Add CLAUDE.md project documentation for session context

Provides automatic context loading for new Claude Code sessions,
documenting architecture, filters, sources, and conventions.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

.claude/settings.local.json

@@ -1,31 +0,0 @@
-{
-  "permissions": {
-    "allow": [
-      "WebFetch(domain:github.com)",
-      "WebFetch(domain:www.sreality.cz)",
-      "WebFetch(domain:webscraping.pro)",
-      "WebFetch(domain:raw.githubusercontent.com)",
-      "Bash(python3:*)",
-      "Bash(open:*)",
-      "WebFetch(domain:www.realingo.cz)",
-      "WebFetch(domain:api.realingo.cz)",
-      "Bash(curl:*)",
-      "Bash(grep:*)",
-      "WebFetch(domain:www.realitni-pes.cz)",
-      "WebFetch(domain:www.bezrealitky.cz)",
-      "WebFetch(domain:apify.com)",
-      "WebFetch(domain:www.bezrealitky.com)",
-      "WebFetch(domain:reality.idnes.cz)",
-      "Bash(# Final checks: robots.txt and response time for rate limiting clues curl -s -L -H \"\"User-Agent: Mozilla/5.0 \\(Windows NT 10.0; Win64; x64\\) AppleWebKit/537.36 \\(KHTML, like Gecko\\) Chrome/120.0.0.0 Safari/537.36\"\" \"\"https://reality.idnes.cz/robots.txt\"\")",
-      "WebFetch(domain:www.cityhome.cz)",
-      "WebFetch(domain:www.psn.cz)",
-      "WebFetch(domain:www.city-home.cz)",
-      "WebFetch(domain:psn.cz)",
-      "WebFetch(domain:api.psn.cz)",
-      "Bash(done)",
-      "Bash(# Final summary: count total units across all projects\n# Get the total count from the unitsCountData we already extracted\necho \"\"From unitsCountData on /prodej page:\"\"\necho \"\"  type_id 0 \\(Prodej bytů a ateliérů\\): 146\"\"\necho \"\"  type_id 1 \\(Prodej komerčních nemovitostí\\): 14\"\"\necho \"\"  type_id 2 \\(Pronájem bytů\\): 3\"\"\necho \"\"  type_id 3 \\(Pronájem komerčních nemovitostí\\): 48\"\"\necho \"\"\"\"\necho \"\"Total for-sale projects: 19\"\"\necho \"\"\"\"\necho \"\"Disposition counts from the data:\"\"\npython3 << 'PYEOF'\n# Extract disposition counts from prodej page\nimport re\n\nwith open\\('/tmp/psn_prodej_p1.html', 'r', encoding='utf-8'\\) as f:\n    html = f.read\\(\\)\n\n# Find disposition data\nidx = html.find\\('\\\\\\\\\"disposition\\\\\\\\\":['\\)\nif idx >= 0:\n    chunk = html[idx:idx+2000].replace\\('\\\\\\\\\"', '\"'\\)\n    # Extract name and count pairs\n    import re\n    pairs = re.findall\\(r'\"name\":\"\\([^\"]+\\)\",\"count\":\\(\\\\d+\\)', chunk\\)\n    for name, count in pairs:\n        print\\(f\"  {name}: {count}\"\\)\nPYEOF)",
-      "Bash(ls:*)",
-      "Bash(chmod:*)"
-    ]
-  }
-}

.gitignore

@@ -1,3 +1,8 @@
 .vscode/
 __pycache__/
+.DS_Store
 byty_*.json
+*.json
+*.log
+mapa_bytu.html
+

CLAUDE.md

@@ -0,0 +1,124 @@
+# Maru hledá byt
+
+Projekt pro hledání bytů v Praze. Scrapuje inzeráty ze 7 realitních portálů, filtruje, deduplikuje a generuje interaktivní mapu.
+
+**Jazyk komunikace:** Čeština (uživatelka Marie). Kód a komentáře v kódu jsou mix CZ/EN.
+
+## Architektura
+
+```
+run_all.sh (orchestrátor)
+  ├─ scrape_and_map.py      → byty_sreality.json     (Sreality API)
+  ├─ scrape_bezrealitky.py  → byty_bezrealitky.json  (HTML Apollo cache)
+  ├─ scrape_idnes.py        → byty_idnes.json        (HTML regex)
+  ├─ scrape_psn.py }        → byty_psn.json          (React API + curl)
+  ├─ scrape_cityhome.py }   → byty_cityhome.json     (HTML tabulky)
+  ├─ scrape_bazos.py        → byty_bazos.json        (HTML regex)
+  └─ scrape_realingo.py     → byty_realingo.json     (Next.js __NEXT_DATA__)
+      ↓
+  merge_and_map.py
+      ├─ byty_merged.json (deduplikovaná data)
+      └─ mapa_bytu.html   (Leaflet.js mapa)
+      ↓
+  generate_status.py → status.json + scraper_history.json
+      ↓
+  server.py (port 8080) → servíruje mapu + status page + ratings API
+```
+
+## Filtry (společné všem scraperům)
+
+| Parametr | Hodnota | Poznámka |
+|----------|---------|----------|
+| Max cena | 13.5M Kč (Sreality/Realingo/Bezrealitky/iDNES), 14M Kč (PSN/CityHome/Bazoš) | Rozdíl je záměrný |
+| Min plocha | 69 m² | |
+| Min patro | 2. NP | 2. NP se na mapě označí varováním |
+| Dispozice | 3+kk, 3+1, 4+kk, 4+1, 5+kk, 5+1, 6+ | |
+| Region | Praha | |
+| Vyloučit | panelové domy, sídliště | regex v popisu/polích |
+
+## Klíčové soubory
+
+- **scrape_and_map.py** — Sreality scraper + `generate_map()` funkce (sdílená, generuje HTML mapu)
+- **merge_and_map.py** — sloučí 7 JSON zdrojů, deduplikuje (klíč: ulice + cena + plocha), volá `generate_map()`
+- **scraper_stats.py** — utility: `validate_listing()` (validace povinných polí + GPS bounds) a `write_stats()`
+- **generate_status.py** — generuje status.json a scraper_history.json z výstupů scraperů
+- **server.py** — HTTP server (port 8080), endpointy: `/mapa_bytu.html`, `/scrapers-status`, `/api/ratings`, `/api/status`
+- **run_all.sh** — orchestrátor, spouští scrapery postupně (PSN+CityHome paralelně), pak merge + status
+
+## Mapa (mapa_bytu.html)
+
+- Leaflet.js + CARTO tiles
+- Barvy markerů podle ceny/m² (modrá < 110k → červená > 165k, šedá = neuvedeno)
+- PSN/CityHome = srdíčkové markery (❤️)
+- Nové inzeráty (≤ 1 den) = žlutý badge "NEW"
+- Zamítnuté = zprůhledněné + 🚫 SVG overlay
+- Oblíbené = hvězdička (⭐)
+- Filtry: patro, max cena (input, default 13.5M, max 14M), datum přidání, skrýt zamítnuté, klik na cenový pás
+- Ratings uložené v localStorage + sync na server `/api/ratings`
+
+## Barvy zdrojů na mapě
+
+```python
+source_colors = {
+    "sreality": "#1976D2",    # modrá
+    "realingo": "#00897B",    # teal
+    "bezrealitky": "#E91E63", # růžová
+    "idnes": "#FF6F00",       # oranžová
+    "psn": "#D32F2F",         # červená
+    "cityhome": "#D32F2F",    # červená
+    "bazos": "#7B1FA2",       # fialová
+}
+```
+
+## Deduplikace (merge_and_map.py)
+
+- Klíč: `normalize_street(locality) + price + area`
+- Normalizace ulice: první část před čárkou, lowercase, odstranění diakritiky, jen alfanumerické znaky
+- PSN a CityHome mají prioritu (načtou se první)
+
+## Vývoj
+
+- **Git remote:** `https://gitea.home.hrajfrisbee.cz/littlemeat/maru-hleda-byt.git`
+- **Gitea API token:** uložen v `.claude/settings.local.json`
+- **Python 3.9+** kompatibilita (`from __future__ import annotations`)
+- **Žádné pip závislosti** — jen stdlib (urllib, json, re, logging, pathlib, subprocess)
+- **Docker:** `build/Dockerfile` (python:3.13-alpine), cron každé 4 hodiny
+- Generované soubory (`byty_*.json`, `mapa_bytu.html`, `*.log`) jsou v `.gitignore`
+
+## Typické úlohy
+
+```bash
+# Rychlý test scraperu
+python3 scrape_bazos.py --max-pages 1 --max-properties 5 --log-level DEBUG
+
+# Lokální validace (všechny scrapery s limity)
+make validation-local
+
+# Vygenerovat mapu z existujících dat
+python3 merge_and_map.py
+
+# Spustit server
+python3 server.py  # nebo: make serve
+
+# Plný scrape
+./run_all.sh
+```
+
+## Pořadí scraperů v run_all.sh
+
+1. Sreality
+2. Bezrealitky
+3. iDNES
+4. PSN + CityHome (paralelně)
+5. Bazoš
+6. Realingo (poslední — uživatelka ho nemá ráda)
+7. Merge + mapa
+8. Status generování
+
+## Konvence
+
+- Commit messages v angličtině, PR popis v angličtině
+- Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
+- PRy přes Gitea API (viz create_pr.sh pattern v historii)
+- Nové scrapery kopírují vzor z `scrape_bezrealitky.py`
+- Každý scraper má argparse s `--max-pages`, `--max-properties`, `--log-level`

Makefile

@@ -3,9 +3,13 @@ CONTAINER_NAME    := maru-hleda-byt
 VOLUME_NAME       := maru-hleda-byt-data
 VALIDATION_CONTAINER := maru-hleda-byt-validation
 VALIDATION_VOLUME  := maru-hleda-byt-validation-data
+DEBUG_CONTAINER   := maru-hleda-byt-debug
+DEBUG_VOLUME      := maru-hleda-byt-debug-data
+DEBUG_PORT        ?= 8082
 PORT              := 8080
+SERVER_PORT       ?= 8080
 
-.PHONY: build run stop logs scrape restart clean help validation validation-local validation-stop validation-local-debug
+.PHONY: build run stop logs scrape restart clean help serve validation validation-local validation-stop validation-local-debug debug debug-stop
 
 help:
 	@echo "Available targets:"
@@ -20,6 +24,9 @@ help:
 	@echo "  validation-local-debug - Run validation locally with DEBUG logging"
 	@echo "  restart             - Restart the container (stop and run again)"
 	@echo "  clean               - Stop container and remove the Docker image"
+	@echo "  serve               - Start server.py locally on port 8080"
+	@echo "  debug               - Build and run debug Docker container with limited scrape (port $(DEBUG_PORT))"
+	@echo "  debug-stop          - Stop and remove the debug Docker container"
 	@echo "  help                - Show this help message"
 
 build:
@@ -59,6 +66,27 @@ validation-stop:
 	@docker rm $(VALIDATION_CONTAINER) 2>/dev/null || true
 	@echo "Validation container stopped and removed"
 
+debug: build
+	@docker stop $(DEBUG_CONTAINER) 2>/dev/null || true
+	@docker rm $(DEBUG_CONTAINER) 2>/dev/null || true
+	docker run -d --name $(DEBUG_CONTAINER) \
+		-p $(DEBUG_PORT):8080 \
+		-v $(DEBUG_VOLUME):/app/data \
+		-e LOG_LEVEL=DEBUG \
+		$(IMAGE_NAME)
+	@sleep 2
+	docker exec $(DEBUG_CONTAINER) bash /app/run_all.sh --max-pages 1 --max-properties 10
+	@echo "Debug app at http://localhost:$(DEBUG_PORT)/mapa_bytu.html"
+	@echo "Debug status at http://localhost:$(DEBUG_PORT)/scrapers-status"
+
+debug-stop:
+	@docker stop $(DEBUG_CONTAINER) 2>/dev/null || true
+	@docker rm $(DEBUG_CONTAINER) 2>/dev/null || true
+	@echo "Debug container stopped and removed"
+
+serve:
+	DATA_DIR=. SERVER_PORT=$(SERVER_PORT) python3 server.py
+
 validation-local:
 	./run_all.sh --max-pages 1 --max-properties 10
 

README.md

@@ -83,10 +83,6 @@ Merges all `byty_*.json` files into `byty_merged.json` and generates `mapa_bytu.
 
 **Deduplication logic:** Two listings are considered duplicates if they share the same normalized street name + price + area. PSN and CityHome have priority during dedup (loaded first), so their listings are kept over duplicates from other portals.
 
-### `regen_map.py`
-
-Regenerates the map from existing `byty_sreality.json` data without re-scraping. Fetches missing area values from the Sreality API, fixes URLs, and re-applies the area filter. Useful for tweaking map output after data has already been collected.
-
 ## Interactive map (`mapa_bytu.html`)
 
 The generated map is a standalone HTML file using Leaflet.js with CARTO basemap tiles. Features:
@@ -151,7 +147,7 @@ The project includes a Docker setup for unattended operation with a cron-based s
 │  PID 1: python3 -m http.server :8080    │
 │         serves /app/data/               │
 │                                         │
-│  crond:  runs run_all.sh at 06:00/18:00 │
+│  crond:  runs run_all.sh every 4 hours  │
 │          Europe/Prague timezone          │
 │                                         │
 │  /app/        -- scripts (.py, .sh)     │
@@ -160,7 +156,7 @@ The project includes a Docker setup for unattended operation with a cron-based s
 └─────────────────────────────────────────┘
 ```
 
-On startup, the HTTP server starts immediately. The initial scrape runs in the background. Subsequent cron runs update data in-place twice daily at 06:00 and 18:00 CET/CEST.
+On startup, the HTTP server starts immediately. The initial scrape runs in the background. Subsequent cron runs update data in-place every 4 hours.
 
 ### Quick start
 
@@ -201,14 +197,13 @@ Validation targets run scrapers with `--max-pages 1 --max-properties 10` for a f
 ├── scrape_psn.py           # PSN scraper
 ├── scrape_cityhome.py      # CityHome scraper
 ├── merge_and_map.py        # Merge all sources + generate final map
-├── regen_map.py            # Regenerate map from cached Sreality data
 ├── run_all.sh              # Orchestrator script (runs all scrapers + merge)
 ├── mapa_bytu.html          # Generated interactive map (output)
 ├── Makefile                # Docker management + validation shortcuts
 ├── build/
 │   ├── Dockerfile          # Container image definition (python:3.13-alpine)
 │   ├── entrypoint.sh       # Container entrypoint (HTTP server + cron + initial scrape)
-│   ├── crontab             # Cron schedule (06:00 and 18:00 CET)
+│   ├── crontab             # Cron schedule (every 4 hours)
 │   └── CONTAINER.md        # Container-specific documentation
 └── .gitignore              # Ignores byty_*.json, __pycache__, .vscode
 ```

build/Dockerfile

@@ -5,12 +5,14 @@ RUN apk add --no-cache curl bash tzdata \
     && echo "Europe/Prague" > /etc/timezone
 
 ENV PYTHONUNBUFFERED=1
+ENV DATA_DIR=/app/data
 
 WORKDIR /app
 
 COPY scrape_and_map.py scrape_realingo.py scrape_bezrealitky.py \
      scrape_idnes.py scrape_psn.py scrape_cityhome.py \
-     merge_and_map.py regen_map.py run_all.sh ratings_server.py ./
+     merge_and_map.py generate_status.py scraper_stats.py \
+     run_all.sh server.py ./
 
 COPY build/crontab /etc/crontabs/root
 COPY build/entrypoint.sh /entrypoint.sh
@@ -18,7 +20,7 @@ RUN chmod +x /entrypoint.sh run_all.sh
 
 RUN mkdir -p /app/data
 
-EXPOSE 8080 8081
+EXPOSE 8080
 
 HEALTHCHECK --interval=60s --timeout=5s --start-period=300s \
     CMD wget -q -O /dev/null http://localhost:8080/ || exit 1

build/crontab

@@ -1 +1 @@
-0 6,18 * * * cd /app && bash /app/run_all.sh >> /proc/1/fd/1 2>> /proc/1/fd/2
+0 */4 * * * cd /app && bash /app/run_all.sh >> /proc/1/fd/1 2>> /proc/1/fd/2

build/entrypoint.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 set -euo pipefail
 
-DATA_DIR="/app/data"
+export DATA_DIR="/app/data"
 
 # Create symlinks so scripts (which write to /app/) persist data to the volume
 for f in byty_sreality.json byty_realingo.json byty_bezrealitky.json \
@@ -18,8 +18,5 @@ crond -b -l 2
 echo "[entrypoint] Starting initial scrape in background..."
 bash /app/run_all.sh &
 
