utils.py

"""
Utility functions for Finn.no scraper.
"""

import re
from typing import Dict, Optional
from urllib.parse import urlencode, urljoin, urlparse

from config import BASE_URL, SEARCH_URL, FUEL_TYPES, TRANSMISSION_TYPES


def build_search_url(
    page: int = 1,
    min_price: Optional[int] = None,
    max_price: Optional[int] = None,
    min_year: Optional[int] = None,
    max_year: Optional[int] = None,
    min_mileage: Optional[int] = None,
    max_mileage: Optional[int] = None,
    fuel_type: Optional[str] = None,
    sort_by: str = "PUBLISHED_DESC"
) -> str:
    """Build a Finn.no search URL with filters."""
    params = {}

    # Add pagination
    if page > 1:
        params["page"] = page

    # Add price filters
    if min_price is not None:
        params["price_from"] = min_price
    if max_price is not None:
        params["price_to"] = max_price

    # Add year filters
    if min_year is not None:
        params["year_from"] = min_year
    if max_year is not None:
        params["year_to"] = max_year

    # Add mileage filters
    if min_mileage is not None:
        params["mileage_from"] = min_mileage
    if max_mileage is not None:
        params["mileage_to"] = max_mileage

    # Add fuel type filter
    if fuel_type:
        params["fuel"] = fuel_type

    # Add sorting
    if sort_by:
        params["sort"] = sort_by

    if params:
        query_string = urlencode(params, doseq=True)
        return f"{SEARCH_URL}?{query_string}"
    return SEARCH_URL


def parse_listing_text(text: str) -> Dict[str, str]:
    """
    Parse listing text to extract details.

    Example text format:
    "Volvo V70 D2 / Momentum / Manuell / Meget pen / Cruise / Dab
     D2 / Momentum / Manuell / Meget pen / Cruise / Dab
     2014 ∙ 210 000 km ∙ Diesel ∙ Manuell
     109 999 kr
     Stange ∙ WIIK AUTO AS
     Forhandler
     1 dg."
    """
    result = {
        "title": "",
        "year": "",
        "mileage": "",
        "fuel_type": "",
        "transmission": "",
        "price": "",
        "location": "",
        "seller_name": "",
        "seller_type": "",
        "time_posted": ""
    }

    if not text:
        return result

    lines = [line.strip() for line in text.split('\n') if line.strip()]

    for line in lines:
        # Extract price (e.g., "109 999 kr", "265 000 kr")
        price_match = re.search(r'([\d\s]+)\s*kr\s*$', line)
        if price_match and not result["price"]:
            result["price"] = price_match.group(1).replace(" ", " ").strip() + " kr"
            continue

        # Extract year, mileage, fuel, transmission (e.g., "2014 ∙ 210 000 km ∙ Diesel ∙ Manuell")
        if '∙' in line and 'km' in line:
            parts = [p.strip() for p in line.split('∙')]
            for part in parts:
                # Year (4 digits)
                if re.match(r'^\d{4}$', part):
                    result["year"] = part
                # Mileage
                elif 'km' in part:
                    result["mileage"] = part
                # Fuel type
                elif part in FUEL_TYPES or part in ["Bensin", "Diesel", "El", "Hybrid", "Plug-in Bensin", "Plug-in Diesel"]:
                    result["fuel_type"] = part
                # Transmission
                elif part in ["Manuell", "Automat"]:
                    result["transmission"] = part
                # Electric range (e.g., "450 km rekkevidde")
                elif 'rekkevidde' in part:
                    result["mileage"] = part if not result["mileage"] else result["mileage"]
            continue

        # Extract seller type
        if line in ["Forhandler", "Privat", "Merkeforhandler"]:
            result["seller_type"] = line
            continue

        # Extract time posted (e.g., "1 dg.", "3 dg.", "Nå", "1 min.")
        if re.match(r'^(\d+\s*(dg|min|time|dag)|Nå)\.?$', line):
            result["time_posted"] = line
            continue

        # Extract location and seller name (e.g., "Stange ∙ WIIK AUTO AS")
        if '∙' in line and 'km' not in line and 'kr' not in line:
            parts = [p.strip() for p in line.split('∙')]
            if len(parts) >= 2:
                result["location"] = parts[0]
                result["seller_name"] = parts[1]
            elif len(parts) == 1:
                # Could be just location
                result["location"] = parts[0]
            continue

    # First non-empty line is usually the title
    for line in lines:
        if line and not any(x in line for x in ['kr', '∙', 'Forhandler', 'Privat', 'dg.', 'min.', 'Nå']):
            if not line.startswith('Betalt'):
                result["title"] = line
                break

    return result


def extract_listing_id(url: str) -> str:
    """Extract listing ID from Finn.no URL."""
    # URL format: https://www.finn.no/mobility/item/436201670
    match = re.search(r'/mobility/item/(\d+)', url)
    return match.group(1) if match else ""


def make_absolute_url(url: str) -> str:
    """Convert relative URL to absolute URL."""
    if url.startswith("http"):
        return url
    return urljoin(BASE_URL, url)


def clean_text(text: str) -> str:
    """Clean and normalize text."""
    if not text:
        return ""
    # Remove extra whitespace
    return " ".join(text.split())


def translate_fuel_type(norwegian: str) -> str:
    """Translate Norwegian fuel type to English."""
    return FUEL_TYPES.get(norwegian, norwegian)


def translate_transmission(norwegian: str) -> str:
    """Translate Norwegian transmission type to English."""
    return TRANSMISSION_TYPES.get(norwegian, norwegian)