path: root/routers/imports.py

import csv
import io
from datetime import datetime

import httpx
from fastapi import APIRouter, Depends, File, Request, UploadFile
from fastapi.responses import RedirectResponse
from fastapi.templating import Jinja2Templates
from sqlalchemy import or_, text
from sqlalchemy.orm import Session

from database import get_db
from models import Film
from services.tmdb import TMDBNotConfiguredError, apply_metadata_to_film, find_movie

router = APIRouter(tags=["imports"])
templates = Jinja2Templates(directory="templates")


def _value(row: dict, *names: str) -> str:
    for name in names:
        value = row.get(name)
        if value:
            return value.strip()
    return ""


def _parse_int(value: str) -> int | None:
    value = value.strip()
    if not value:
        return None
    try:
        return int(value)
    except ValueError:
        return None


def _parse_bool(value: str) -> bool:
    return value.strip().lower() in {"1", "true", "yes", "y", "on", "rewatched"}


def _parse_explicit_stars(value: str) -> int | None:
    value = value.strip()
    if not value:
        return None
    try:
        stars = int(value)
    except ValueError:
        return None
    return stars if stars in {0, 1, 2, 3} else None


def _parse_rating_to_stars(value: str) -> int:
    value = value.strip()
    if not value:
        return 0
    try:
        rating = float(value)
    except ValueError:
        return 0

    if rating < 0 or rating > 5:
        return 0
    if rating >= 5:
        return 3
    if rating >= 4.5:
        return 2
    if rating >= 3.5:
        return 1
    return 0


def _stars_from_row(row: dict) -> int:
    explicit_stars = _parse_explicit_stars(_value(row, "Stars"))
    if explicit_stars is not None:
        return explicit_stars
    return _parse_rating_to_stars(_value(row, "Rating"))


def _parse_shelf(value: str) -> str:
    shelf = value.strip().lower()
    return shelf if shelf in {"diary", "queue", "abandoned"} else "diary"


def _parse_date(value: str):
    value = value.strip()
    if not value:
        return None
    for date_format in ("%Y-%m-%d", "%m/%d/%Y", "%d/%m/%Y"):
        try:
            return datetime.strptime(value, date_format).date()
        except ValueError:
            continue
    return None


def _normalize_title(value: str | None) -> str:
    return (value or "").casefold().strip()


def _date_key(value) -> str:
    return value.isoformat() if value else ""


def _duplicate_keys_for_film(film: Film) -> set[tuple]:
    keys = set()
    watched_date = _date_key(film.date_watched)
    if film.tmdb_id:
        keys.add(("tmdb", str(film.tmdb_id), watched_date))

    title = _normalize_title(film.title)
    if title:
        keys.add(("title", title, str(film.year or ""), watched_date))

    return keys


def _existing_duplicate_keys(db: Session) -> set[tuple]:
    keys = set()
    for film in db.query(Film).all():
        keys.update(_duplicate_keys_for_film(film))
    return keys


def _decode_csv(content: bytes) -> str:
    for encoding in ("utf-8-sig", "utf-8", "latin-1"):
        try:
            return content.decode(encoding)
        except UnicodeDecodeError:
            continue
    return content.decode("utf-8", errors="replace")


def _film_from_row(row: dict, shelf: str) -> Film | None:
    title = _value(row, "Name", "Title")
    if not title:
        return None

    is_diary = shelf == "diary"
    return Film(
        title=title,
        original_title=_value(row, "Original Title") or None,
        director=_value(row, "Director") or None,
        year=_parse_int(_value(row, "Year")),
        country=_value(row, "Country") or None,
        language=_value(row, "Language") or None,
        runtime=_parse_int(_value(row, "Runtime", "Runtime (mins)")),
        date_watched=_parse_date(_value(row, "Watched Date", "Date")) if is_diary else None,
        rewatch=_parse_bool(_value(row, "Rewatch")) if is_diary else False,
        rewatch_count=(_parse_int(_value(row, "Rewatch Count")) or 0) if is_diary else 0,
        stars=_stars_from_row(row) if is_diary else 0,
        watched_with=_value(row, "Watched With", "Watched_with") or None,
        shelf=shelf,
        how_found=_value(row, "How Found", "How_found") or None,
        context=_value(row, "Context", "Tags") or None,
        notes=_value(row, "Review", "Notes") or None,
        poster_url=_value(row, "Poster URL", "Poster") or None,
        tmdb_id=_parse_int(_value(row, "TMDB ID", "tmdb_id")),
    )


async def _dedupe_enrich_and_save(
    films: list[Film],
    db: Session,
) -> tuple[list[Film], int, int]:
    skipped = 0
    existing_keys = _existing_duplicate_keys(db)
    pending_keys = set()
    deduped_before_enrichment = []

    for film in films:
        film_keys = _duplicate_keys_for_film(film)
        if film_keys & (existing_keys | pending_keys):
            skipped += 1
            continue