-echo "[entrypoint] Starting ratings API server on port 8081..."
+echo "[entrypoint] Starting server on port 8080..."
-DATA_DIR="$DATA_DIR" python3 /app/ratings_server.py &
+exec python3 /app/server.py
-
-echo "[entrypoint] Starting HTTP server on port 8080..."
-exec python3 -m http.server 8080 --directory "$DATA_DIR"

byty_bezrealitky.json

@@ -1,427 +0,0 @@
-[
-  {
-    "hash_id": 990183,
-    "name": "Prodej bytu 3+kk 86 m²",
-    "price": 10385000,
-    "price_formatted": "10 385 000 Kč",
-    "locality": "Ke Tvrzi, Praha - Královice",
-    "lat": 50.0390519,
-    "lon": 14.63862,
-    "disposition": "3+kk",
-    "floor": 2,
-    "area": 86,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/990183-nabidka-prodej-bytu-ke-tvrzi-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 989862,
-    "name": "Prodej bytu 3+kk 73 m²",
-    "price": 12790000,
-    "price_formatted": "12 790 000 Kč",
-    "locality": "Vrázova, Praha - Smíchov",
-    "lat": 50.0711312,
-    "lon": 14.4076652,
-    "disposition": "3+kk",
-    "floor": 3,
-    "area": 73,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/989862-nabidka-prodej-bytu-vrazova-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 981278,
-    "name": "Prodej bytu 3+kk 70 m²",
-    "price": 11890000,
-    "price_formatted": "11 890 000 Kč",
-    "locality": "Argentinská, Praha - Holešovice",
-    "lat": 50.1026043,
-    "lon": 14.4435365,
-    "disposition": "3+kk",
-    "floor": 3,
-    "area": 70,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/981278-nabidka-prodej-bytu-argentinska-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 989817,
-    "name": "Prodej bytu 3+kk 88 m²",
-    "price": 13490000,
-    "price_formatted": "13 490 000 Kč",
-    "locality": "Miroslava Hajna, Praha - Letňany",
-    "lat": 50.1406487,
-    "lon": 14.5207541,
-    "disposition": "3+kk",
-    "floor": 2,
-    "area": 88,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/989817-nabidka-prodej-bytu-miroslava-hajna-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 970257,
-    "name": "Prodej bytu 3+1 106 m²",
-    "price": 12950000,
-    "price_formatted": "12 950 000 Kč",
-    "locality": "Novákových, Praha - Libeň",
-    "lat": 50.1034771,
-    "lon": 14.4758735,
-    "disposition": "3+1",
-    "floor": 5,
-    "area": 106,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/970257-nabidka-prodej-bytu-novakovych-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 972406,
-    "name": "Prodej bytu 3+kk 83 m²",
-    "price": 10490000,
-    "price_formatted": "10 490 000 Kč",
-    "locality": "Na Výrovně, Praha - Stodůlky",
-    "lat": 50.0396067,
-    "lon": 14.3167022,
-    "disposition": "3+kk",
-    "floor": 2,
-    "area": 83,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/972406-nabidka-prodej-bytu-na-vyrovne",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 967142,
-    "name": "Prodej bytu 3+kk 78 m²",
-    "price": 11648000,
-    "price_formatted": "11 648 000 Kč",
-    "locality": "Na Míčánkách, Praha - Vršovice",
-    "lat": 50.0713284,
-    "lon": 14.4638722,
-    "disposition": "3+kk",
-    "floor": 6,
-    "area": 78,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/967142-nabidka-prodej-bytu-na-micankach",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 955977,
-    "name": "Prodej bytu 4+kk 75 m²",
-    "price": 10363000,
-    "price_formatted": "10 363 000 Kč",
-    "locality": "Karla Guta, Praha - Uhříněves",
-    "lat": 50.03017,
-    "lon": 14.5940072,
-    "disposition": "4+kk",
-    "floor": 4,
-    "area": 75,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/955977-nabidka-prodej-bytu-karla-guta",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 974557,
-    "name": "Prodej bytu 4+kk 94 m²",
-    "price": 13499900,
-    "price_formatted": "13 499 900 Kč",
-    "locality": "V Dolině, Praha - Michle",
-    "lat": 50.0579963,
-    "lon": 14.4682887,
-    "disposition": "4+kk",
-    "floor": 8,
-    "area": 94,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/974557-nabidka-prodej-bytu-v-doline-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 988498,
-    "name": "Prodej bytu 3+1 75 m²",
-    "price": 11400000,
-    "price_formatted": "11 400 000 Kč",
-    "locality": "5. května, Praha - Nusle",
-    "lat": 50.0604096,
-    "lon": 14.4326302,
-    "disposition": "3+1",
-    "floor": 4,
-    "area": 75,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/988498-nabidka-prodej-bytu-5-kvetna-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 985285,
-    "name": "Prodej bytu 3+kk 70 m²",
-    "price": 12200000,
-    "price_formatted": "12 200 000 Kč",
-    "locality": "Klausova, Praha - Stodůlky",
-    "lat": 50.0370204,
-    "lon": 14.3432643,
-    "disposition": "3+kk",
-    "floor": 5,
-    "area": 70,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/985285-nabidka-prodej-bytu-klausova-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 965526,
-    "name": "Prodej bytu 3+kk 77 m²",
-    "price": 11890000,
-    "price_formatted": "11 890 000 Kč",
-    "locality": "Vinohradská, Praha - Strašnice",
-    "lat": 50.0776726,
-    "lon": 14.4870072,
-    "disposition": "3+kk",
-    "floor": 16,
-    "area": 77,
-    "building_type": "Smíšená",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/965526-nabidka-prodej-bytu-vinohradska-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 924811,
-    "name": "Prodej bytu 3+kk 75 m²",
-    "price": 13390000,
-    "price_formatted": "13 390 000 Kč",
-    "locality": "Waltariho, Praha - Hloubětín",
-    "lat": 50.1076717,
-    "lon": 14.5248559,
-    "disposition": "3+kk",
-    "floor": 4,
-    "area": 75,
-    "building_type": "Smíšená",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/924811-nabidka-prodej-bytu-waltariho-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 985859,
-    "name": "Prodej bytu 3+1 80 m²",
-    "price": 9000000,
-    "price_formatted": "9 000 000 Kč",
-    "locality": "Staňkova, Praha - Háje",
-    "lat": 50.0377128,
-    "lon": 14.5311557,
-    "disposition": "3+1",
-    "floor": 2,
-    "area": 80,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/985859-nabidka-prodej-bytu-stankova-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 985583,
-    "name": "Prodej bytu 3+kk 76 m²",
-    "price": 10850000,
-    "price_formatted": "10 850 000 Kč",
-    "locality": "Boloňská, Praha - Horní Měcholupy",
-    "lat": 50.047328,
-    "lon": 14.5565277,
-    "disposition": "3+kk",
-    "floor": 4,
-    "area": 76,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/985583-nabidka-prodej-bytu-bolonska-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 981178,
-    "name": "Prodej bytu 4+kk 86 m²",
-    "price": 11990000,
-    "price_formatted": "11 990 000 Kč",
-    "locality": "Sušilova, Praha - Uhříněves",
-    "lat": 50.032081,
-    "lon": 14.5885148,
-    "disposition": "4+kk",
-    "floor": 2,
-    "area": 86,
-    "building_type": "SKELET",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/981178-nabidka-prodej-bytu-susilova-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 973216,
-    "name": "Prodej bytu 4+1 82 m²",
-    "price": 11357000,
-    "price_formatted": "11 357 000 Kč",
-    "locality": "Nad Kapličkou, Praha - Strašnice",
-    "lat": 50.0839509,
-    "lon": 14.4904493,
-    "disposition": "4+1",
-    "floor": 2,
-    "area": 82,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/973216-nabidka-prodej-bytu-nad-kaplickou-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 868801,
-    "name": "Prodej bytu 3+kk 109 m²",
-    "price": 7299000,
-    "price_formatted": "7 299 000 Kč",
-    "locality": "Pod Karlovem, Praha - Vinohrady",
-    "lat": 50.0676313,
-    "lon": 14.432498,
-    "disposition": "3+kk",
-    "floor": 5,
-    "area": 109,
-    "building_type": "Cihlová",
-    "ownership": "Družstevní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/868801-nabidka-prodej-bytu-pod-karlovem-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 868795,
-    "name": "Prodej bytu 3+kk 106 m²",
-    "price": 6299000,
-    "price_formatted": "6 299 000 Kč",
-    "locality": "Pod Karlovem, Praha - Vinohrady",
-    "lat": 50.0676313,
-    "lon": 14.432498,
-    "disposition": "3+kk",
-    "floor": 2,
-    "area": 106,
-    "building_type": "Cihlová",
-    "ownership": "Družstevní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/868795-nabidka-prodej-bytu-pod-karlovem-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 981890,
-    "name": "Prodej bytu 3+1 84 m²",
-    "price": 12980000,
-    "price_formatted": "12 980 000 Kč",
-    "locality": "Novákových, Praha - Libeň",
-    "lat": 50.103273,
-    "lon": 14.4746894,
-    "disposition": "3+1",
-    "floor": 2,
-    "area": 84,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/981890-nabidka-prodej-bytu-novakovych-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 976276,
-    "name": "Prodej bytu 3+kk 75 m²",
-    "price": 13490000,
-    "price_formatted": "13 490 000 Kč",
-    "locality": "Svornosti, Praha - Smíchov",
-    "lat": 50.0673284,
-    "lon": 14.4095087,
-    "disposition": "3+kk",
-    "floor": 2,
-    "area": 75,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/976276-nabidka-prodej-bytu-svornosti-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 950787,
-    "name": "Prodej bytu 3+kk 70 m²",
-    "price": 9999000,
-    "price_formatted": "9 999 000 Kč",
-    "locality": "Sečská, Praha - Strašnice",
-    "lat": 50.071191,
-    "lon": 14.5035501,
-    "disposition": "3+kk",
-    "floor": 3,
-    "area": 70,
-    "building_type": "Smíšená",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/950787-nabidka-prodej-bytu-secska-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 978045,
-    "name": "Prodej bytu 3+kk 76 m²",
-    "price": 11133000,
-    "price_formatted": "11 133 000 Kč",
-    "locality": "K Vinoři, Praha - Kbely",
-    "lat": 50.1329656,
-    "lon": 14.5618499,
-    "disposition": "3+kk",
-    "floor": 2,
-    "area": 76,
-    "building_type": "Smíšená",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/978045-nabidka-prodej-bytu-k-vinori",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 974552,
-    "name": "Prodej bytu 3+1 75 m²",
-    "price": 11000000,
-    "price_formatted": "11 000 000 Kč",
-    "locality": "Vejražkova, Praha - Košíře",
-    "lat": 50.0637808,
-    "lon": 14.3612275,
-    "disposition": "3+1",
-    "floor": 2,
-    "area": 75,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/974552-nabidka-prodej-bytu-vejrazkova-praha",
-    "source": "bezrealitky",
-    "image": ""
-  },
-  {
-    "hash_id": 955010,
-    "name": "Prodej bytu 3+kk 70 m²",
-    "price": 12290000,
-    "price_formatted": "12 290 000 Kč",
-    "locality": "Břeclavská, Praha - Kyje",
-    "lat": 50.0951045,
-    "lon": 14.5454237,
-    "disposition": "3+kk",
-    "floor": 2,
-    "area": 70,
-    "building_type": "Cihlová",
-    "ownership": "Osobní",
-    "url": "https://www.bezrealitky.cz/nemovitosti-byty-domy/955010-nabidka-prodej-bytu-breclavska-hlavni-mesto-praha",
-    "source": "bezrealitky",
-    "image": ""
-  }
-]

byty_cityhome.json

@@ -1 +0,0 @@
-[]

byty_psn.json

@@ -1 +0,0 @@
-[]

docs/validation.md

@@ -0,0 +1,123 @@
+# Validation Recipe
+
+End-to-end check that scraping, data persistence, history, and the status page all work correctly in Docker.
+
+## What it verifies
+
+- All scrapers run and write output to `DATA_DIR` (`/app/data`)
+- `stats_*.json` land in `/app/data/` (not in `/app/`)
+- `status.json` and `scraper_history.json` land in `/app/data/`
+- `/api/status`, `/api/status/history`, and `/scrapers-status` serve correct data
+- History accumulates across runs
+
+## Steps
+
+### 1. Build the image
+
+```bash
+make build
+```
+
+### 2. Start a clean validation container
+
+```bash
+# Stop/remove any leftover container and volume from a previous run
+docker stop maru-hleda-byt-validation 2>/dev/null; docker rm maru-hleda-byt-validation 2>/dev/null
+docker volume rm maru-hleda-byt-validation-data 2>/dev/null
+
+docker run -d --name maru-hleda-byt-validation \
+  -p 8081:8080 \
+  -v maru-hleda-byt-validation-data:/app/data \
+  maru-hleda-byt
+```
+
+Give the container ~3 seconds to start. The entrypoint launches a background full scrape automatically — suppress it so only controlled runs execute:
+
+```bash
+sleep 3
+docker exec maru-hleda-byt-validation pkill -f run_all.sh 2>/dev/null || true
+docker exec maru-hleda-byt-validation rm -f /app/data/scraper_running.json 2>/dev/null || true
+```
+
+### 3. Run a limited scrape (run 1)
+
+```bash
+docker exec maru-hleda-byt-validation bash /app/run_all.sh --max-pages 1 --max-properties 10
+```
+
+Expected output (last few lines):
+```
+Status uložen: /app/data/status.json
+Historie uložena: /app/data/scraper_history.json (1 záznamů)
+```
+
+### 4. Verify data files are in `/app/data/`
+
+```bash
+docker exec maru-hleda-byt-validation ls /app/data/
+```
+
+Expected files:
+```
+byty_cityhome.json   byty_idnes.json   byty_merged.json
+byty_realingo.json   byty_sreality.json
+mapa_bytu.html
+scraper_history.json
+stats_bezrealitky.json  stats_cityhome.json  stats_idnes.json
+stats_realingo.json     stats_sreality.json
+status.json
+```
+
+### 5. Run a second limited scrape (run 2)
+
+```bash
+docker exec maru-hleda-byt-validation bash /app/run_all.sh --max-pages 1 --max-properties 10
+```
+
+Expected last line: `Historie uložena: /app/data/scraper_history.json (2 záznamů)`
+
+### 6. Verify history via API
+
+```bash
+curl -s http://localhost:8081/api/status/history | python3 -c "
+import json, sys
+h = json.load(sys.stdin)
+print(f'{len(h)} entries:')
+for i, e in enumerate(h):
+    print(f'  [{i}] {e[\"timestamp\"]} total={e[\"total_accepted\"]}')
+"
+```
+
+Expected: 2 entries with different timestamps.
+
+```bash
+curl -s http://localhost:8081/api/status | python3 -c "
+import json, sys; s=json.load(sys.stdin)
+print(f'status={s[\"status\"]} total={s[\"total_accepted\"]} ts={s[\"timestamp\"]}')
+"
+```
+
+Expected: `status=done total=<N> ts=<latest timestamp>`
+
+### 7. Check the status page
+
+Open http://localhost:8081/scrapers-status in a browser (or `curl -s http://localhost:8081/scrapers-status | grep -c "clickable-row"` — should print `2`).
+
+### 8. Clean up
+
+```bash
+docker stop maru-hleda-byt-validation && docker rm maru-hleda-byt-validation
+docker volume rm maru-hleda-byt-validation-data
+```
+
+Or use the Makefile shortcut:
+
+```bash
+make validation-stop
+```
+
+## Notes
+
+- PSN scraper does not support `--max-pages` and will always fail with this command; `success=False` in history is expected during validation.
+- Bezrealitky may return 0 results with a 1-page limit; `byty_bezrealitky.json` will be absent from `/app/data/` in that case — this is normal.
+- `make validation` (the Makefile target) runs the same limited scrape but does not suppress the background startup scrape, so two concurrent runs may occur. Use the manual steps above for a clean controlled test.

generate_status.py

@@ -1,16 +1,15 @@
 #!/usr/bin/env python3
-"""Generate status.json from scraper JSON outputs and run log."""
+"""Generate status.json from scraper JSON outputs and per-scraper stats files."""
 from __future__ import annotations
 
+import argparse
 import json
 import os
-import re
-import sys
 from datetime import datetime
 from pathlib import Path
-from typing import Optional
 
 HERE = Path(__file__).parent
+DATA_DIR = Path(os.environ.get("DATA_DIR", HERE))
 
 SOURCE_FILES = {
     "Sreality":    "byty_sreality.json",
@@ -21,7 +20,17 @@ SOURCE_FILES = {
     "CityHome":    "byty_cityhome.json",
 }
 
+STATS_FILES = {
+    "Sreality":    "stats_sreality.json",
+    "Realingo":    "stats_realingo.json",
+    "Bezrealitky": "stats_bezrealitky.json",
+    "iDNES":       "stats_idnes.json",
+    "PSN":         "stats_psn.json",
+    "CityHome":    "stats_cityhome.json",
+}
+
 MERGED_FILE = "byty_merged.json"
+HISTORY_FILE = "scraper_history.json"
 
 
 def count_source(path: Path) -> dict:
@@ -36,105 +45,51 @@ def count_source(path: Path) -> dict:
         return {"accepted": 0, "error": str(e)}
 
 
-def parse_log(log_path: str) -> dict[str, dict]:
+def read_scraper_stats(path: Path) -> dict:
-    """Parse scraper run log and extract per-source statistics.
+    """Load a per-scraper stats JSON. Returns {} on missing or corrupt file."""
-
+    if not path.exists():
-    Scrapers log summary lines like:
+        return {}
-      ✓ Vyhovující byty:    12
+    try:
-      Vyloučeno (prodáno):   5
+        data = json.loads(path.read_text(encoding="utf-8"))
-      Staženo stránek:       3
+        return data if isinstance(data, dict) else {}
-      Staženo inzerátů:      48
+    except Exception:
-      Celkem bytů v cache:   120
-    and section headers like:
-      [2/6] Realingo
-    """
-    if not log_path or not os.path.exists(log_path):
         return {}
 
-    with open(log_path, encoding="utf-8") as f:
-        content = f.read()
 
-    # Split into per-source sections by the [N/6] Step header
+def append_to_history(status: dict, keep: int) -> None:
-    # Each section header looks like "[2/6] Realingo\n----..."
+    """Append the current status entry to scraper_history.json, keeping only `keep` latest."""
-    section_pattern = re.compile(r'\[(\d+)/\d+\]\s+(.+)\n-+', re.MULTILINE)
+    history_path = DATA_DIR / HISTORY_FILE
-    sections_found = list(section_pattern.finditer(content))
+    history: list = []
+    if history_path.exists():
+        try:
+            history = json.loads(history_path.read_text(encoding="utf-8"))
+            if not isinstance(history, list):
+                history = []
+        except Exception:
+            history = []
 
-    if not sections_found:
+    history.append(status)
-        return {}
 
-    stats = {}
+    # Keep only the N most recent entries
-    for i, match in enumerate(sections_found):
+    if keep > 0 and len(history) > keep:
-        step_name = match.group(2).strip()
+        history = history[-keep:]
-        start = match.end()
-        end = sections_found[i + 1].start() if i + 1 < len(sections_found) else len(content)
-        section_text = content[start:end]
 
