app/models/schemas.py

from __future__ import annotations

import re
from enum import Enum
from typing import Any, Dict, List, Literal, Optional

from pydantic import BaseModel, Field, field_validator, model_validator

from app.core.config import settings

_PAGE_SPEC_RE = re.compile(r"^\s*(\d+(-\d+)?)(\s*,\s*(\d+(-\d+)?))*\s*$")


class ImageFormat(str, Enum):
    PNG = "PNG"
    JPEG = "JPEG"


class JobStatus(str, Enum):
    PENDING = "pending"
    PROCESSING = "processing"
    COMPLETED = "completed"
    FAILED = "failed"


class ConversionParams(BaseModel):
    format: ImageFormat = Field(ImageFormat.PNG)
    dpi: int = Field(settings.DEFAULT_DPI, ge=72, le=settings.MAX_DPI)
    quality: int = Field(85, ge=1, le=100)
    pages: Optional[str] = Field(None)
    grayscale: bool = Field(False)
    transparent_bg: bool = Field(False)
    split_page: Optional[bool] = Field(
        False,
        description=(
            "Controls output granularity. "
            "True: each page → separate image URL. "
            "False (default): all pages stitched into one tall image → single URL."
        ),
    )

    @field_validator("format", mode="before")
    @classmethod
    def normalise_format(cls, v: str) -> str:
        if isinstance(v, str):
            upper = v.upper()
            if upper == "JPG":
                return "JPEG"
            return upper
        return v

    @field_validator("pages")
    @classmethod
    def validate_page_spec_syntax(cls, v: Optional[str]) -> Optional[str]:
        if v is None or v.strip() == "":
            return None
        if not _PAGE_SPEC_RE.match(v):
            raise ValueError(
                f"Invalid page spec '{v}'. Use comma-separated numbers or ranges, "
                "e.g. '1', '1-3', '1,3,5-7'."
            )
        for token in v.split(","):
            token = token.strip()
            if "-" in token:
                parts = token.split("-", 1)
                start, end = int(parts[0]), int(parts[1])
                if start < 1:
                    raise ValueError(f"Page numbers are 1-indexed; got '{token}'")
                if start > end:
                    raise ValueError(f"Invalid range '{token}': start must be <= end")
            else:
                if int(token) < 1:
                    raise ValueError(f"Page numbers are 1-indexed; got '{token}'")
        return v

    @model_validator(mode="after")
    def validate_transparent_bg(self) -> ConversionParams:
        if self.transparent_bg and self.format != ImageFormat.PNG:
            raise ValueError("transparent_bg is only supported for PNG output")
        return self

    def parse_pages(self, total_pages: int) -> List[int]:
        if not self.pages:
            return list(range(total_pages))
        indices: set[int] = set()
        for token in self.pages.split(","):
            token = token.strip()
            if "-" in token:
                parts = token.split("-", 1)
                start, end = int(parts[0]) - 1, int(parts[1]) - 1
                indices.update(range(start, end + 1))
            else:
                indices.add(int(token) - 1)
        valid = sorted(i for i in indices if 0 <= i < total_pages)
        if not valid:
            raise ValueError(
                f"Page spec '{self.pages}' contains no pages within "
                f"the document's {total_pages} page(s)"
            )
        return valid


class PageResult(BaseModel):
    page_number: int
    download_url: str
    width: int
    height: int
    size_bytes: int
    format: str
    file_id: Optional[str] = None
    file_url: Optional[str] = None
    upload_error: Optional[str] = None


class UploadSummary(BaseModel):
    upload_mode: str
    total_uploaded: int
    failed_uploads: int
    files: List["UploadedFileInfo"]


class ConversionResult(BaseModel):
    job_id: str
    status: JobStatus
    total_pages: int
    converted_pages: int
    pages: List[PageResult]
    duration_ms: float
    upload_summary: Optional[UploadSummary] = None


class AsyncJobResponse(BaseModel):
    job_id: str
    status: JobStatus
    message: str
    status_url: str


class JobStatusResponse(BaseModel):
    job_id: str
    status: JobStatus
    progress: float = Field(..., ge=0, le=100)
    result: Optional[ConversionResult] = None
    error: Optional[str] = None


class HealthResponse(BaseModel):
    status: str
    version: str
    environment: str


class ErrorResponse(BaseModel):
    error: str
    request_id: Optional[str] = None


class UploadedFileInfo(BaseModel):
    file_id: str
    file_url: str
    filename: str
    size_bytes: int