 | + columns_metadata: List[ColumnMetadata] |
| 260 | + status: ProjectStatusEnum |
| 261 | + created_at: str |
| 262 | + updated_at: str |
| 263 | + |
| 264 | + @classmethod |
| 265 | + def from_db_project(cls, project: ProjectInDB) -> "ProjectPublic": |
| 266 | + """Convert ProjectInDB to ProjectPublic""" |
| 267 | + return cls( |
| 268 | + id=str(project.id), |
| 269 | + user_id=str(project.user_id), |
| 270 | + name=project.name, |
| 271 | + description=project.description, |
| 272 | + csv_filename=project.csv_filename, |
| 273 | + csv_path=project.csv_path, |
| 274 | + row_count=project.row_count, |
| 275 | + column_count=project.column_count, |
| 276 | + columns_metadata=project.columns_metadata, |
| 277 | + status=project.status, |
| 278 | + created_at=project.created_at.isoformat(), |
| 279 | + updated_at=project.updated_at.isoformat(), |
| 280 | + ) |
| 281 | + |
| 282 | + class Config: |
| 283 | + from_attributes = True |
0 commit comments