-        # Identify which sources this section covers
+    history_path.write_text(json.dumps(history, ensure_ascii=False, indent=2), encoding="utf-8")
-        # "PSN + CityHome" covers both
+    print(f"Historie uložena: {history_path} ({len(history)} záznamů)")
-        source_names = []
-        for name in SOURCE_FILES:
-            if name.lower() in step_name.lower():
-                source_names.append(name)
-        if not source_names:
-            continue
-
-        # Parse numeric summary lines
-        def extract(pattern: str) -> Optional[int]:
-            m = re.search(pattern, section_text)
-            return int(m.group(1)) if m else None
-
-        # Lines present in all/most scrapers
-        accepted = extract(r'Vyhovující byty[:\s]+(\d+)')
-        fetched = extract(r'Staženo inzerátů[:\s]+(\d+)')
-        pages = extract(r'Staženo stránek[:\s]+(\d+)')
-        cached = extract(r'Celkem bytů v cache[:\s]+(\d+)')
-        cache_hits = extract(r'Cache hit[:\s]+(\d+)')
-
-        # Rejection reasons — collect all into a dict
-        excluded = {}
-        for m in re.finditer(r'Vyloučeno\s+\(([^)]+)\)[:\s]+(\d+)', section_text):
-            excluded[m.group(1)] = int(m.group(2))
-        # Also PSN-style "Vyloučeno (prodáno): N"
-        total_excluded = sum(excluded.values()) if excluded else extract(r'Vyloučen\w*[:\s]+(\d+)')
-
-        entry = {}
-        if accepted is not None:
-            entry["accepted"] = accepted
-        if fetched is not None:
-            entry["fetched"] = fetched
-        if pages is not None:
-            entry["pages"] = pages
-        if cached is not None:
-            entry["cached"] = cached
-        if cache_hits is not None:
-            entry["cache_hits"] = cache_hits
-        if excluded:
-            entry["excluded"] = excluded
-        elif total_excluded is not None:
-            entry["excluded_total"] = total_excluded
-
-        for name in source_names:
-            stats[name] = entry
-
-    return stats
 
 
 def main():
-    start_time = None
+    parser = argparse.ArgumentParser(description="Generate status.json from scraper outputs.")
-    duration_sec = None
+    parser.add_argument("--start-time", dest="start_time", default=None,
+                        help="ISO timestamp of scrape start (default: now)")
+    parser.add_argument("--duration", dest="duration", type=int, default=None,
+                        help="Run duration in seconds")
+    parser.add_argument("--keep", dest="keep", type=int, default=20,
+                        help="Number of history entries to keep (default: 20, 0=unlimited)")
+    args = parser.parse_args()
 
-    if len(sys.argv) >= 3:
+    start_time = args.start_time or datetime.now().isoformat(timespec="seconds")
-        start_time = sys.argv[1]
+    duration_sec = args.duration
-        try:
-            duration_sec = int(sys.argv[2])
-        except ValueError:
-            pass
-
-    if not start_time:
-        start_time = datetime.now().isoformat(timespec="seconds")
-
-    log_path = sys.argv[3] if len(sys.argv) >= 4 else None
-    log_stats = parse_log(log_path)
 
     sources = []
     for name, filename in SOURCE_FILES.items():
@@ -142,14 +97,12 @@ def main():
         info = count_source(path)
         info["name"] = name
 
-        # Merge log stats
+        # Merge in stats from the per-scraper stats file (authoritative for run data)
-        ls = log_stats.get(name, {})
+        stats = read_scraper_stats(DATA_DIR / STATS_FILES[name])
-        for k in ("fetched", "pages", "cached", "cache_hits", "excluded", "excluded_total"):
+        for key in ("accepted", "fetched", "pages", "cache_hits", "excluded", "excluded_total",
-            if k in ls:
+                    "success", "duration_sec", "error"):
-                info[k] = ls[k]
+            if key in stats:
-        # Override accepted from log if available (log is authoritative for latest run)
+                info[key] = stats[key]
-        if "accepted" in ls:
-            info["accepted"] = ls["accepted"]
 
         sources.append(info)
 
@@ -168,17 +121,21 @@ def main():
 
     duplicates_removed = total_accepted - deduplicated if deduplicated else 0
 
+    # Top-level success: True if no source has an error
+    success = not any("error" in s for s in sources)
+
     status = {
         "status": "done",
         "timestamp": start_time,
         "duration_sec": duration_sec,
+        "success": success,
         "total_accepted": total_accepted,
         "deduplicated": deduplicated,
         "duplicates_removed": duplicates_removed,
         "sources": sources,
     }
 
-    out = HERE / "status.json"
+    out = DATA_DIR / "status.json"
     out.write_text(json.dumps(status, ensure_ascii=False, indent=2), encoding="utf-8")
     print(f"Status uložen: {out}")
     print(f"  Celkem bytů (před dedup): {total_accepted}")
@@ -197,6 +154,8 @@ def main():
             parts.append(f"[CHYBA: {err}]")
         print("  " + "  ".join(parts))
 
+    append_to_history(status, args.keep)
+
 
 if __name__ == "__main__":
     main()

merge_and_map.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 """
-Sloučí data ze Sreality, Realinga, Bezrealitek, iDNES, PSN a CityHome,
+Sloučí data ze Sreality, Realinga, Bezrealitek, iDNES, PSN, CityHome a Bazoše,
 deduplikuje a vygeneruje mapu.
 Deduplikace: stejná ulice (z locality) + stejná cena + stejná plocha = duplikát.
 PSN a CityHome mají při deduplikaci prioritu (načtou se první).
@@ -9,6 +9,7 @@ from __future__ import annotations
 
 import json
 import re
+import unicodedata
 from pathlib import Path
 
 from scrape_and_map import generate_map, format_price
@@ -19,14 +20,8 @@ def normalize_street(locality: str) -> str:
     # "Studentská, Praha 6 - Dejvice" → "studentska"
     # "Rýnská, Praha" → "rynska"
     street = locality.split(",")[0].strip().lower()
-    # Remove diacritics (simple Czech)
+    # Remove diacritics using Unicode decomposition (handles all Czech characters)
-    replacements = {
+    street = unicodedata.normalize("NFKD", street).encode("ascii", "ignore").decode("ascii")
-        "á": "a", "č": "c", "ď": "d", "é": "e", "ě": "e",
-        "í": "i", "ň": "n", "ó": "o", "ř": "r", "š": "s",
-        "ť": "t", "ú": "u", "ů": "u", "ý": "y", "ž": "z",
-    }
-    for src, dst in replacements.items():
-        street = street.replace(src, dst)
     # Remove non-alphanumeric
     street = re.sub(r"[^a-z0-9]", "", street)
     return street
@@ -49,6 +44,7 @@ def main():
         ("Realingo",    "byty_realingo.json"),
         ("Bezrealitky", "byty_bezrealitky.json"),
         ("iDNES",       "byty_idnes.json"),
+        ("Bazoš",       "byty_bazos.json"),
     ]
 
     all_estates = []
@@ -79,6 +75,10 @@ def main():
         if key in seen_keys:
             dupes += 1
             existing = seen_keys[key]
+            # Preserve earliest first_seen across sources
+            dup_fs = e.get("first_seen", "")
+            if dup_fs and (not existing.get("first_seen") or dup_fs < existing["first_seen"]):
+                existing["first_seen"] = dup_fs
             # Log it
             print(f"  Duplikát: {e['locality']} | {format_price(e['price'])} | {e.get('area', '?')} m² "
                   f"({e.get('source', '?')} vs {existing.get('source', '?')})")

ratings_server.py

@@ -1,116 +0,0 @@
-#!/usr/bin/env python3
-"""
-Minimal HTTP API server for persisting apartment ratings.
-
-GET  /api/ratings        → returns ratings.json contents
-POST /api/ratings        → saves entire ratings object
-GET  /api/ratings/export → same as GET, but with download header
-
-Ratings file: /app/data/ratings.json (or ./ratings.json locally)
-"""
-
-import json
-import logging
-import os
-import sys
-from http.server import BaseHTTPRequestHandler, HTTPServer
-from pathlib import Path
-
-PORT = int(os.environ.get("RATINGS_PORT", 8081))
-DATA_DIR = Path(os.environ.get("DATA_DIR", "."))
-RATINGS_FILE = DATA_DIR / "ratings.json"
-
-logging.basicConfig(
-    level=logging.INFO,
-    format="%(asctime)s [ratings] %(levelname)s %(message)s",
-    datefmt="%Y-%m-%dT%H:%M:%S",
-)
-log = logging.getLogger(__name__)
-
-
-def load_ratings() -> dict:
-    try:
-        if RATINGS_FILE.exists():
-            return json.loads(RATINGS_FILE.read_text(encoding="utf-8"))
-    except Exception as e:
-        log.error("Failed to load ratings: %s", e)
-    return {}
-
-
-def save_ratings(data: dict) -> None:
-    RATINGS_FILE.write_text(
-        json.dumps(data, ensure_ascii=False, indent=2),
-        encoding="utf-8",
-    )
-
-
-class RatingsHandler(BaseHTTPRequestHandler):
-    def log_message(self, format, *args):
-        # Suppress default HTTP access log (we use our own)
-        pass
-
-    def _send_json(self, status: int, body: dict, extra_headers=None):
-        payload = json.dumps(body, ensure_ascii=False).encode("utf-8")
-        self.send_response(status)
-        self.send_header("Content-Type", "application/json; charset=utf-8")
-        self.send_header("Content-Length", str(len(payload)))
-        self.send_header("Access-Control-Allow-Origin", "*")
-        self.send_header("Access-Control-Allow-Methods", "GET, POST, OPTIONS")
-        self.send_header("Access-Control-Allow-Headers", "Content-Type")
-        if extra_headers:
-            for k, v in extra_headers.items():
-                self.send_header(k, v)
-        self.end_headers()
-        self.wfile.write(payload)
-
-    def do_OPTIONS(self):
-        # CORS preflight
-        self.send_response(204)
-        self.send_header("Access-Control-Allow-Origin", "*")
-        self.send_header("Access-Control-Allow-Methods", "GET, POST, OPTIONS")
-        self.send_header("Access-Control-Allow-Headers", "Content-Type")
-        self.end_headers()
-
-    def do_GET(self):
-        if self.path in ("/api/ratings", "/api/ratings/export"):
-            ratings = load_ratings()
-            extra = None
-            if self.path == "/api/ratings/export":
-                extra = {"Content-Disposition": 'attachment; filename="ratings.json"'}
-            log.info("GET %s → %d ratings", self.path, len(ratings))
-            self._send_json(200, ratings, extra)
-        else:
-            self._send_json(404, {"error": "not found"})
-
-    def do_POST(self):
-        if self.path == "/api/ratings":
-            length = int(self.headers.get("Content-Length", 0))
-            if length == 0:
-                self._send_json(400, {"error": "empty body"})
-                return
-            try:
-                raw = self.rfile.read(length)
-                data = json.loads(raw.decode("utf-8"))
-            except Exception as e:
-                log.warning("Bad request body: %s", e)
-                self._send_json(400, {"error": "invalid JSON"})
-                return
-            if not isinstance(data, dict):
-                self._send_json(400, {"error": "expected JSON object"})
-                return
-            save_ratings(data)
-            log.info("POST /api/ratings → saved %d ratings", len(data))
-            self._send_json(200, {"ok": True, "count": len(data)})
-        else:
-            self._send_json(404, {"error": "not found"})
-
-
-if __name__ == "__main__":
-    log.info("Ratings server starting on port %d, data dir: %s", PORT, DATA_DIR)
-    log.info("Ratings file: %s", RATINGS_FILE)
-    server = HTTPServer(("0.0.0.0", PORT), RatingsHandler)
-    try:
-        server.serve_forever()
-    except KeyboardInterrupt:
-        log.info("Stopped.")
-        sys.exit(0)

regen_map.py

@@ -1,114 +0,0 @@
-#!/usr/bin/env python3
-"""
-Přegeneruje mapu z již stažených dat (byty_sreality.json).
-Doplní chybějící plochy ze Sreality API, opraví URL, aplikuje filtry.
-"""
-from __future__ import annotations
-
-import json
-import time
-import urllib.request
-from pathlib import Path
-
-from scrape_and_map import (
-    generate_map, format_price, MIN_AREA, HEADERS, DETAIL_API
-)
-
-
-def api_get(url: str) -> dict:
-    req = urllib.request.Request(url, headers=HEADERS)
-    with urllib.request.urlopen(req, timeout=30) as resp:
-        return json.loads(resp.read().decode("utf-8"))
-
-
-def fix_sreality_url(estate: dict) -> str:
-    """Fix the Sreality URL to include disposition segment (only if missing)."""
-    disp = estate.get("disposition", "")
-    slug_map = {
-        "1+kk": "1+kk", "1+1": "1+1", "2+kk": "2+kk", "2+1": "2+1",
-        "3+kk": "3+kk", "3+1": "3+1", "4+kk": "4+kk", "4+1": "4+1",
-        "5+kk": "5+kk", "5+1": "5+1", "6+": "6-a-vice", "Atypický": "atypicky",
-    }
-    slug = slug_map.get(disp, "byt")
-    old_url = estate.get("url", "")
-    parts = old_url.split("/")
-    try:
-        byt_idx = parts.index("byt")
-        # Only insert if disposition slug is not already there
-        if byt_idx + 1 < len(parts) and parts[byt_idx + 1] == slug:
-            return old_url  # already correct
-        parts.insert(byt_idx + 1, slug)
-        return "/".join(parts)
-    except ValueError:
-        return old_url
-
-
-def fetch_area(hash_id: int) -> int | None:
-    """Fetch area from detail API."""
-    try:
-        url = DETAIL_API.format(hash_id)
-        detail = api_get(url)
-        for item in detail.get("items", []):
-            name = item.get("name", "")
-            if "žitná ploch" in name or "zitna ploch" in name.lower():
-                return int(item["value"])
-    except Exception:
-        pass
-    return None
-
-
-def main():
-    json_path = Path("byty_sreality.json")
-    if not json_path.exists():
-        print("Soubor byty_sreality.json nenalezen. Nejprve spusť scrape_and_map.py")
-        return
-
-    estates = json.loads(json_path.read_text(encoding="utf-8"))
-    print(f"Načteno {len(estates)} bytů z byty_sreality.json")
-
-    # Step 1: Fetch missing areas
-    missing_area = [e for e in estates if e.get("area") is None]
-    print(f"Doplňuji plochu u {len(missing_area)} bytů...")
-
-    for i, e in enumerate(missing_area):
-        time.sleep(0.3)
-        area = fetch_area(e["hash_id"])
-        if area is not None:
-            e["area"] = area
-        if (i + 1) % 50 == 0:
-            print(f"  {i + 1}/{len(missing_area)} ...")
-
-    # Count results
-    with_area = sum(1 for e in estates if e.get("area") is not None)
-    print(f"Plocha doplněna: {with_area}/{len(estates)}")
-
-    # Step 2: Fix URLs
-    for e in estates:
-        e["url"] = fix_sreality_url(e)
-
-    # Step 3: Filter by min area
-    filtered = []
-    excluded = 0
-    for e in estates:
-        area = e.get("area")
-        if area is not None and area < MIN_AREA:
-            excluded += 1
-            continue
-        filtered.append(e)
-
-    print(f"Vyloučeno (< {MIN_AREA} m²): {excluded}")
-    print(f"Zbývá: {len(filtered)} bytů")
-
-    # Save updated data
-    filtered_path = Path("byty_sreality.json")
-    filtered_path.write_text(
-        json.dumps(filtered, ensure_ascii=False, indent=2),
-        encoding="utf-8",
-    )
-
-    # Generate map
-    generate_map(filtered)
-
-
-if __name__ == "__main__":
-    main()

run_all.sh

@@ -13,15 +13,17 @@ RED='\033[0;31m'
 BOLD='\033[1m'
 NC='\033[0m'
 
-TOTAL=6
+TOTAL=7
 CURRENT=0
 FAILED=0
 START_TIME=$(date -u +"%Y-%m-%dT%H:%M:%S")
 START_EPOCH=$(date +%s)
 LOG_FILE="$(pwd)/scrape_run.log"
 
-# Mark status as running
+# Mark scraper as running; cleaned up on exit (even on error/kill)
-echo '{"status":"running"}' > status.json
+LOCK_FILE="${DATA_DIR:-.}/scraper_running.json"
+echo '{"running":true,"started_at":"'"$START_TIME"'"}' > "$LOCK_FILE"
+trap 'rm -f "$LOCK_FILE"' EXIT
 
 show_help() {
     echo "Usage: ./run_all.sh [OPTIONS]"
@@ -32,16 +34,19 @@ show_help() {
     echo "  --max-pages N         Maximální počet stránek ke stažení z každého zdroje"
     echo "  --max-properties N    Maximální počet nemovitostí ke stažení z každého zdroje"
     echo "  --log-level LEVEL     Úroveň logování (DEBUG, INFO, WARNING, ERROR)"
+    echo "  --keep N              Počet běhů v historii (výchozí: 5, 0=neomezeno)"
     echo "  -h, --help            Zobrazí tuto nápovědu"
     echo ""
     echo "Examples:"
     echo "  ./run_all.sh                                  # plný běh"
     echo "  ./run_all.sh --max-pages 1 --max-properties 10  # rychlý test"
     echo "  ./run_all.sh --log-level DEBUG                # s debug logováním"
+    echo "  ./run_all.sh --keep 10                        # uchovej 10 běhů v historii"
 }
 
 # Parse arguments
 SCRAPER_ARGS=""
+KEEP_ARG=""
 while [[ $# -gt 0 ]]; do
     case $1 in
         -h|--help)
@@ -52,6 +57,10 @@ while [[ $# -gt 0 ]]; do
             SCRAPER_ARGS="$SCRAPER_ARGS $1 $2"
             shift 2
             ;;
+        --keep)
+            KEEP_ARG="--keep $2"
+            shift 2
+            ;;
         *)
             echo "Unknown argument: $1"
             echo ""
@@ -75,9 +84,6 @@ exec > >(tee -a "$LOG_FILE") 2>&1
 step "Sreality"
 python3 scrape_and_map.py $SCRAPER_ARGS || { echo -e "${RED}✗ Sreality selhalo${NC}"; FAILED=$((FAILED + 1)); }
 
-step "Realingo"
-python3 scrape_realingo.py $SCRAPER_ARGS || { echo -e "${RED}✗ Realingo selhalo${NC}"; FAILED=$((FAILED + 1)); }
-
 step "Bezrealitky"
 python3 scrape_bezrealitky.py $SCRAPER_ARGS || { echo -e "${RED}✗ Bezrealitky selhalo${NC}"; FAILED=$((FAILED + 1)); }
 
@@ -92,6 +98,12 @@ PID_CH=$!
 wait $PID_PSN || { echo -e "${RED}✗ PSN selhalo${NC}"; FAILED=$((FAILED + 1)); }
 wait $PID_CH  || { echo -e "${RED}✗ CityHome selhalo${NC}"; FAILED=$((FAILED + 1)); }
 
+step "Bazoš"
+python3 scrape_bazos.py $SCRAPER_ARGS || { echo -e "${RED}✗ Bazoš selhalo${NC}"; FAILED=$((FAILED + 1)); }
+
+step "Realingo"
+python3 scrape_realingo.py $SCRAPER_ARGS || { echo -e "${RED}✗ Realingo selhalo${NC}"; FAILED=$((FAILED + 1)); }
+
 # ── Sloučení + mapa ──────────────────────────────────────────
 
 step "Sloučení dat a generování mapy"
@@ -103,12 +115,12 @@ python3 merge_and_map.py || { echo -e "${RED}✗ Merge selhal${NC}"; FAILED=$((F
 
