scoring and analyzsis

finn_eiendom/analysis.py

@@ -19,6 +19,12 @@ def _normalize_description(text: str | None) -> str:
     return text.lower() if text else ""
 
 
+def _is_resale_listing(url: str) -> bool:
+    """True for ordinary resale ads. Project / new-build ads use different URL
+    paths that fetch_ad_details cannot resolve (it builds a /homes/ URL)."""
+    return "/realestate/homes/" in url
+
+
 def _build_ad_summary(
     ad: FinnAd,
     enriched: EiendomUnit | None,
@@ -95,12 +101,17 @@ async def analyze_ad(
             if enriched is not None:
                 cache.save_eiendom_unit(conn, enriched)
 
-    if enriched and enriched.unit_vector:
-        similar_units = cache.get_similar_units(conn, enriched.unit_code, "RECENTLY_SOLD")
-        if not similar_units:
-            similar_units = await eiendom_no.get_similar_units(enriched.unit_vector)
-            if similar_units:
-                cache.save_similar_units(conn, enriched.unit_code, "RECENTLY_SOLD", similar_units)
+    if enriched:
+        # EiendomUnit.unit_vector is NOT populated by get_unit / enrich -- the
+        # field comes back None. Reading enriched.unit_vector directly leaves
+        # this block dead and similar_units permanently empty. Build the vector
+        # from the unit fields instead (fall back to the field if a future
+        # endpoint ever populates it).
+        vector = enriched.unit_vector or eiendom_no.build_unit_vector(enriched)
+        if vector:
+            # No dedicated cache table for similar units (per PRD) -- fetch
+            # fresh each call, consistent with service.get_or_fetch_similar_units.
+            similar_units = await eiendom_no.get_similar_units(vector)
 
     scores = scoring.score_ad(finn_ad, enriched, similar_units)
     categories = scoring.classify_ad(scores)
@@ -120,6 +131,26 @@ async def analyze_ad(
     return result
 
 
+async def _analyze_card(card, conn, *, include_eiendom_no: bool, client) -> dict:
+    """Fetch details + enrich a single search card. Raises on unrecoverable
+    errors; the caller is responsible for catching and skipping."""
+    finn_ad = cache.get_finn_ad(conn, card.finnkode, ttl_hours=FINN_CACHE_TTL_AD_HOURS)
+    if finn_ad is None:
+        finn_ad = await ad_module.fetch_ad_details(card.finnkode, client=client)
+
+    unit_code = None
+    if include_eiendom_no:
+        try:
+            matched_unit = await eiendom_no.search_unit_from_finn_url(card.url)
+            unit_code = matched_unit.unit_code if matched_unit else None
+        except Exception as exc:
+            # A failed unit resolution is non-fatal -- proceed without enrichment.
+            logger.warning("Eiendom.no unit search failed for %s: %s", card.finnkode, exc)
+            unit_code = None
+
+    return await analyze_ad(finn_ad, unit_code=unit_code)
+
+
 async def analyze_search(
     search_url: str,
     max_pages: int = FINN_MAX_SEARCH_PAGES,
@@ -139,21 +170,28 @@ async def analyze_search(
     )
     results = []
     enriched_count = 0
+    skipped_count = 0
 
     if fetch_details:
         for card in cards[:detail_limit]:
-            finn_ad = cache.get_finn_ad(conn, card.finnkode, ttl_hours=FINN_CACHE_TTL_AD_HOURS)
-            if finn_ad is None:
-                finn_ad = await ad_module.fetch_ad_details(card.finnkode, client=client)
-            unit_code = None
-            if include_eiendom_no:
-                try:
-                    matched_unit = await eiendom_no.search_unit_from_finn_url(card.url)
-                    unit_code = matched_unit.unit_code if matched_unit else None
-                except Exception as exc:
-                    logger.warning("Eiendom.no unit search failed: %s", exc)
-                    unit_code = None
-            result = await analyze_ad(finn_ad, unit_code=unit_code)
+            # Project / new-build ads are not resale listings and fetch_ad_details
+            # cannot resolve them -- skip up front rather than 404 mid-run.
+            if not _is_resale_listing(card.url):
+                logger.info("Skipping non-resale card %s (%s)", card.finnkode, card.url)
+                skipped_count += 1
+                continue
+
+            # One bad card (stale finnkode, removed ad, transient network error)
+            # must not abort the whole search -- isolate each card.
+            try:
+                result = await _analyze_card(
+                    card, conn, include_eiendom_no=include_eiendom_no, client=client
+                )
+            except Exception as exc:
+                logger.warning("Skipping card %s: %s", card.finnkode, exc)
+                skipped_count += 1
+                continue
+
             if result.get("eiendom_unit"):
                 enriched_count += 1
             results.append(result)
@@ -166,6 +204,7 @@ async def analyze_search(
         "summary": {
             "total_listings": len(cards),
             "analyzed_listings": len(results),
+            "skipped_listings": skipped_count,
             "eiendom_enriched": enriched_count,
         },
     }