reuse query params when exporting to immoweb and allow filtering from available date

crawler/rec/query.py

@@ -1,8 +1,11 @@
 # from diskcache import Cache
 import asyncio
+from dataclasses import dataclass
+from datetime import datetime
 import enum
 from typing import Any
 import aiohttp
+from data_access import Listing
 
 
 class ListingType(enum.StrEnum):
@@ -16,6 +19,57 @@ class FurnishType(enum.StrEnum):
     PART_FURNISHED = "partFurnished"
 
 
+@dataclass(frozen=True)
+class QueryParameters:
+    listing_type: ListingType
+    min_bedrooms: int
+    max_bedrooms: int
+    min_price: int
+    max_price: int
+    district_names: set[str]
+    radius: float = 0
+    page_size: int = 500  # items per page
+    max_days_since_added: int = 30
+    furnish_types: list[FurnishType] | None = None
+    # The values below are not supported by rightmove
+    # hence we apply them after fetching
+    # available from; council tax
+    let_date_available_from: datetime | None = None
+    last_seen_days: int = 14
+
+
+async def filter_listings(
+    listings: list[Listing],
+    query_parameters: QueryParameters,
+) -> list[Listing]:
+    """
+    Filter listings based on the provided query parameters.
+    """
+    filtered_listings = []
+    for listing in listings:
+        if (
+            listing.bedrooms > query_parameters.max_bedrooms
+            or listing.bedrooms < query_parameters.min_bedrooms
+        ):
+            continue
+        if (
+            listing.price < query_parameters.min_price
+            or listing.price > query_parameters.max_price
+        ):
+            continue
+        if listing.last_seen > query_parameters.last_seen_days:
+            continue
+        if (
+            listing.letDateAvailable is not None
+            and query_parameters.let_date_available_from is not None
+            and listing.letDateAvailable < query_parameters.let_date_available_from
+        ):
+            continue
+        filtered_listings.append(listing)
+
+    return filtered_listings
+
+
 headers = {
     "Host": "api.rightmove.co.uk",
     # 'Accept-Encoding': 'gzip, deflate, br',