 END_EPOCH=$(date +%s)
 DURATION=$((END_EPOCH - START_EPOCH))
-python3 generate_status.py "$START_TIME" "$DURATION" "$LOG_FILE"
+python3 generate_status.py --start-time "$START_TIME" --duration "$DURATION" $KEEP_ARG
 
 echo ""
 echo "============================================================"
 if [ $FAILED -eq 0 ]; then
-    echo -e "${GREEN}${BOLD}Hotovo! Všech 6 zdrojů úspěšně staženo.${NC}"
+    echo -e "${GREEN}${BOLD}Hotovo! Všech 7 zdrojů úspěšně staženo.${NC}"
 else
     echo -e "${RED}${BOLD}Hotovo s $FAILED chybami.${NC}"
 fi

scrape_and_map.py

@@ -13,8 +13,11 @@ import math
 import time
 import urllib.request
 import urllib.parse
-from datetime import datetime
+from datetime import datetime, timedelta
 from pathlib import Path
+from scraper_stats import write_stats, validate_listing
+
+STATS_FILE = "stats_sreality.json"
 
 logger = logging.getLogger(__name__)
 
@@ -42,19 +45,26 @@ HEADERS = {
 
 
 def api_get(url: str) -> dict:
-    """Fetch JSON from Sreality API."""
+    """Fetch JSON from Sreality API with retry."""
-    logger.debug(f"HTTP GET request: {url}")
+    for attempt in range(3):
-    logger.debug(f"Headers: {HEADERS}")
+        logger.debug(f"HTTP GET request (attempt {attempt + 1}/3): {url}")
-    req = urllib.request.Request(url, headers=HEADERS)
+        req = urllib.request.Request(url, headers=HEADERS)
-    try:
+        try:
-        with urllib.request.urlopen(req, timeout=30) as resp:
+            with urllib.request.urlopen(req, timeout=30) as resp:
-            response_data = resp.read().decode("utf-8")
+                response_data = resp.read().decode("utf-8")
-            logger.debug(f"HTTP response: status={resp.status}, size={len(response_data)} bytes")
+                logger.debug(f"HTTP response: status={resp.status}, size={len(response_data)} bytes")
-            logger.debug(f"Response preview: {response_data[:200]}")
+                logger.debug(f"Response preview: {response_data[:200]}")
-            return json.loads(response_data)
+                return json.loads(response_data)
-    except (urllib.error.URLError, ConnectionError, OSError) as e:
+        except urllib.error.HTTPError:
-        logger.error(f"HTTP request failed for {url}: {e}", exc_info=True)
+            raise
-        raise
+        except (urllib.error.URLError, ConnectionError, OSError) as e:
+            if attempt < 2:
+                wait = (attempt + 1) * 2
+                logger.warning(f"Connection error (retry {attempt + 1}/3 after {wait}s): {e}")
+                time.sleep(wait)
+            else:
+                logger.error(f"HTTP request failed after 3 attempts: {e}", exc_info=True)
+                raise
 
 
 def build_list_url(disposition: int, page: int = 1) -> str:
@@ -209,6 +219,8 @@ def load_cache(json_path: str = "byty_sreality.json") -> dict[int, dict]:
 
 def scrape(max_pages: int | None = None, max_properties: int | None = None):
     """Main scraping function. Returns list of filtered estates."""
+    _run_start = time.time()
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     all_estates_raw = []
     cache = load_cache()
 
@@ -348,7 +360,11 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
             "url": sreality_url(hash_id, seo),
             "image": (estate.get("_links", {}).get("images", [{}])[0].get("href", "") if estate.get("_links", {}).get("images") else ""),
             "scraped_at": datetime.now().strftime("%Y-%m-%d"),
+            "first_seen": cached.get("first_seen", datetime.now().strftime("%Y-%m-%d")) if cached else datetime.now().strftime("%Y-%m-%d"),
+            "last_changed": datetime.now().strftime("%Y-%m-%d"),
         }
+        if not validate_listing(result, "sreality"):
+            continue
         results.append(result)
         details_fetched += 1
 
@@ -366,6 +382,21 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
     logger.info(f"  ✓ Vyhovující byty:    {len(results)}")
     logger.info(f"{'=' * 60}")
 
+    write_stats(STATS_FILE, {
+        "source": "Sreality",
+        "timestamp": _run_ts,
+        "duration_sec": round(time.time() - _run_start, 1),
+        "success": True,
+        "accepted": len(results),
+        "fetched": len(unique_estates),
+        "cache_hits": cache_hits,
+        "excluded": {
+            "panel/síd": excluded_panel,
+            "<69 m²": excluded_small,
+            "bez GPS": excluded_no_gps,
+            "bez detailu": excluded_no_detail,
+        },
+    })
     return results
 
 
@@ -409,18 +440,30 @@ def generate_map(estates: list[dict], output_path: str = "mapa_bytu.html"):
     ]
     for bcolor, blabel in bands:
         price_legend_items += (
-            f'<div style="display:flex;align-items:center;gap:6px;margin:2px 0;">'
+            f'<div class="price-band" data-color="{bcolor}" onclick="toggleColorFilter(\'{bcolor}\')" '
+            f'style="display:flex;align-items:center;gap:6px;margin:2px 0;padding:2px 4px;'
+            f'border-radius:4px;border:2px solid transparent;">'
             f'<span style="width:14px;height:14px;border-radius:50%;background:{bcolor};'
             f'display:inline-block;border:2px solid white;box-shadow:0 1px 3px rgba(0,0,0,0.3);flex-shrink:0;"></span>'
             f'<span>{blabel}</span></div>'
         )
+    price_legend_items += (
+        '<div id="price-filter-reset" style="display:none;margin:3px 0 0 4px;">'
+        '<a href="#" onclick="resetColorFilter();return false;" '
+        'style="font-size:11px;color:#1976D2;text-decoration:none;">✕ Zobrazit všechny ceny</a>'
+        '</div>'
+    )
     # New marker indicator — bigger dot, no extra border
     price_legend_items += (
         '<div style="display:flex;align-items:center;gap:6px;margin:6px 0 0 0;'
         'padding-top:6px;border-top:1px solid #eee;">'
-        '<span style="width:18px;height:18px;border-radius:50%;background:#66BB6A;'
+        '<span style="display:inline-flex;align-items:center;gap:3px;flex-shrink:0;">'
-        'display:inline-block;box-shadow:0 1px 4px rgba(0,0,0,0.35);flex-shrink:0;"></span>'
+        '<span style="width:14px;height:14px;border-radius:50%;background:#66BB6A;'
-        '<span>Nové (z dnešního scrapu) — větší</span></div>'
+        'display:inline-block;box-shadow:0 1px 3px rgba(0,0,0,0.3);"></span>'
+        '<span style="font-size:8px;font-weight:700;background:#FFD600;color:#333;'
+        'padding:1px 3px;border-radius:2px;">NEW</span>'
+        '</span>'
+        '<span>Nové (≤ 1 den)</span></div>'
     )
 
     markers_js = ""
@@ -437,23 +480,37 @@ def generate_map(estates: list[dict], output_path: str = "mapa_bytu.html"):
             floor_note = '<br><span style="color:#FF9800;font-weight:bold;">⚠ 2. NP — zvážit klidnost lokality</span>'
 
         source = e.get("source", "sreality")
-        source_labels = {"sreality": "Sreality", "realingo": "Realingo", "bezrealitky": "Bezrealitky", "idnes": "iDNES", "psn": "PSN", "cityhome": "CityHome"}
+        source_labels = {"sreality": "Sreality", "realingo": "Realingo", "bezrealitky": "Bezrealitky", "idnes": "iDNES", "psn": "PSN", "cityhome": "CityHome", "bazos": "Bazoš"}
-        source_colors = {"sreality": "#1976D2", "realingo": "#00897B", "bezrealitky": "#E91E63", "idnes": "#FF6F00", "psn": "#D32F2F", "cityhome": "#D32F2F"}
+        source_colors = {"sreality": "#1976D2", "realingo": "#00897B", "bezrealitky": "#E91E63", "idnes": "#FF6F00", "psn": "#D32F2F", "cityhome": "#D32F2F", "bazos": "#7B1FA2"}
         source_label = source_labels.get(source, source)
         source_color = source_colors.get(source, "#999")
 
-        hash_id = e.get("hash_id", "")
+        hash_id = f"{source}_{e.get('hash_id', '')}"
 
-        scraped_at = e.get("scraped_at", "")
+        first_seen = e.get("first_seen", "")
-        is_new = scraped_at == datetime.now().strftime("%Y-%m-%d")
+        last_changed = e.get("last_changed", "")
+        today = datetime.now().strftime("%Y-%m-%d")
+        yesterday = (datetime.now() - timedelta(days=1)).strftime("%Y-%m-%d")
+        is_new = first_seen in (today, yesterday)
 
         new_badge = (
             '<span style="margin-left:6px;font-size:11px;background:#FFD600;color:#333;'
             'padding:1px 6px;border-radius:3px;font-weight:bold;">NOVÉ</span>'
             if is_new else ""
         )
+
+        date_parts = []
+        if first_seen:
+            date_parts.append(f'Přidáno: {first_seen}')
+        if last_changed and last_changed != first_seen:
+            date_parts.append(f'Změněno: {last_changed}')
+        date_row = (
+            f'<span style="font-size:11px;color:#888;">{"&nbsp;·&nbsp;".join(date_parts)}</span><br>'
+            if date_parts else ""
+        )
+
         popup = (
-            f'<div style="min-width:280px;font-family:system-ui,sans-serif;" data-hashid="{hash_id}">'
+            f'<div style="min-width:280px;font-family:system-ui,sans-serif;" data-hashid="{hash_id}" data-first-seen="{first_seen}" data-last-changed="{last_changed}">'
             f'<b style="font-size:14px;">{format_price(e["price"])}</b>'
             f'<span style="margin-left:8px;font-size:11px;background:{source_color};color:white;'
             f'padding:1px 6px;border-radius:3px;">{source_label}</span>{new_badge}<br>'
@@ -461,7 +518,9 @@ def generate_map(estates: list[dict], output_path: str = "mapa_bytu.html"):
             f'{floor_note}<br><br>'
             f'<b>{e["locality"]}</b><br>'
             f'Stavba: {building_text}<br>'
-            f'Vlastnictví: {ownership_text}<br><br>'
+            f'Vlastnictví: {ownership_text}<br>'
+            f'{date_row}'
+            f'<br>'
             f'<a href="{e["url"]}" target="_blank" '
             f'style="color:{source_color};text-decoration:none;font-weight:bold;">'
             f'→ Otevřít na {source_label}</a>'
@@ -493,7 +552,7 @@ def generate_map(estates: list[dict], output_path: str = "mapa_bytu.html"):
         else:
             marker_fn = "addMarker"
         markers_js += (
-            f"  {marker_fn}({e['lat']}, {e['lon']}, '{color}', '{popup}', '{hash_id}');\n"
+            f"  {marker_fn}({e['lat']}, {e['lon']}, '{color}', '{popup}', '{hash_id}', '{first_seen}', '{last_changed}');\n"
         )
 
     # Build legend — price per m² bands + disposition counts
@@ -559,12 +618,12 @@ def generate_map(estates: list[dict], output_path: str = "mapa_bytu.html"):
   .heart-icon-fav svg path {{ stroke: gold !important; stroke-width: 2.5 !important; filter: drop-shadow(0 0 4px rgba(255,193,7,0.7)); }}
   .heart-icon-rej {{ opacity: 0.4 !important; filter: grayscale(1); }}
   .reject-overlay {{ background: none !important; border: none !important; pointer-events: none !important; }}
-  @keyframes pulse-new {{
+  .new-badge-icon {{ background: none !important; border: none !important; pointer-events: none !important; }}
-    0%   {{ stroke-opacity: 1;   stroke-width: 3px; r: 11; }}
+  .new-badge {{
-    50%  {{ stroke-opacity: 0.4; stroke-width: 6px; r: 12; }}
+    font-size: 9px; font-weight: 700; color: #333; background: #FFD600;
-    100% {{ stroke-opacity: 1;   stroke-width: 3px; r: 11; }}
+    padding: 1px 4px; border-radius: 3px; white-space: nowrap;
+    box-shadow: 0 1px 3px rgba(0,0,0,0.3); letter-spacing: 0.5px;
   }}
-  .marker-new {{ animation: pulse-new 2s ease-in-out infinite; }}
   .info-panel {{
     position: absolute; top: 10px; right: 10px; z-index: 1000;
     background: white; padding: 16px; border-radius: 10px;
@@ -597,6 +656,10 @@ def generate_map(estates: list[dict], output_path: str = "mapa_bytu.html"):
   .info-panel .stats {{ color: #666; margin-bottom: 10px; padding-bottom: 10px; border-bottom: 1px solid #eee; }}
   .filter-section {{ margin-top: 10px; padding-top: 10px; border-top: 1px solid #eee; }}
   .filter-section label {{ display: flex; align-items: center; gap: 6px; margin: 3px 0; cursor: pointer; }}
+  .price-band {{ cursor: pointer; transition: background 0.12s; }}
+  .price-band:hover {{ background: #f0f0f0; }}
+  .price-band.active {{ border-color: #333 !important; background: #e8f0fe; }}
+  .price-band.dimmed {{ opacity: 0.35; }}
   .filter-section input[type="checkbox"] {{ accent-color: #1976D2; }}
   #floor-filter {{ margin-top: 8px; }}
   #floor-filter select {{ width: 100%; padding: 4px; border-radius: 4px; border: 1px solid #ccc; }}
@@ -635,11 +698,23 @@ def generate_map(estates: list[dict], output_path: str = "mapa_bytu.html"):
     </div>
     <div style="margin-top:6px;">
       <label>Max cena:
-        <select id="max-price" onchange="applyFilters()">
+        <input type="number" id="max-price" value="13500000" max="14000000" step="500000"
-          <option value="13500000">13 500 000 Kč</option>
+          style="width:130px;padding:2px 4px;border:1px solid #ccc;border-radius:3px;"
-          <option value="12000000">12 000 000 Kč</option>
+          onchange="applyFilters()" onkeyup="applyFilters()"> Kč
-          <option value="10000000">10 000 000 Kč</option>
+      </label>
-          <option value="8000000">8 000 000 Kč</option>
+    </div>
+    <div style="margin-top:6px;">
+      <label>Přidáno / změněno:
+        <select id="days-filter" onchange="applyFilters()" style="width:100%;padding:4px;border-radius:4px;border:1px solid #ccc;">
+          <option value="0">Vše</option>
+          <option value="1">za 1 den</option>
+          <option value="2">za 2 dny</option>
+          <option value="3">za 3 dny</option>
+          <option value="4">za 4 dny</option>
+          <option value="5">za 5 dní</option>
+          <option value="7">za 7 dní</option>
+          <option value="14">za 14 dní</option>
+          <option value="30">za 30 dní</option>
         </select>
       </label>
     </div>
@@ -653,7 +728,7 @@ def generate_map(estates: list[dict], output_path: str = "mapa_bytu.html"):
       Skrýt zamítnuté
     </label>
   </div>
-  <div class="status-link"><a href="status.html">Scraper status</a></div>
+  <div class="status-link"><a href="/scrapers-status">Scraper status</a></div>
 </div>
 
 <script>
@@ -673,9 +748,39 @@ L.tileLayer('https://{{s}}.basemaps.cartocdn.com/light_only_labels/{{z}}/{{x}}/{
   pane: 'shadowPane',
 }}).addTo(map);
 
+var selectedColors = [];
+
+function toggleColorFilter(color) {{
+  var idx = selectedColors.indexOf(color);
+  if (idx >= 0) selectedColors.splice(idx, 1);
+  else selectedColors.push(color);
+  document.querySelectorAll('.price-band').forEach(function(el) {{
+    var c = el.getAttribute('data-color');
+    if (selectedColors.length === 0) {{
+      el.classList.remove('active', 'dimmed');
+    }} else if (selectedColors.indexOf(c) >= 0) {{
+      el.classList.add('active'); el.classList.remove('dimmed');
+    }} else {{
+      el.classList.add('dimmed'); el.classList.remove('active');
+    }}
+  }});
+  document.getElementById('price-filter-reset').style.display =
+    selectedColors.length > 0 ? 'block' : 'none';
+  applyFilters();
+}}
+
+function resetColorFilter() {{
+  selectedColors = [];
+  document.querySelectorAll('.price-band').forEach(function(el) {{
+    el.classList.remove('active', 'dimmed');
+  }});
+  document.getElementById('price-filter-reset').style.display = 'none';
+  applyFilters();
+}}
+
 var allMarkers = [];
 
-function addMarker(lat, lon, color, popup, hashId) {{
+function addMarker(lat, lon, color, popup, hashId, firstSeen, lastChanged) {{
   var marker = L.circleMarker([lat, lon], {{
     radius: 8,
     fillColor: color,
@@ -684,26 +789,35 @@ function addMarker(lat, lon, color, popup, hashId) {{
     opacity: 1,
     fillOpacity: 0.85,
   }}).bindPopup(popup);
-  marker._data = {{ lat: lat, lon: lon, color: color, hashId: hashId }};
+  marker._data = {{ lat: lat, lon: lon, color: color, hashId: hashId, firstSeen: firstSeen || '', lastChanged: lastChanged || '' }};
   allMarkers.push(marker);
   marker.addTo(map);
 }}
 
-function addNewMarker(lat, lon, color, popup, hashId) {{
+function addNewMarker(lat, lon, color, popup, hashId, firstSeen, lastChanged) {{
   var marker = L.circleMarker([lat, lon], {{
-    radius: 12,
+    radius: 8,
     fillColor: color,
-    color: color,
+    color: '#fff',
-    weight: 4,
+    weight: 2,
-    opacity: 0.35,
+    opacity: 1,
-    fillOpacity: 0.95,
+    fillOpacity: 0.85,
   }}).bindPopup(popup);
-  marker._data = {{ lat: lat, lon: lon, color: color, hashId: hashId, isNew: true }};
+  marker._data = {{ lat: lat, lon: lon, color: color, hashId: hashId, isNew: true, firstSeen: firstSeen || '', lastChanged: lastChanged || '' }};
   allMarkers.push(marker);
   marker.addTo(map);
-  marker.on('add', function() {{
+  var badge = L.marker([lat, lon], {{
-    if (marker._path) marker._path.classList.add('marker-new');
+    icon: L.divIcon({{
+      className: 'new-badge-icon',
+      html: '<span class="new-badge">NEW</span>',
+      iconSize: [32, 14],
+      iconAnchor: [-6, 7],
+    }}),
+    interactive: false,
+    pane: 'markerPane',
   }});
+  badge.addTo(map);
+  marker._newBadge = badge;
 }}
 
 function heartIcon(color) {{
@@ -736,11 +850,11 @@ function starIcon() {{
   }});
 }}
 