        deduped_before_enrichment.append(film)
        pending_keys.update(film_keys)

    enriched = await _enrich_films_from_tmdb(deduped_before_enrichment)
    deduped_after_enrichment = []
    final_keys = set(existing_keys)
    for film in deduped_before_enrichment:
        film_keys = _duplicate_keys_for_film(film)
        if film_keys & final_keys:
            skipped += 1
            continue
        deduped_after_enrichment.append(film)
        final_keys.update(film_keys)

    if deduped_after_enrichment:
        db.add_all(deduped_after_enrichment)
        db.commit()

    return deduped_after_enrichment, skipped, enriched


async def _enrich_films_from_tmdb(films: list[Film]) -> int:
    enriched = 0
    if not films:
        return enriched

    async with httpx.AsyncClient(timeout=10.0) as client:
        for film in films:
            if film.poster_url and film.tmdb_id:
                continue
            try:
                metadata = await find_movie(film.title, year=film.year, client=client)
            except TMDBNotConfiguredError:
                return enriched
            except httpx.HTTPError:
                continue
            if metadata and apply_metadata_to_film(film, metadata):
                enriched += 1

    return enriched


@router.get("/import")
def import_page(request: Request):
    return templates.TemplateResponse(
        request=request,
        name="import.html",
        context={"request": request, "error": None, "active_page": "import"},
    )


@router.post("/data/clear")
def clear_data(db: Session = Depends(get_db)):
    deleted = db.query(Film).delete(synchronize_session=False)
    bind = db.get_bind()
    if bind.dialect.name == "sqlite":
        db.execute(text("DELETE FROM sqlite_sequence WHERE name = 'films'"))
    db.commit()
    return RedirectResponse(f"/?cleared={deleted}", status_code=303)


@router.post("/data/clear-duplicates")
def clear_duplicates(db: Session = Depends(get_db)):
    seen_keys = set()
    duplicate_ids = []
    for film in db.query(Film).order_by(Film.id.asc()).all():
        film_keys = _duplicate_keys_for_film(film)
        if film_keys & seen_keys:
            duplicate_ids.append(film.id)
            continue
        seen_keys.update(film_keys)

    deleted = 0
    if duplicate_ids:
        deleted = (
            db.query(Film)
            .filter(Film.id.in_(duplicate_ids))
            .delete(synchronize_session=False)
        )
        db.commit()

    return RedirectResponse(f"/?deduped={deleted}", status_code=303)


@router.post("/data/enrich-posters")
async def enrich_missing_posters(db: Session = Depends(get_db)):
    films = (
        db.query(Film)
        .filter(or_(Film.poster_url.is_(None), Film.poster_url == "", Film.tmdb_id.is_(None)))
        .order_by(Film.year.asc(), Film.title.asc())
        .all()
    )
    enriched = await _enrich_films_from_tmdb(films)
    if enriched:
        db.commit()
    return RedirectResponse(f"/?enriched={enriched}", status_code=303)


@router.post("/import/letterboxd")
async def import_letterboxd(
    request: Request,
    file: UploadFile = File(...),
    db: Session = Depends(get_db),
):
    if not file.filename.lower().endswith(".csv"):
        return templates.TemplateResponse(
            request=request,
            name="import.html",
            context={"request": request, "error": "Upload a CSV file.", "active_page": "import"},
            status_code=400,
        )

    content = await file.read()
    reader = csv.DictReader(io.StringIO(_decode_csv(content)))
    films = []

    for row in reader:
        if not any(value.strip() for value in row.values() if value):
            continue

        film = _film_from_row(row, _parse_shelf(_value(row, "Shelf")))
        if film is None:
            continue
        films.append(film)

    films, skipped, enriched = await _dedupe_enrich_and_save(films, db)

    return RedirectResponse(
        f"/?imported={len(films)}&skipped={skipped}&enriched={enriched}",
        status_code=303,
    )


@router.post("/import/watchlist")
async def import_watchlist(
    request: Request,
    file: UploadFile = File(...),
    db: Session = Depends(get_db),
):
    if not file.filename.lower().endswith(".csv"):
        return templates.TemplateResponse(
            request=request,
            name="import.html",
            context={"request": request, "error": "Upload a CSV file.", "active_page": "import"},
            status_code=400,
        )

    content = await file.read()
    reader = csv.DictReader(io.StringIO(_decode_csv(content)))
    films = []

    for row in reader:
        if not any(value.strip() for value in row.values() if value):
            continue

        film = _film_from_row(row, "queue")
        if film is None:
            continue

        films.append(film)

    films, skipped, enriched = await _dedupe_enrich_and_save(films, db)

    return RedirectResponse(
        f"/queue?imported={len(films)}&skipped={skipped}&enriched={enriched}",
        status_code=303,
    )