-function addHeartMarker(lat, lon, color, popup, hashId) {{
+function addHeartMarker(lat, lon, color, popup, hashId, firstSeen, lastChanged) {{
   var marker = L.marker([lat, lon], {{
     icon: heartIcon(color),
   }}).bindPopup(popup);
-  marker._data = {{ lat: lat, lon: lon, color: color, hashId: hashId, isHeart: true }};
+  marker._data = {{ lat: lat, lon: lon, color: color, hashId: hashId, isHeart: true, firstSeen: firstSeen || '', lastChanged: lastChanged || '' }};
   allMarkers.push(marker);
   marker.addTo(map);
 }}
@@ -759,6 +873,11 @@ function loadRatings() {{
 
 function saveRatings(ratings) {{
   localStorage.setItem(RATINGS_KEY, JSON.stringify(ratings));
+  fetch('/api/ratings', {{
+    method: 'POST',
+    headers: {{'Content-Type': 'application/json'}},
+    body: JSON.stringify(ratings)
+  }}).catch(function() {{}});
 }}
 
 function addRejectStrike(marker) {{
@@ -806,6 +925,7 @@ function applyMarkerStyle(marker, status) {{
   }} else {{
     if (status === 'fav') {{
       removeRejectStrike(marker);
+      if (marker._newBadge && map.hasLayer(marker._newBadge)) map.removeLayer(marker._newBadge);
       if (!marker._data._origCircle) marker._data._origCircle = true;
       var popup = marker.getPopup();
       var popupContent = popup ? popup.getContent() : '';
@@ -829,6 +949,7 @@ function applyMarkerStyle(marker, status) {{
       }}
       // Add strikethrough line over the marker
       addRejectStrike(marker);
+      if (marker._newBadge && map.hasLayer(marker._newBadge)) map.removeLayer(marker._newBadge);
     }} else {{
       if (marker._data._origCircle && !(marker instanceof L.CircleMarker)) {{
         revertToCircle(marker, {{ radius: 8, fillColor: marker._data.color, color: '#fff', weight: 2, fillOpacity: 0.85 }});
@@ -841,6 +962,7 @@ function applyMarkerStyle(marker, status) {{
       }}
       if (marker._path) marker._path.classList.remove('marker-rejected');
       removeRejectStrike(marker);
+      if (marker._newBadge && !map.hasLayer(marker._newBadge)) marker._newBadge.addTo(map);
     }}
   }}
 }}
@@ -996,11 +1118,21 @@ map.on('popupopen', function(e) {{
 // ── Filters ────────────────────────────────────────────────────
 function applyFilters() {{
   var minFloor = parseInt(document.getElementById('min-floor').value);
-  var maxPrice = parseInt(document.getElementById('max-price').value);
+  var maxPriceEl = document.getElementById('max-price');
+  var maxPrice = parseInt(maxPriceEl.value) || 14000000;
+  if (maxPrice > 14000000) {{ maxPrice = 14000000; maxPriceEl.value = 14000000; }}
   var hideRejected = document.getElementById('hide-rejected').checked;
+  var daysFilter = parseInt(document.getElementById('days-filter').value) || 0;
   var ratings = loadRatings();
   var visible = 0;
 
+  var cutoff = null;
+  if (daysFilter > 0) {{
+    cutoff = new Date();
+    cutoff.setDate(cutoff.getDate() - daysFilter);
+    cutoff.setHours(0, 0, 0, 0);
+  }}
+
   allMarkers.forEach(function(m) {{
     var popup = m.getPopup().getContent();
     var floorMatch = popup.match(/(\\d+)\\. NP/);
@@ -1013,6 +1145,14 @@ function applyFilters() {{
     if (floor !== null && floor < minFloor) show = false;
     if (price > maxPrice) show = false;
 
+    if (cutoff) {{
+      var fs = m._data.firstSeen ? new Date(m._data.firstSeen) : null;
+      var lc = m._data.lastChanged ? new Date(m._data.lastChanged) : null;
+      if (!((fs && fs >= cutoff) || (lc && lc >= cutoff))) show = false;
+    }}
+
+    if (selectedColors.length > 0 && selectedColors.indexOf(m._data.color) < 0) show = false;
+
     var r = ratings[m._data.hashId];
     if (hideRejected && r && r.status === 'reject') show = false;
 
@@ -1021,10 +1161,12 @@ function applyFilters() {{
       visible++;
       // Show strike line if rejected and visible
       if (m._rejectStrike && !map.hasLayer(m._rejectStrike)) m._rejectStrike.addTo(map);
+      if (m._newBadge && !map.hasLayer(m._newBadge)) m._newBadge.addTo(map);
     }} else {{
       if (map.hasLayer(m)) map.removeLayer(m);
       // Hide strike line when marker hidden
       if (m._rejectStrike && map.hasLayer(m._rejectStrike)) map.removeLayer(m._rejectStrike);
+      if (m._newBadge && map.hasLayer(m._newBadge)) map.removeLayer(m._newBadge);
     }}
   }});
 
@@ -1039,8 +1181,25 @@ function applyFilters() {{
   document.getElementById('visible-count').textContent = visible;
 }}
 
-// Initialize ratings on load
+// Initialize ratings: load from server, merge with localStorage, then restore
-restoreRatings();
+function initRatings() {{
+  var local = loadRatings();
+  fetch('/api/ratings')
+    .then(function(r) {{ return r.ok ? r.json() : null; }})
+    .then(function(server) {{
+      if (server && typeof server === 'object') {{
+        var merged = Object.assign({{}}, local, server);
+        localStorage.setItem(RATINGS_KEY, JSON.stringify(merged));
+      }}
+      restoreRatings();
+      updateRatingCounts();
+    }})
+    .catch(function() {{
+      restoreRatings();
+      updateRatingCounts();
+    }});
+}}
+initRatings();
 
 // ── Panel toggle ──────────────────────────────────────────────
 function togglePanel() {{
@@ -1089,8 +1248,22 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    try:
+        estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    except Exception as e:
+        logger.error(f"Scraper failed: {e}", exc_info=True)
+        write_stats(STATS_FILE, {
+            "source": "Sreality",
+            "timestamp": _run_ts,
+            "duration_sec": round(time.time() - start, 1),
+            "success": False,
+            "accepted": 0,
+            "fetched": 0,
+            "error": str(e),
+        })
+        raise
 
     if estates:
         # Save raw data as JSON backup

scrape_bazos.py

@@ -0,0 +1,560 @@
+#!/usr/bin/env python3
+"""
+Bazoš.cz scraper.
+Stáhne byty na prodej v Praze a vyfiltruje podle kritérií.
+Výstup: byty_bazos.json
+"""
+from __future__ import annotations
+
+import argparse
+from datetime import datetime
+import json
+import logging
+import math
+import re
+import time
+import urllib.request
+import urllib.parse
+from pathlib import Path
+from scraper_stats import write_stats, validate_listing
+
+STATS_FILE = "stats_bazos.json"
+
+logger = logging.getLogger(__name__)
+
+# ── Konfigurace ─────────────────────────────────────────────────────────────
+
+MAX_PRICE = 14_000_000
+MIN_AREA = 69
+MIN_FLOOR = 2
+PER_PAGE = 20  # Bazoš vrací 20 na stránku
+
+WANTED_DISPOSITIONS = {"3+kk", "3+1", "4+kk", "4+1", "5+kk", "5+1", "6+kk", "6+1"}
+
+# Regex patterns pro parsování dispozice, plochy a patra z textu
+DISP_RE = re.compile(r'(\d)\s*\+\s*(kk|1)', re.IGNORECASE)
+AREA_RE = re.compile(r'(\d+(?:[.,]\d+)?)\s*m[²2\s,.]', re.IGNORECASE)
+FLOOR_RE = re.compile(r'(\d+)\s*[./]\s*(\d+)\s*(?:NP|patr|podlaž|floor)', re.IGNORECASE)
+FLOOR_RE2 = re.compile(r'(\d+)\.\s*(?:NP|patr[eouě]|podlaž[ií])', re.IGNORECASE)
+FLOOR_RE3 = re.compile(r'(?:patr[eouě]|podlaž[ií]|NP)\s*[:\s]*(\d+)', re.IGNORECASE)
+PANEL_RE = re.compile(r'panel(?:ov|ák|\.)', re.IGNORECASE)
+SIDLISTE_RE = re.compile(r'sídliště|sidliste|panelák', re.IGNORECASE)
+
+HEADERS = {
+    "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
+    "Accept": "text/html,application/xhtml+xml",
+    "Accept-Language": "cs,en;q=0.9",
+}
+
+BASE_URL = "https://reality.bazos.cz"
+SEARCH_PARAMS = "hledat=&rubriky=reality&hlokalita=Praha&humkreis=25&cenado={max_price}&kitx=ano"
+
+
+def fetch_url(url: str, retries: int = 3) -> str:
+    """Fetch URL and return HTML string with retry on transient errors."""
+    for attempt in range(retries):
+        try:
+            logger.debug(f"HTTP GET request (attempt {attempt + 1}/{retries}): {url}")
+            req = urllib.request.Request(url, headers=HEADERS)
+            resp = urllib.request.urlopen(req, timeout=30)
+            html = resp.read().decode("utf-8", errors="replace")
+            logger.debug(f"HTTP response: status={resp.status}, size={len(html)} bytes")
+            return html
+        except urllib.error.HTTPError:
+            raise
+        except (ConnectionResetError, ConnectionError, urllib.error.URLError, OSError) as e:
+            if attempt < retries - 1:
+                wait = (attempt + 1) * 3
+                logger.warning(f"Connection error (retry {attempt + 1}/{retries} after {wait}s): {e}")
+                time.sleep(wait)
+            else:
+                logger.error(f"HTTP request failed after {retries} attempts: {e}", exc_info=True)
+                raise
+
+
+def format_price(price: int) -> str:
+    s = str(price)
+    parts = []
+    while s:
+        parts.append(s[-3:])
+        s = s[:-3]
+    return " ".join(reversed(parts)) + " Kč"
+
+
+def parse_price(text: str) -> int:
+    """Parse price from text like '5 250 000 Kč' → 5250000."""
+    cleaned = re.sub(r'[^\d]', '', text)
+    return int(cleaned) if cleaned else 0
+
+
+def parse_disposition(text: str) -> str | None:
+    """Parse disposition from title/description like '3+kk', '4+1'."""
+    m = DISP_RE.search(text)
+    if m:
+        rooms = m.group(1)
+        suffix = m.group(2).lower()
+        return f"{rooms}+{suffix}"
+    return None
+
+
+def parse_area(text: str) -> float | None:
+    """Parse area from text like '82 m²' → 82.0."""
+    m = AREA_RE.search(text)
+    if m:
+        return float(m.group(1).replace(',', '.'))
+    return None
+
+
+def parse_floor(text: str) -> int | None:
+    """Parse floor number from description."""
+    for pattern in [FLOOR_RE, FLOOR_RE2, FLOOR_RE3]:
+        m = pattern.search(text)
+        if m:
+            return int(m.group(1))
+    return None
+
+
+def is_panel(text: str) -> bool:
+    """Check if description mentions panel construction."""
+    return bool(PANEL_RE.search(text))
+
+
+def is_sidliste(text: str) -> bool:
+    """Check if description mentions housing estate."""
+    return bool(SIDLISTE_RE.search(text))
+
+
+def fetch_listing_page(offset: int = 0, pagination_params: str | None = None) -> tuple[list[dict], int, str | None]:
+    """
+    Fetch a page of listings from Bazoš.
+    Returns (list of basic listing dicts, total count, pagination_params for next pages).
+    """
+    if pagination_params and offset > 0:
+        # Use resolved numeric params from first page's pagination links
+        url = f"{BASE_URL}/prodam/byt/{offset}/?{pagination_params}"
+    else:
+        params = SEARCH_PARAMS.format(max_price=MAX_PRICE)
+        if offset > 0:
+            url = f"{BASE_URL}/prodam/byt/{offset}/?{params}"
+        else:
+            url = f"{BASE_URL}/prodam/byt/?{params}"
+
+    html = fetch_url(url)
+
+    # Parse total count: "Zobrazeno 1-20 z 727"
+    total = 0
+    total_match = re.search(r'z\s+([\d\s]+)\s', html)
+    if total_match:
+        total = int(total_match.group(1).replace(' ', ''))
+
+    # Extract resolved pagination params from first page (Bazoš converts
+    # hlokalita=Praha → hlokalita=11000, and pagination only works with numeric form)
+    resolved_params = None
+    pag_link = re.search(r'href="/prodam/byt/\d+/\?([^"]+)"', html)
+    if pag_link:
+        resolved_params = pag_link.group(1)
+
+    # Parse listings — split by listing blocks (class="inzeraty inzeratyflex")
+    listings = []
+    all_blocks = re.split(r'<div class="inzeraty\s+inzeratyflex">', html)[1:]  # skip before first
+
+    for block in all_blocks:
+        # Extract URL and ID from first link (/inzerat/XXXXXX/slug.php)
+        url_match = re.search(r'href="(/inzerat/(\d+)/[^"]*)"', block)
+        if not url_match:
+            continue
+        detail_path = url_match.group(1)
+        listing_id = int(url_match.group(2))
+
+        # Title — class=nadpis (without quotes) or class="nadpis"
+        title_match = re.search(r'class=.?nadpis.?[^>]*>\s*<a[^>]*>([^<]+)</a>', block)
+        title = title_match.group(1).strip() if title_match else ""
+
+        # Price — inside <span translate="no"> within inzeratycena
+        price_match = re.search(r'class="inzeratycena"[^>]*>.*?<span[^>]*>([^<]+)</span>', block, re.DOTALL)
+        if not price_match:
+            # Fallback: direct text in inzeratycena
+            price_match = re.search(r'class="inzeratycena"[^>]*>\s*(?:<b>)?([^<]+)', block)
+        price_text = price_match.group(1).strip() if price_match else ""
+        price = parse_price(price_text)
+
+        # Location
+        loc_match = re.search(r'class="inzeratylok"[^>]*>(.*?)</div>', block, re.DOTALL)
+        location = ""
+        if loc_match:
+            location = re.sub(r'<[^>]+>', ' ', loc_match.group(1)).strip()
+            location = re.sub(r'\s+', ' ', location)
+
+        # Date — [5.3. 2026]
+        date_match = re.search(r'\[(\d+\.\d+\.\s*\d{4})\]', block)
+        date_str = date_match.group(1).strip() if date_match else ""
+
+        # Description preview — class=popis (without quotes) or class="popis"
+        desc_match = re.search(r'class=.?popis.?[^>]*>(.*?)</div>', block, re.DOTALL)
+        description = ""
+        if desc_match:
+            description = re.sub(r'<[^>]+>', ' ', desc_match.group(1)).strip()
+            description = re.sub(r'\s+', ' ', description)
+
+        # Image — <img ... class="obrazek" ... src="...">
+        img_match = re.search(r'<img[^>]*src="([^"]+)"[^>]*class="obrazek"', block)
+        if not img_match:
+            img_match = re.search(r'class="obrazek"[^>]*src="([^"]+)"', block)
+        image = img_match.group(1) if img_match else ""
+        if "empty.gif" in image:
+            image = ""
+
+        listings.append({
+            "id": listing_id,
+            "title": title,
+            "price": price,
+            "location": location,
+            "date": date_str,
+            "description": description,
+            "detail_path": detail_path,
+            "image": image,
+        })
+
+    logger.debug(f"Offset {offset}: found {len(listings)} listings, total={total}")
+    return listings, total, resolved_params
+
+
+def fetch_detail(path: str) -> dict | None:
+    """Fetch listing detail page and extract GPS, full description."""
+    try:
+        url = f"{BASE_URL}{path}"
+        html = fetch_url(url)
+
+        result = {}
+
+        # GPS from Google Maps link
+        gps_match = re.search(r'google\.com/maps[^"]*place/([\d.]+),([\d.]+)', html)
+        if gps_match:
+            result["lat"] = float(gps_match.group(1))
+            result["lon"] = float(gps_match.group(2))
+
+        # Full description — Bazoš uses unquoted class=popisdetail
+        desc_match = re.search(r'class=.?popisdetail.?[^>]*>(.*?)</div>', html, re.DOTALL)
+        if desc_match:
+            desc = re.sub(r'<[^>]+>', ' ', desc_match.group(1)).strip()
+            desc = re.sub(r'\s+', ' ', desc)
+            result["description"] = desc
+
+        # Location from detail
+        loc_match = re.search(r'Lokalita:</td>\s*<td[^>]*>(.*?)</td>', html, re.DOTALL)
+        if loc_match:
+            loc = re.sub(r'<[^>]+>', ' ', loc_match.group(1)).strip()
+            loc = re.sub(r'\s+', ' ', loc)
+            result["detail_location"] = loc
+
+        return result
+
+    except Exception as e:
+        logger.warning(f"Detail fetch failed for {path}: {e}")
+        return None
+
+
+def load_cache(json_path: str = "byty_bazos.json") -> dict[int, dict]:
+    """Load previously scraped data as cache keyed by hash_id."""
+    path = Path(json_path)
+    if not path.exists():
+        return {}
+    try:
+        data = json.loads(path.read_text(encoding="utf-8"))
+        return {e["hash_id"]: e for e in data if "hash_id" in e}
+    except (json.JSONDecodeError, KeyError):
+        return {}
+
+
+def scrape(max_pages: int | None = None, max_properties: int | None = None):
+    _run_start = time.time()
+    _run_ts = datetime.now().isoformat(timespec="seconds")
+    cache = load_cache()
+    today = datetime.now().strftime("%Y-%m-%d")
+
+    logger.info("=" * 60)
+    logger.info("Stahuji inzeráty z Bazoš.cz")
+    logger.info(f"Cena: do {format_price(MAX_PRICE)}")
+    logger.info(f"Min. plocha: {MIN_AREA} m²")
+    logger.info(f"Patro: od {MIN_FLOOR}. NP")
+    logger.info(f"Region: Praha")
+    if cache:
+        logger.info(f"Cache: {len(cache)} bytů z minulého běhu")
+    if max_pages:
+        logger.info(f"Max. stran: {max_pages}")
+    if max_properties:
+        logger.info(f"Max. bytů: {max_properties}")
+    logger.info("=" * 60)
+
+    # Step 1: Fetch listing pages
+    logger.info("\nFáze 1: Stahování seznamu inzerátů...")
+    all_listings = {}  # id -> listing dict (dedup)
+    page = 1
+    offset = 0
+    total = None
+    pagination_params = None  # resolved numeric params from first page
+
+    while True:
+        if max_pages and page > max_pages:
+            logger.debug(f"Max pages limit reached: {max_pages}")
+            break
+
+        logger.info(f"Strana {page} (offset {offset}) ...")
+        listings, total_count, resolved = fetch_listing_page(offset, pagination_params)
+        if resolved and not pagination_params:
+            pagination_params = resolved
+            logger.debug(f"Resolved pagination params: {pagination_params}")
+
+        if total is None and total_count > 0:
+            total = total_count
+            total_pages = math.ceil(total / PER_PAGE)
+            logger.info(f"→ Celkem {total} inzerátů, ~{total_pages} stran")
+
+        if not listings:
+            logger.debug(f"No listings found on page {page}, stopping")
+            break
+
+        for lst in listings:
+            lid = lst["id"]
+            if lid not in all_listings:
+                all_listings[lid] = lst
+
+        page += 1
+        offset += PER_PAGE
+        if total and offset >= total:
+            break
+        time.sleep(0.5)
+
+    logger.info(f"\nStaženo: {len(all_listings)} unikátních inzerátů")
+
+    # Step 2: Pre-filter by disposition, price, area from listing data
+    pre_filtered = []
+    excluded_disp = 0
+    excluded_price = 0
+    excluded_area = 0
+    excluded_no_disp = 0
+
+    for lst in all_listings.values():
+        title_and_desc = f"{lst['title']} {lst['description']}"
+
+        # Parse disposition
+        disp = parse_disposition(title_and_desc)
+        if not disp:
+            excluded_no_disp += 1
+            logger.debug(f"Filter: id={lst['id']} - excluded (no disposition found in '{lst['title']}')")
+            continue
+        if disp not in WANTED_DISPOSITIONS:
+            excluded_disp += 1
+            logger.debug(f"Filter: id={lst['id']} - excluded (disposition {disp})")
+            continue
+
+        # Price
+        price = lst["price"]
+        if price <= 0 or price > MAX_PRICE:
+            excluded_price += 1
+            logger.debug(f"Filter: id={lst['id']} - excluded (price {price})")
+            continue
+
+        # Area (if parseable from listing)
+        area = parse_area(title_and_desc)
+        if area is not None and area < MIN_AREA:
+            excluded_area += 1
+            logger.debug(f"Filter: id={lst['id']} - excluded (area {area} m²)")
+            continue
+
+        lst["_disposition"] = disp
+        lst["_area"] = area
+        pre_filtered.append(lst)
+
+    logger.info(f"\nPo předfiltraci:")
+    logger.info(f"  Vyloučeno (bez dispozice): {excluded_no_disp}")
+    logger.info(f"  Vyloučeno (dispozice):     {excluded_disp}")
+    logger.info(f"  Vyloučeno (cena):          {excluded_price}")
+    logger.info(f"  Vyloučeno (plocha):        {excluded_area}")
+    logger.info(f"  Zbývá:                     {len(pre_filtered)}")
+
+    # Step 3: Fetch details (for GPS + full description)
+    logger.info(f"\nFáze 2: Stahování detailů ({len(pre_filtered)} bytů)...")
+    results = []
+    excluded_panel = 0
+    excluded_floor = 0
+    excluded_no_gps = 0
+    excluded_detail = 0
+    excluded_area_detail = 0
+    cache_hits = 0
+    properties_fetched = 0
+
+    for i, lst in enumerate(pre_filtered):
+        if max_properties and properties_fetched >= max_properties:
+            logger.debug(f"Max properties limit reached: {max_properties}")
+            break
+
+        listing_id = lst["id"]
+        price = lst["price"]
+
+        # Check cache
+        cached = cache.get(listing_id)
+        if cached and cached.get("price") == price:
+            cache_hits += 1
+            logger.debug(f"Cache hit for id={listing_id}")
+            results.append(cached)
+            continue
+
+        time.sleep(0.4)
+        detail = fetch_detail(lst["detail_path"])
+
+        if not detail:
+            excluded_detail += 1
+            logger.debug(f"Filter: id={listing_id} - excluded (detail fetch failed)")
+            continue
+
+        # GPS required
+        lat = detail.get("lat")
+        lon = detail.get("lon")
+        if not lat or not lon:
+            excluded_no_gps += 1
+            logger.debug(f"Filter: id={listing_id} - excluded (no GPS)")
+            continue
+
+        # Full text for filtering
+        full_desc = detail.get("description", "")
+        full_text = f"{lst['title']} {lst['description']} {full_desc}"
+
+        # Panel check
+        if is_panel(full_text):
+            excluded_panel += 1
+            logger.info(f"✗ Vyloučen #{listing_id}: panelová stavba")
+            continue
+
+        # Sídliště check
+        if is_sidliste(full_text):
+            excluded_panel += 1
+            logger.info(f"✗ Vyloučen #{listing_id}: sídliště")
+            continue
+
+        # Floor
+        floor = parse_floor(full_text)
+        if floor is not None and floor < MIN_FLOOR:
+            excluded_floor += 1
+            logger.debug(f"Filter: id={listing_id} - excluded (floor {floor})")
+            continue
+
+        # Area — re-check from detail if not found before
+        area = lst.get("_area") or parse_area(full_desc)
+        if area is not None and area < MIN_AREA:
+            excluded_area_detail += 1
+            logger.debug(f"Filter: id={listing_id} - excluded (area {area} m² from detail)")
+            continue
+
+        disp = lst["_disposition"]
+        locality = detail.get("detail_location") or lst["location"]
+
+        result = {
+            "hash_id": listing_id,
+            "name": f"Prodej bytu {disp} {int(area) if area else '?'} m²",
+            "price": price,
+            "price_formatted": format_price(price),
+            "locality": locality,
+            "lat": lat,
+            "lon": lon,
+            "disposition": disp,
+            "floor": floor,
+            "area": area,
+            "building_type": "neuvedeno",
+            "ownership": "neuvedeno",
+            "url": f"{BASE_URL}{lst['detail_path']}",
+            "source": "bazos",
+            "image": lst.get("image", ""),
+            "scraped_at": today,
+            "first_seen": cached.get("first_seen", today) if cached else today,
+            "last_changed": today if not cached or cached.get("price") != price else cached.get("last_changed", today),
+        }
+
+        if not validate_listing(result, "bazos"):
+            continue
+
+        results.append(result)
+        properties_fetched += 1
+
+        if (i + 1) % 20 == 0:
+            logger.info(f"Zpracováno {i + 1}/{len(pre_filtered)} ...")
+
+    logger.info(f"\n{'=' * 60}")
+    logger.info(f"Výsledky Bazoš:")
+    logger.info(f"  Předfiltrováno:        {len(pre_filtered)}")
+    logger.info(f"  Z cache (přeskočeno): {cache_hits}")
+    logger.info(f"  Vyloučeno (panel/síd): {excluded_panel}")
+    logger.info(f"  Vyloučeno (patro):     {excluded_floor}")
+    logger.info(f"  Vyloučeno (bez GPS):   {excluded_no_gps}")
+    logger.info(f"  Vyloučeno (bez detailu): {excluded_detail}")
+    logger.info(f"  Vyloučeno (plocha det): {excluded_area_detail}")
+    logger.info(f"  ✓ Vyhovující byty:    {len(results)}")
+    logger.info(f"{'=' * 60}")
+
+    write_stats(STATS_FILE, {
+        "source": "Bazoš",
+        "timestamp": _run_ts,
+        "duration_sec": round(time.time() - _run_start, 1),
+        "success": True,
+        "accepted": len(results),
+        "fetched": len(all_listings),
+        "pages": page - 1,
+        "cache_hits": cache_hits,
+        "excluded": {
+            "bez dispozice": excluded_no_disp,
+            "dispozice": excluded_disp,
+            "cena": excluded_price,
+            "plocha": excluded_area + excluded_area_detail,
+            "bez GPS": excluded_no_gps,
+            "panel/síd": excluded_panel,
+            "patro": excluded_floor,
+            "bez detailu": excluded_detail,
+        },
+    })
+    return results
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Scrape apartments from Bazoš.cz")
+    parser.add_argument("--max-pages", type=int, default=None,
+                        help="Maximum number of listing pages to scrape")
+    parser.add_argument("--max-properties", type=int, default=None,
+                        help="Maximum number of properties to fetch details for")
+    parser.add_argument("--log-level", type=str, default="INFO", choices=["DEBUG", "INFO", "WARNING", "ERROR"],
+                        help="Logging level (default: INFO)")
+    args = parser.parse_args()
+
+    logging.basicConfig(
+        level=getattr(logging, args.log_level),
+        format="[%(levelname)s] %(asctime)s - %(name)s - %(message)s",
+        handlers=[logging.StreamHandler()]
+    )
+
+    _run_ts = datetime.now().isoformat(timespec="seconds")
+    start = time.time()
+    try:
+        estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    except Exception as e:
+        logger.error(f"Scraper failed: {e}", exc_info=True)
+        write_stats(STATS_FILE, {
+            "source": "Bazoš",
+            "timestamp": _run_ts,
+            "duration_sec": round(time.time() - start, 1),
+            "success": False,
+            "accepted": 0,
+            "fetched": 0,
+            "error": str(e),
+        })
+        raise
+
+    if estates:
+        json_path = Path("byty_bazos.json")
+        json_path.write_text(
+            json.dumps(estates, ensure_ascii=False, indent=2),
+            encoding="utf-8",
+        )
+        elapsed = time.time() - start
+        logger.info(f"\n✓ Data uložena: {json_path.resolve()}")
+        logger.info(f"⏱  Celkový čas: {elapsed:.0f} s")
+    else:
+        logger.info("\nŽádné byty z Bazoše neodpovídají kritériím :(")

scrape_bezrealitky.py

@@ -15,6 +15,9 @@ import re
 import time
 import urllib.request
 from pathlib import Path
+from scraper_stats import write_stats, validate_listing
+
+STATS_FILE = "stats_bezrealitky.json"
 
 logger = logging.getLogger(__name__)
 
@@ -68,62 +71,71 @@ HEADERS = {
 BASE_URL = "https://www.bezrealitky.cz"
 
 
+def fetch_url(url: str, retries: int = 3) -> str:
+    """Fetch URL and return HTML string with retry on transient errors."""
+    for attempt in range(retries):
+        try:
+            logger.debug(f"HTTP GET request (attempt {attempt + 1}/{retries}): {url}")
+            req = urllib.request.Request(url, headers=HEADERS)
+            resp = urllib.request.urlopen(req, timeout=30)
+            html = resp.read().decode("utf-8")
+            logger.debug(f"HTTP response: status={resp.status}, size={len(html)} bytes")
+            return html
+        except urllib.error.HTTPError:
+            raise
+        except (ConnectionResetError, ConnectionError, urllib.error.URLError, OSError) as e:
+            if attempt < retries - 1:
+                wait = (attempt + 1) * 2
+                logger.warning(f"Connection error (retry {attempt + 1}/{retries} after {wait}s): {e}")
+                time.sleep(wait)
+            else:
+                logger.error(f"HTTP request failed after {retries} attempts: {e}", exc_info=True)
+                raise
+
+
 def fetch_page(page: int) -> tuple[list[dict], int]:
     """
     Fetch a listing page from Bezrealitky.
     Returns (list of advert dicts from Apollo cache, total count).
     """
     url = f"{BASE_URL}/vypis/nabidka-prodej/byt/praha?page={page}"
-    logger.debug(f"HTTP GET request: {url}")
+    html = fetch_url(url)
-    logger.debug(f"Headers: {HEADERS}")
-    req = urllib.request.Request(url, headers=HEADERS)
-    try:
-        resp = urllib.request.urlopen(req, timeout=30)
-        html = resp.read().decode("utf-8")
-        logger.debug(f"HTTP response: status={resp.status}, size={len(html)} bytes")
 
-        match = re.search(
+    match = re.search(
-            r'<script id="__NEXT_DATA__" type="application/json">(.*?)</script>',
+        r'<script id="__NEXT_DATA__" type="application/json">(.*?)</script>',
-            html, re.DOTALL
+        html, re.DOTALL
-        )
+    )
-        if not match:
+    if not match:
-            logger.debug("No __NEXT_DATA__ script found in HTML")
+        logger.debug("No __NEXT_DATA__ script found in HTML")
-            return [], 0
+        return [], 0
 
-        data = json.loads(match.group(1))
+    data = json.loads(match.group(1))
-        cache = data["props"]["pageProps"]["apolloCache"]
+    cache = data["props"]["pageProps"]["apolloCache"]
 
-        # Extract adverts from cache
+    # Extract adverts from cache
-        adverts = []
+    adverts = []
-        for key, val in cache.items():
+    for key, val in cache.items():
-            if key.startswith("Advert:") and isinstance(val, dict) and val.get("__typename") == "Advert":
+        if key.startswith("Advert:") and isinstance(val, dict) and val.get("__typename") == "Advert":
-                adverts.append(val)
+            adverts.append(val)
 
-        # Get total count from ROOT_QUERY
+    # Get total count from ROOT_QUERY
-        total = 0
+    total = 0
-        root = cache.get("ROOT_QUERY", {})
+    root = cache.get("ROOT_QUERY", {})
-        for key, val in root.items():
+    for key, val in root.items():
-            if "listAdverts" in key and isinstance(val, dict):
+        if "listAdverts" in key and isinstance(val, dict):
-                tc = val.get("totalCount")
+            tc = val.get("totalCount")
-                if tc and tc > total:
+            if tc and tc > total:
-                    total = tc
+                total = tc
 
-        logger.debug(f"Page {page}: found {len(adverts)} adverts, total={total}")
+    logger.debug(f"Page {page}: found {len(adverts)} adverts, total={total}")
-        return adverts, total
+    return adverts, total
-    except (urllib.error.URLError, ConnectionError, OSError) as e:
-        logger.error(f"HTTP request failed for {url}: {e}", exc_info=True)
-        raise
 
 
 def fetch_detail(uri: str) -> dict | None:
     """Fetch detail page for a listing."""
     try:
         url = f"{BASE_URL}/nemovitosti-byty-domy/{uri}"
-        logger.debug(f"HTTP GET request: {url}")
+        html = fetch_url(url)
-        req = urllib.request.Request(url, headers=HEADERS)
-        resp = urllib.request.urlopen(req, timeout=30)
-        html = resp.read().decode("utf-8")
-        logger.debug(f"HTTP response: status={resp.status}, size={len(html)} bytes")
 
         match = re.search(
             r'<script id="__NEXT_DATA__" type="application/json">(.*?)</script>',
@@ -171,6 +183,8 @@ def load_cache(json_path: str = "byty_bezrealitky.json") -> dict[int, dict]:
 
 
 def scrape(max_pages: int | None = None, max_properties: int | None = None):
+    _run_start = time.time()
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     cache = load_cache()
 
     logger.info("=" * 60)
@@ -357,7 +371,11 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
             "source": "bezrealitky",
             "image": "",
             "scraped_at": datetime.now().strftime("%Y-%m-%d"),
+            "first_seen": cached.get("first_seen", datetime.now().strftime("%Y-%m-%d")) if cached else datetime.now().strftime("%Y-%m-%d"),
+            "last_changed": datetime.now().strftime("%Y-%m-%d"),
         }
+        if not validate_listing(result, "bezrealitky"):
+            continue
         results.append(result)
         properties_fetched += 1
 
@@ -374,6 +392,25 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
     logger.info(f"  ✓ Vyhovující byty:    {len(results)}")
     logger.info(f"{'=' * 60}")
 
+    write_stats(STATS_FILE, {
+        "source": "Bezrealitky",
+        "timestamp": _run_ts,
+        "duration_sec": round(time.time() - _run_start, 1),
+        "success": True,
+        "accepted": len(results),
+        "fetched": len(all_adverts),
+        "pages": page - 1,
+        "cache_hits": cache_hits,
+        "excluded": {
+            "dispozice": excluded_disp,
+            "cena": excluded_price,
+            "plocha": excluded_area,
+            "bez GPS": excluded_no_gps,
+            "panel/síd": excluded_panel,
+            "patro": excluded_floor,
+            "bez detailu": excluded_detail,
+        },
+    })
     return results
 
 
@@ -394,8 +431,22 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    try:
+        estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    except Exception as e:
+        logger.error(f"Scraper failed: {e}", exc_info=True)
+        write_stats(STATS_FILE, {
+            "source": "Bezrealitky",
+            "timestamp": _run_ts,
+            "duration_sec": round(time.time() - start, 1),
+            "success": False,
+            "accepted": 0,
+            "fetched": 0,
+            "error": str(e),
+        })
+        raise
 
     if estates:
         json_path = Path("byty_bezrealitky.json")

scrape_cityhome.py

@@ -14,6 +14,9 @@ import time
 import urllib.request
 from datetime import datetime
 from pathlib import Path
+from scraper_stats import write_stats, validate_listing
+
+STATS_FILE = "stats_cityhome.json"
 
 logger = logging.getLogger(__name__)
 
@@ -203,6 +206,8 @@ def extract_project_gps(html: str) -> tuple[float, float] | None:
 
 
 def scrape(max_pages: int | None = None, max_properties: int | None = None):
+    _run_start = time.time()
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     logger.info("=" * 60)
     logger.info("Stahuji inzeráty z CityHome (city-home.cz)")
     logger.info(f"Cena: do {format_price(MAX_PRICE)}")
@@ -250,6 +255,16 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
         else:
             logger.info(f"✗ {slug}: GPS nenalezeno")
 
+    # Load previous output for first_seen/last_changed tracking
+    _prev_cache: dict[str, dict] = {}
+    _prev_path = Path("byty_cityhome.json")
+    if _prev_path.exists():
+        try:
+            for _item in json.loads(_prev_path.read_text(encoding="utf-8")):
+                _prev_cache[str(_item["hash_id"])] = _item
+        except Exception:
+            pass
+
     # Step 3: Filter listings
     logger.info(f"\nFáze 3: Filtrování...")
     results = []
@@ -357,7 +372,11 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
             "source": "cityhome",
             "image": "",
             "scraped_at": datetime.now().strftime("%Y-%m-%d"),
+            "first_seen": _prev_cache.get(f"cityhome_{slug}_{listing['unit_name']}", {}).get("first_seen", datetime.now().strftime("%Y-%m-%d")),
+            "last_changed": datetime.now().strftime("%Y-%m-%d") if _prev_cache.get(f"cityhome_{slug}_{listing['unit_name']}", {}).get("price") != price else _prev_cache[f"cityhome_{slug}_{listing['unit_name']}"].get("last_changed", datetime.now().strftime("%Y-%m-%d")),
         }
+        if not validate_listing(result, "cityhome"):
+            continue
         results.append(result)
         properties_fetched += 1
 
@@ -374,6 +393,23 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
     logger.info(f"  ✓ Vyhovující byty:    {len(results)}")
     logger.info(f"{'=' * 60}")
 
+    write_stats(STATS_FILE, {
+        "source": "CityHome",
+        "timestamp": _run_ts,
+        "duration_sec": round(time.time() - _run_start, 1),
+        "success": True,
+        "accepted": len(results),
+        "fetched": len(all_listings),
+        "excluded": {
+            "prodáno": excluded_sold,
+            "typ": excluded_type,
+            "dispozice": excluded_disp,
+            "cena": excluded_price,
+            "plocha": excluded_area,
+            "patro": excluded_floor,
+            "bez GPS": excluded_no_gps,
+        },
+    })
     return results
 
 
@@ -394,8 +430,22 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    try:
+        estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    except Exception as e:
+        logger.error(f"Scraper failed: {e}", exc_info=True)
+        write_stats(STATS_FILE, {
+            "source": "CityHome",
+            "timestamp": _run_ts,
+            "duration_sec": round(time.time() - start, 1),
+            "success": False,
+            "accepted": 0,
+            "fetched": 0,
+            "error": str(e),
+        })
+        raise
 
     if estates:
         json_path = Path("byty_cityhome.json")

scrape_idnes.py

@@ -15,8 +15,10 @@ import re
 import time
 import urllib.request
 import urllib.parse
-from html.parser import HTMLParser
 from pathlib import Path
+from scraper_stats import write_stats, validate_listing
+
+STATS_FILE = "stats_idnes.json"
 
 logger = logging.getLogger(__name__)
 
@@ -279,6 +281,8 @@ def load_cache(json_path: str = "byty_idnes.json") -> dict[str, dict]:
 
 
 def scrape(max_pages: int | None = None, max_properties: int | None = None):
+    _run_start = time.time()
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     cache = load_cache()
 
     logger.info("=" * 60)
@@ -460,7 +464,11 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
             "source": "idnes",
             "image": "",
             "scraped_at": datetime.now().strftime("%Y-%m-%d"),
+            "first_seen": cached.get("first_seen", datetime.now().strftime("%Y-%m-%d")) if cached else datetime.now().strftime("%Y-%m-%d"),
+            "last_changed": datetime.now().strftime("%Y-%m-%d"),
         }
+        if not validate_listing(result, "idnes"):
+            continue
         results.append(result)
         properties_fetched += 1
 
@@ -478,6 +486,25 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
     logger.info(f"  ✓ Vyhovující byty:    {len(results)}")
     logger.info(f"{'=' * 60}")
 
+    write_stats(STATS_FILE, {
+        "source": "iDNES",
+        "timestamp": _run_ts,
+        "duration_sec": round(time.time() - _run_start, 1),
+        "success": True,
+        "accepted": len(results),
+        "fetched": len(all_listings),
+        "pages": page,
+        "cache_hits": cache_hits,
+        "excluded": {
+            "cena": excluded_price,
+            "plocha": excluded_area,
+            "dispozice": excluded_disp,
+            "panel/síd": excluded_panel,
+            "patro": excluded_floor,
+            "bez GPS": excluded_no_gps,
+            "bez detailu": excluded_detail,
+        },
+    })
     return results
 
 
@@ -498,8 +525,22 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    try:
+        estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    except Exception as e:
+        logger.error(f"Scraper failed: {e}", exc_info=True)
+        write_stats(STATS_FILE, {
+            "source": "iDNES",
+            "timestamp": _run_ts,
+            "duration_sec": round(time.time() - start, 1),
+            "success": False,
+            "accepted": 0,
+            "fetched": 0,
+            "error": str(e),
+        })
+        raise
 
     if estates:
         json_path = Path("byty_idnes.json")

scrape_psn.py

@@ -15,6 +15,9 @@ import time
 from datetime import datetime
 from pathlib import Path
 from urllib.parse import urlencode
+from scraper_stats import write_stats, validate_listing
+
+STATS_FILE = "stats_psn.json"
 
 logger = logging.getLogger(__name__)
 
@@ -35,19 +38,25 @@ BASE_URL = "https://psn.cz"
 UNITS_API = f"{BASE_URL}/api/units-list"
 
 
-def fetch_json(url: str) -> dict:
+def fetch_json(url: str, retries: int = 3) -> dict:
-    """Fetch JSON via curl (urllib SSL may fail on Cloudflare)."""
+    """Fetch JSON via curl (urllib SSL may fail on Cloudflare) with retry."""
-    logger.debug(f"HTTP GET: {url}")
+    for attempt in range(retries):
-    result = subprocess.run(
+        logger.debug(f"HTTP GET (attempt {attempt + 1}/{retries}): {url}")
-        ["curl", "-s", "-L", "--max-time", "30",
+        result = subprocess.run(
-         "-H", f"User-Agent: {UA}",
+            ["curl", "-s", "-L", "--max-time", "30",
-         "-H", "Accept: application/json",
+             "-H", f"User-Agent: {UA}",
-         url],
+             "-H", "Accept: application/json",
-        capture_output=True, text=True, timeout=60
+             url],
-    )
+            capture_output=True, text=True, timeout=60
-    if result.returncode != 0:
+        )
-        raise RuntimeError(f"curl failed ({result.returncode}): {result.stderr[:200]}")
+        if result.returncode == 0:
-    return json.loads(result.stdout)
+            return json.loads(result.stdout)
+        if attempt < retries - 1:
+            wait = (attempt + 1) * 2
+            logger.warning(f"curl failed (retry {attempt + 1}/{retries} after {wait}s): {result.stderr[:200]}")
+            time.sleep(wait)
+        else:
+            raise RuntimeError(f"curl failed after {retries} attempts ({result.returncode}): {result.stderr[:200]}")
 
 
 def fix_gps(lat, lng):
@@ -67,6 +76,8 @@ def format_price(price: int) -> str:
 
 
 def scrape(max_properties: int | None = None):
+    _run_start = time.time()
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     logger.info("=" * 60)
     logger.info("Stahuji inzeráty z PSN.cz")
     logger.info(f"Cena: do {format_price(MAX_PRICE)}")
@@ -93,11 +104,30 @@ def scrape(max_properties: int | None = None):
         data = fetch_json(url)
     except Exception as e:
         logger.error(f"Chyba při stahování: {e}", exc_info=True)
+        write_stats(STATS_FILE, {
+            "source": "PSN",
+            "timestamp": _run_ts,
+            "duration_sec": round(time.time() - _run_start, 1),
+            "success": False,
+            "accepted": 0,
+            "fetched": 0,
+            "error": str(e),
+        })
         return []
 
     all_units = data.get("units", {}).get("data", [])
     logger.info(f"Staženo jednotek celkem: {len(all_units)}")
 
+    # Load previous output for first_seen/last_changed tracking
+    _prev_cache: dict[str, dict] = {}
+    _prev_path = Path("byty_psn.json")
+    if _prev_path.exists():
+        try:
+            for _item in json.loads(_prev_path.read_text(encoding="utf-8")):
+                _prev_cache[str(_item["hash_id"])] = _item
+        except Exception:
+            pass
+
     # Filtrování
     results = []
     excluded = {
@@ -228,7 +258,11 @@ def scrape(max_properties: int | None = None):
             "source": "psn",
             "image": "",
             "scraped_at": datetime.now().strftime("%Y-%m-%d"),
+            "first_seen": _prev_cache.get(str(unit_id), {}).get("first_seen", datetime.now().strftime("%Y-%m-%d")),
+            "last_changed": datetime.now().strftime("%Y-%m-%d") if _prev_cache.get(str(unit_id), {}).get("price") != int(price) else _prev_cache[str(unit_id)].get("last_changed", datetime.now().strftime("%Y-%m-%d")),
         }
+        if not validate_listing(result, "psn"):
+            continue
         results.append(result)
         properties_fetched += 1
 
@@ -241,6 +275,15 @@ def scrape(max_properties: int | None = None):
     logger.info(f"  ✓ Vyhovující byty:    {len(results)}")
     logger.info(f"{'=' * 60}")
 
+    write_stats(STATS_FILE, {
+        "source": "PSN",
+        "timestamp": _run_ts,
+        "duration_sec": round(time.time() - _run_start, 1),
+        "success": True,
+        "accepted": len(results),
+        "fetched": len(all_units),
+        "excluded": excluded,
+    })
     return results
 
 
@@ -259,8 +302,22 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     start = time.time()
-    estates = scrape(max_properties=args.max_properties)
+    try:
+        estates = scrape(max_properties=args.max_properties)
+    except Exception as e:
+        logger.error(f"Scraper failed: {e}", exc_info=True)
+        write_stats(STATS_FILE, {
+            "source": "PSN",
+            "timestamp": _run_ts,
+            "duration_sec": round(time.time() - start, 1),
+            "success": False,
+            "accepted": 0,
+            "fetched": 0,
+            "error": str(e),
+        })
+        raise
 
     if estates:
         json_path = Path("byty_psn.json")

scrape_realingo.py

@@ -15,6 +15,9 @@ import re
 import time
 import urllib.request
 from pathlib import Path
+from scraper_stats import write_stats, validate_listing
+
+STATS_FILE = "stats_realingo.json"
 
 logger = logging.getLogger(__name__)
 
@@ -53,6 +56,28 @@ HEADERS = {
 BASE_URL = "https://www.realingo.cz"
 
 
+def fetch_url(url: str, retries: int = 3) -> str:
+    """Fetch URL and return HTML string with retry on transient errors."""
+    for attempt in range(retries):
+        try:
+            logger.debug(f"HTTP GET request (attempt {attempt + 1}/{retries}): {url}")
+            req = urllib.request.Request(url, headers=HEADERS)
+            resp = urllib.request.urlopen(req, timeout=30)
+            html = resp.read().decode("utf-8")
+            logger.debug(f"HTTP response: status={resp.status}, size={len(html)} bytes")
+            return html
+        except urllib.error.HTTPError:
+            raise
+        except (ConnectionResetError, ConnectionError, urllib.error.URLError, OSError) as e:
+            if attempt < retries - 1:
+                wait = (attempt + 1) * 2
+                logger.warning(f"Connection error (retry {attempt + 1}/{retries} after {wait}s): {e}")
+                time.sleep(wait)
+            else:
+                logger.error(f"HTTP request failed after {retries} attempts: {e}", exc_info=True)
+                raise
+
+
 def fetch_listing_page(page: int = 1) -> tuple[list[dict], int]:
     """Fetch a page of Prague listings. Returns (items, total_count)."""
     if page == 1:
@@ -60,41 +85,26 @@ def fetch_listing_page(page: int = 1) -> tuple[list[dict], int]:
     else:
         url = f"{BASE_URL}/prodej_byty/praha/{page}_strana/"
 
-    logger.debug(f"HTTP GET request: {url}")
+    html = fetch_url(url)
-    logger.debug(f"Headers: {HEADERS}")
+    match = re.search(
-    req = urllib.request.Request(url, headers=HEADERS)
+        r'<script id="__NEXT_DATA__" type="application/json">(.*?)</script>',
-    try:
+        html, re.DOTALL
-        resp = urllib.request.urlopen(req, timeout=30)
+    )
-        html = resp.read().decode("utf-8")
+    if not match:
-        logger.debug(f"HTTP response: status={resp.status}, size={len(html)} bytes")
+        logger.debug("No __NEXT_DATA__ script found in HTML")
+        return [], 0
 
-        match = re.search(
+    data = json.loads(match.group(1))
-            r'<script id="__NEXT_DATA__" type="application/json">(.*?)</script>',
+    offer_list = data["props"]["pageProps"]["store"]["offer"]["list"]
-            html, re.DOTALL
+    logger.debug(f"Page {page}: found {len(offer_list['data'])} items, total={offer_list['total']}")
-        )
+    return offer_list["data"], offer_list["total"]
-        if not match:
-            logger.debug("No __NEXT_DATA__ script found in HTML")
-            return [], 0
-
-        data = json.loads(match.group(1))
-        offer_list = data["props"]["pageProps"]["store"]["offer"]["list"]
-        logger.debug(f"Page {page}: found {len(offer_list['data'])} items, total={offer_list['total']}")
-        return offer_list["data"], offer_list["total"]
-    except (urllib.error.URLError, ConnectionError, OSError) as e:
-        logger.error(f"HTTP request failed for {url}: {e}", exc_info=True)
-        raise
 
 
 def fetch_detail(listing_url: str) -> dict | None:
     """Fetch detail page for a listing to get floor, building type, etc."""
     try:
         url = f"{BASE_URL}{listing_url}"
-        logger.debug(f"HTTP GET request: {url}")
+        html = fetch_url(url)
-        req = urllib.request.Request(url, headers=HEADERS)
-        resp = urllib.request.urlopen(req, timeout=30)
-        html = resp.read().decode("utf-8")
-        logger.debug(f"HTTP response: status={resp.status}, size={len(html)} bytes")
-
         match = re.search(
             r'<script id="__NEXT_DATA__" type="application/json">(.*?)</script>',
             html, re.DOTALL
@@ -136,6 +146,8 @@ def load_cache(json_path: str = "byty_realingo.json") -> dict[int, dict]:
 
 
 def scrape(max_pages: int | None = None, max_properties: int | None = None):
+    _run_start = time.time()
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     cache = load_cache()
 
     logger.info("=" * 60)
@@ -316,7 +328,11 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
             "source": "realingo",
             "image": "",
             "scraped_at": datetime.now().strftime("%Y-%m-%d"),
+            "first_seen": cached.get("first_seen", datetime.now().strftime("%Y-%m-%d")) if cached else datetime.now().strftime("%Y-%m-%d"),
+            "last_changed": datetime.now().strftime("%Y-%m-%d"),
         }
+        if not validate_listing(result, "realingo"):
+            continue
         results.append(result)
         properties_fetched += 1
 
@@ -333,6 +349,25 @@ def scrape(max_pages: int | None = None, max_properties: int | None = None):
     logger.info(f"  ✓ Vyhovující byty:    {len(results)}")
     logger.info(f"{'=' * 60}")
 
+    write_stats(STATS_FILE, {
+        "source": "Realingo",
+        "timestamp": _run_ts,
+        "duration_sec": round(time.time() - _run_start, 1),
+        "success": True,
+        "accepted": len(results),
+        "fetched": len(all_listings),
+        "pages": page - 1,
+        "cache_hits": cache_hits,
+        "excluded": {
+            "dispozice": excluded_category,
+            "cena": excluded_price,
+            "plocha": excluded_area,
+            "bez GPS": excluded_no_gps,
+            "panel/síd": excluded_panel,
+            "patro": excluded_floor,
+            "bez detailu": excluded_detail,
+        },
+    })
     return results
 
 
@@ -353,8 +388,22 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    _run_ts = datetime.now().isoformat(timespec="seconds")
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    try:
+        estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    except Exception as e:
+        logger.error(f"Scraper failed: {e}", exc_info=True)
+        write_stats(STATS_FILE, {
+            "source": "Realingo",
+            "timestamp": _run_ts,
+            "duration_sec": round(time.time() - start, 1),
+            "success": False,
+            "accepted": 0,
+            "fetched": 0,
+            "error": str(e),
+        })
+        raise
 
     if estates:
         json_path = Path("byty_realingo.json")

scraper_stats.py

@@ -0,0 +1,55 @@
+"""Shared utilities for scraper run statistics and listing validation."""
+from __future__ import annotations
+
+import json
+import logging
+import os
+from pathlib import Path
+
+HERE = Path(__file__).parent
+DATA_DIR = Path(os.environ.get("DATA_DIR", HERE))
+
+_val_log = logging.getLogger(__name__)
+
+_REQUIRED_FIELDS = ("hash_id", "price", "locality", "lat", "lon", "url", "source")
+
+
+def validate_listing(listing: dict, context: str = "") -> bool:
+    """
+    Validate a listing dict before it is written to the output JSON.
+    Returns True if valid, False if the listing should be skipped.
+    Logs a warning for each invalid listing.
+    """
+    prefix = f"[{context}] " if context else ""
+
+    for field in _REQUIRED_FIELDS:
+        val = listing.get(field)
+        if val is None or val == "":
+            _val_log.warning(f"{prefix}Skipping listing — missing field '{field}': {listing.get('hash_id', '?')}")
+            return False
+
+    price = listing.get("price")
+    if not isinstance(price, (int, float)) or price <= 0:
+        _val_log.warning(f"{prefix}Skipping listing — invalid price={price!r}: {listing.get('hash_id', '?')}")
+        return False
+
+    lat, lon = listing.get("lat"), listing.get("lon")
+    if not isinstance(lat, (int, float)) or not isinstance(lon, (int, float)):
+        _val_log.warning(f"{prefix}Skipping listing — non-numeric GPS lat={lat!r} lon={lon!r}: {listing.get('hash_id', '?')}")
+        return False
+    if not (47.0 <= lat <= 52.0) or not (12.0 <= lon <= 19.0):
+        _val_log.warning(f"{prefix}Skipping listing — GPS outside Czech Republic lat={lat} lon={lon}: {listing.get('hash_id', '?')}")
+        return False
+
+    area = listing.get("area")
+    if area is not None and (not isinstance(area, (int, float)) or area <= 0):
+        _val_log.warning(f"{prefix}Skipping listing — invalid area={area!r}: {listing.get('hash_id', '?')}")
+        return False
+
+    return True
+
+
+def write_stats(filename: str, stats: dict) -> None:
+    """Write scraper run stats dict to the data directory."""
+    path = DATA_DIR / filename
+    path.write_text(json.dumps(stats, ensure_ascii=False, indent=2), encoding="utf-8")

server.py

@@ -0,0 +1,477 @@
+#!/usr/bin/env python3
+"""
+General-purpose HTTP server for maru-hleda-byt.
+
+Serves static files from DATA_DIR and additionally handles:
+  GET  /scrapers-status        → SSR scraper status page
+  GET  /api/ratings            → ratings.json contents
+  POST /api/ratings            → save entire ratings object
+  GET  /api/ratings/export     → same as GET, with download header
+  GET  /api/status             → status.json contents (JSON)
+  GET  /api/status/history     → scraper_history.json contents (JSON)
+"""
+
+from __future__ import annotations
+
+import functools
+import json
+import logging
+import os
+import sys
+from datetime import datetime
+from http.server import HTTPServer, SimpleHTTPRequestHandler
+from pathlib import Path
+
+PORT = int(os.environ.get("SERVER_PORT", 8080))
+DATA_DIR = Path(os.environ.get("DATA_DIR", "."))
+RATINGS_FILE = DATA_DIR / "ratings.json"
+_LOG_LEVEL = getattr(logging, os.environ.get("LOG_LEVEL", "INFO").upper(), logging.INFO)
+
+logging.basicConfig(
+    level=_LOG_LEVEL,
+    format="%(asctime)s [server] %(levelname)s %(message)s",
+    datefmt="%Y-%m-%dT%H:%M:%S",
+)
+log = logging.getLogger(__name__)
+
+# ── Helpers ──────────────────────────────────────────────────────────────────
+
+COLORS = {
+    "sreality":    "#1976D2",
+    "realingo":    "#7B1FA2",
+    "bezrealitky": "#E65100",
+    "idnes":       "#C62828",
+    "psn":         "#2E7D32",
+    "cityhome":    "#00838F",
+}
+
+MONTHS_CZ = [
+    "ledna", "února", "března", "dubna", "května", "června",
+    "července", "srpna", "září", "října", "listopadu", "prosince",
+]
+
+
+def _load_json(path: Path, default=None):
+    """Read and parse JSON file; return default on missing or parse error."""
+    log.debug("_load_json: %s", path.resolve())
+    try:
+        if path.exists():
+            return json.loads(path.read_text(encoding="utf-8"))
+    except Exception as e:
+        log.warning("Failed to load %s: %s", path, e)
+    return default
+
+
+def _fmt_date(iso_str: str) -> str:
+    """Format ISO timestamp as Czech date string."""
+    try:
+        d = datetime.fromisoformat(iso_str)
+        return f"{d.day}. {MONTHS_CZ[d.month - 1]} {d.year}, {d.hour:02d}:{d.minute:02d}"
+    except Exception:
+        return iso_str
+
+
+def load_ratings() -> dict:
+    return _load_json(RATINGS_FILE, default={})
+
+
+def save_ratings(data: dict) -> None:
+    RATINGS_FILE.write_text(
+        json.dumps(data, ensure_ascii=False, indent=2),
+        encoding="utf-8",
+    )
+
+
+# ── SSR status page ──────────────────────────────────────────────────────────
+
+_CSS = """\
+* { margin: 0; padding: 0; box-sizing: border-box; }
+body {
+  font-family: system-ui, -apple-system, sans-serif;
+  background: #f5f5f5; color: #333;
+  padding: 24px; max-width: 640px; margin: 0 auto;
+}
+h1 { font-size: 22px; margin-bottom: 4px; }
+.subtitle { color: #888; font-size: 13px; margin-bottom: 24px; }
+.card {
+  background: white; border-radius: 12px; padding: 20px;
+  box-shadow: 0 1px 4px rgba(0,0,0,0.08); margin-bottom: 16px;
+}
+.card h2 { font-size: 15px; margin-bottom: 12px; color: #555; }
+.timestamp { font-size: 28px; font-weight: 700; color: #1976D2; }
+.timestamp-sub { font-size: 13px; color: #999; margin-top: 2px; }
+.summary-row {
+  display: flex; justify-content: space-between; align-items: center;
+  padding: 10px 0; border-bottom: 1px solid #f0f0f0;
+}
+.summary-row:last-child { border-bottom: none; }
+.summary-label { font-size: 13px; color: #666; }
+.summary-value { font-size: 18px; font-weight: 700; }
+.badge {
+  display: inline-block; padding: 2px 8px; border-radius: 4px;
+  font-size: 11px; font-weight: 600; color: white;
+}
+.badge-ok   { background: #4CAF50; }
+.badge-err  { background: #F44336; }
+.badge-skip { background: #FF9800; }
+.bar-row { display: flex; align-items: center; gap: 8px; margin: 4px 0; }
+.bar-track { flex: 1; height: 20px; background: #f0f0f0; border-radius: 4px; overflow: hidden; }
+.bar-fill  { height: 100%; border-radius: 4px; }
+.bar-count { font-size: 12px; width: 36px; font-variant-numeric: tabular-nums; }
+.loader-wrap {
+  display: flex; flex-direction: column; align-items: center;
+  justify-content: center; padding: 60px 0;
+}
+.spinner {
+  width: 40px; height: 40px; border: 4px solid #e0e0e0;
+  border-top-color: #1976D2; border-radius: 50%;
+  animation: spin 0.8s linear infinite;
+}
+@keyframes spin { to { transform: rotate(360deg); } }
+.loader-text { margin-top: 16px; color: #999; font-size: 14px; }
+.link-row { text-align: center; margin-top: 8px; }
+.link-row a { color: #1976D2; text-decoration: none; font-size: 14px; }
+.history-table { width: 100%; border-collapse: collapse; font-size: 12px; }
+.history-table th {
+  text-align: left; font-weight: 600; color: #999; font-size: 11px;
+  padding: 4px 6px 8px 6px; border-bottom: 2px solid #f0f0f0;
+}
+.history-table td { padding: 7px 6px; border-bottom: 1px solid #f5f5f5; vertical-align: middle; }
+.history-table tr:last-child td { border-bottom: none; }
+.history-table tr.latest td { background: #f8fbff; font-weight: 600; }
+.src-nums { display: flex; gap: 4px; flex-wrap: wrap; }
+.src-chip {
+  display: inline-block; padding: 1px 5px; border-radius: 3px;
+  font-size: 10px; color: white; font-variant-numeric: tabular-nums;
+}
+.clickable-row { cursor: pointer; }
+.clickable-row:hover td { background: #f0f7ff !important; }
+/* Modal */
+#md-overlay {
+  position: fixed; inset: 0; background: rgba(0,0,0,0.45);
+  display: flex; align-items: flex-start; justify-content: center;
+  z-index: 1000; padding: 40px 16px; overflow-y: auto;
+}
+#md-box {
+  background: white; border-radius: 12px; padding: 24px;
+  width: 100%; max-width: 620px; position: relative;
+  box-shadow: 0 8px 32px rgba(0,0,0,0.24); margin: auto;
+}
+#md-close {
+  position: absolute; top: 10px; right: 14px;
+  background: none; border: none; font-size: 26px; cursor: pointer;
+  color: #aaa; line-height: 1;
+}
+#md-close:hover { color: #333; }
+#md-box h3 { font-size: 15px; margin-bottom: 14px; padding-right: 24px; }
+.md-summary { display: flex; gap: 20px; flex-wrap: wrap; font-size: 13px; margin-bottom: 16px; color: #555; }
+.md-summary b { color: #333; }
+.detail-table { width: 100%; border-collapse: collapse; font-size: 12px; }
+.detail-table th {
+  text-align: left; color: #999; font-size: 11px; font-weight: 600;
+  padding: 4px 8px 6px 0; border-bottom: 2px solid #f0f0f0; white-space: nowrap;
+}
+.detail-table td { padding: 6px 8px 6px 0; border-bottom: 1px solid #f5f5f5; vertical-align: top; }
+.detail-table tr:last-child td { border-bottom: none; }
+"""
+
+_SOURCE_ORDER = ["Sreality", "Realingo", "Bezrealitky", "iDNES", "PSN", "CityHome"]
+_SOURCE_ABBR  = ["Sre", "Rea", "Bez", "iDN", "PSN", "CH"]
+
+
+def _sources_html(sources: list) -> str:
+    if not sources:
+        return ""
+    max_count = max((s.get("accepted", 0) for s in sources), default=1) or 1
+    parts = ['<div class="card"><h2>Zdroje</h2>']
+    for s in sources:
+        name = s.get("name", "?")
+        accepted = s.get("accepted", 0)
+        error = s.get("error")
+        exc = s.get("excluded", {})
+        excluded_total = sum(exc.values()) if isinstance(exc, dict) else s.get("excluded_total", 0)
+        color = COLORS.get(name.lower(), "#999")
+        pct = round(accepted / max_count * 100) if max_count else 0
+        if error:
+            badge = '<span class="badge badge-err">chyba</span>'
+        elif accepted == 0:
+            badge = '<span class="badge badge-skip">0</span>'
+        else:
+            badge = '<span class="badge badge-ok">OK</span>'
+        parts.append(
+            f'<div style="margin-bottom:12px;">'
+            f'<div style="display:flex;justify-content:space-between;align-items:center;margin-bottom:4px;">'
+            f'<span style="font-weight:600;font-size:14px;">{name} {badge}</span>'
+            f'<span style="font-size:12px;color:#999;">{excluded_total} vyloučených</span>'
+            f'</div>'
+            f'<div class="bar-row">'
+            f'<div class="bar-track"><div class="bar-fill" style="width:{pct}%;background:{color};"></div></div>'
+            f'<span class="bar-count">{accepted}</span>'
+            f'</div></div>'
+        )
+    parts.append("</div>")
+    return "".join(parts)
+
+
+def _history_html(history: list) -> str:
+    if not history:
+        return ""
+    rows = list(reversed(history))
+    parts = [
+        '<div class="card">'
+        '<h2>Historie běhů <span style="font-size:11px;font-weight:400;color:#bbb;">– klikni pro detaily</span></h2>',
+        '<table class="history-table"><thead><tr>',
+        '<th>Datum</th><th>Trvání</th><th>Přijato&nbsp;/&nbsp;Dedup</th><th>Zdroje</th><th>OK</th>',
+        '</tr></thead><tbody>',
+    ]
+    for i, entry in enumerate(rows):
+        row_class = ' class="latest clickable-row"' if i == 0 else ' class="clickable-row"'
+        src_map = {s["name"]: s for s in entry.get("sources", []) if "name" in s}
+        chips = "".join(
+            f'<span class="src-chip" style="background:{"#F44336" if (src_map.get(name) or {}).get("error") else COLORS.get(name.lower(), "#999")}" title="{name}">'
+            f'{abbr}&nbsp;{src_map[name].get("accepted", 0) if name in src_map else "-"}</span>'
+            for name, abbr in zip(_SOURCE_ORDER, _SOURCE_ABBR)
+        )
+        ok_badge = (
+            '<span class="badge badge-err">chyba</span>'
+            if entry.get("success") is False
+            else '<span class="badge badge-ok">OK</span>'
+        )
+        dur = f'{entry["duration_sec"]}s' if entry.get("duration_sec") is not None else "-"
+        parts.append(
+            f'<tr{row_class} data-idx="{i}">'
+            f'<td>{_fmt_date(entry.get("timestamp", ""))}</td>'
+            f'<td>{dur}</td>'
+            f'<td>{entry.get("total_accepted", "-")}&nbsp;/&nbsp;{entry.get("deduplicated", "-")}</td>'
+            f'<td><div class="src-nums">{chips}</div></td>'
+            f'<td>{ok_badge}</td>'
+            f'</tr>'
+        )
+    parts.append("</tbody></table></div>")
+    return "".join(parts)
+
+
+def _modal_script(rows_json: str) -> str:
+    """Return the modal overlay HTML + JS for the history detail popup."""
+    return (
+        '<div id="md-overlay" style="display:none">'
+        '<div id="md-box"><button id="md-close">&times;</button>'
+        '<div id="md-body"></div></div></div>\n'
+        '<script>\n(function(){\n'
+        f'var H={rows_json};\n'
+        'var C={"sreality":"#1976D2","realingo":"#7B1FA2","bezrealitky":"#E65100","idnes":"#C62828","psn":"#2E7D32","cityhome":"#00838F"};\n'
+        'var MN=["ledna","února","března","dubna","května","června","července","srpna","září","října","listopadu","prosince"];\n'
+        'function fd(s){var d=new Date(s);return d.getDate()+". "+MN[d.getMonth()]+" "+d.getFullYear()+", "+String(d.getHours()).padStart(2,"0")+":"+String(d.getMinutes()).padStart(2,"0");}\n'
+        'function openModal(idx){\n'
+        '  var e=H[idx],src=e.sources||[];\n'
+        '  var h="<h3>Detaily b\u011bhu \u2013 "+fd(e.timestamp)+"</h3>";\n'
+        '  h+="<div class=\\"md-summary\\">";\n'
+        '  if(e.duration_sec!=null) h+="<span><b>Trvání:</b> "+e.duration_sec+"s</span>";\n'
+        '  if(e.total_accepted!=null) h+="<span><b>Přijato:</b> "+e.total_accepted+"</span>";\n'
+        '  if(e.deduplicated!=null) h+="<span><b>Po dedup:</b> "+e.deduplicated+"</span>";\n'
+        '  h+="</div>";\n'
+        '  h+="<table class=\\"detail-table\\"><thead><tr>";\n'
+        '  h+="<th>Zdroj</th><th>Přijato</th><th>Staženo</th><th>Stránky</th><th>Cache</th><th>Vyloučeno</th><th>Čas</th><th>OK</th>";\n'
+        '  h+="</tr></thead><tbody>";\n'
+        '  src.forEach(function(s){\n'
+        '    var nm=s.name||"?",col=C[nm.toLowerCase()]||"#999";\n'
+        '    var exc=s.excluded||{};\n'
+        '    var excStr=Object.entries(exc).filter(function(kv){return kv[1]>0;}).map(function(kv){return kv[0]+":&nbsp;"+kv[1];}).join(", ")||"\u2013";\n'
+        '    var ok=s.error?"<span class=\\"badge badge-err\\" title=\\""+s.error+"\\">chyba</span>":"<span class=\\"badge badge-ok\\">OK</span>";\n'
+        '    var dot="<span style=\\"display:inline-block;width:8px;height:8px;border-radius:50%;background:"+col+";margin-right:5px;\\"></span>";\n'
+        '    h+="<tr>";\n'
+        '    h+="<td>"+dot+nm+"</td>";\n'
+        '    h+="<td>"+(s.accepted!=null?s.accepted:"\u2013")+"</td>";\n'
+        '    h+="<td>"+(s.fetched!=null?s.fetched:"\u2013")+"</td>";\n'
+        '    h+="<td>"+(s.pages!=null?s.pages:"\u2013")+"</td>";\n'
+        '    h+="<td>"+(s.cache_hits!=null?s.cache_hits:"\u2013")+"</td>";\n'
+        '    h+="<td style=\\"font-size:11px;color:#666;\\">"+excStr+"</td>";\n'
+        '    h+="<td>"+(s.duration_sec!=null?s.duration_sec+"s":"\u2013")+"</td>";\n'
+        '    h+="<td>"+ok+"</td></tr>";\n'
+        '  });\n'
+        '  h+="</tbody></table>";\n'
+        '  document.getElementById("md-body").innerHTML=h;\n'
+        '  document.getElementById("md-overlay").style.display="flex";\n'
+        '}\n'
+        'function closeModal(){document.getElementById("md-overlay").style.display="none";}\n'
+        'var tb=document.querySelector(".history-table tbody");\n'
+        'if(tb)tb.addEventListener("click",function(e){var tr=e.target.closest("tr[data-idx]");if(tr)openModal(parseInt(tr.dataset.idx,10));});\n'
+        'document.getElementById("md-close").addEventListener("click",closeModal);\n'
+        'document.getElementById("md-overlay").addEventListener("click",function(e){if(e.target===this)closeModal();});\n'
+        'document.addEventListener("keydown",function(e){if(e.key==="Escape")closeModal();});\n'
+        '})();\n</script>'
+    )
+
+
+def _render_status_html(status: dict | None, history: list, is_running: bool = False) -> str:
+    """Generate the complete HTML page for /scrapers-status."""
+    head_open = (
+        '<!DOCTYPE html>\n<html lang="cs">\n<head>\n'
+        '<meta charset="UTF-8">\n'
+        '<meta name="viewport" content="width=device-width, initial-scale=1.0">\n'
+        f'<title>Scraper status</title>\n<style>{_CSS}</style>\n'
+    )
+    page_header = '<h1>Scraper status</h1>\n<div class="subtitle">maru-hleda-byt</div>\n'
+    footer = '<div class="link-row"><a href="/mapa_bytu.html">Otevřít mapu</a></div>'
+
+    if status is None:
+        return (
+            head_open + '</head>\n<body>\n' + page_header
+            + '<div class="card"><p style="color:#F44336">Status není k dispozici.</p></div>\n'
+            + footer + '\n</body>\n</html>'
+        )
+
+    if is_running:
+        return (
+            head_open
+            + '<meta http-equiv="refresh" content="30">\n'
+            + '</head>\n<body>\n' + page_header
+            + '<div class="loader-wrap"><div class="spinner"></div>'
+            + '<div class="loader-text">Scraper právě běží…</div></div>\n'
+            + footer + '\n</body>\n</html>'
+        )
+
+    # ── Done state ────────────────────────────────────────────────────────────
+    ts = status.get("timestamp", "")
+    duration = status.get("duration_sec")
+    total_accepted = status.get("total_accepted", 0)
+    deduplicated = status.get("deduplicated")
+
+    ts_card = (
+        '<div class="card"><h2>Poslední scrape</h2>'
+        f'<div class="timestamp">{_fmt_date(ts)}</div>'
+        + (f'<div class="timestamp-sub">Trvání: {round(duration)}s</div>' if duration is not None else "")
+        + '</div>'
+    )
+
+    sum_card = (
+        '<div class="card"><h2>Souhrn</h2>'
+        f'<div class="summary-row"><span class="summary-label">Vyhovujících bytů</span>'
+        f'<span class="summary-value" style="color:#4CAF50">{total_accepted}</span></div>'
+        + (
+            f'<div class="summary-row"><span class="summary-label">Po deduplikaci (v mapě)</span>'
+            f'<span class="summary-value" style="color:#1976D2">{deduplicated}</span></div>'
+            if deduplicated is not None else ""
+        )
+        + '</div>'
+    )
+
+    rows_for_js = list(reversed(history))
+    body = (
+        page_header
+        + ts_card + "\n"
+        + sum_card + "\n"
+        + _sources_html(status.get("sources", [])) + "\n"
+        + _history_html(history) + "\n"
+        + footer
+    )
+    modal = _modal_script(json.dumps(rows_for_js, ensure_ascii=False))
+    return head_open + '</head>\n<body>\n' + body + '\n' + modal + '\n</body>\n</html>'
+
+
+# ── HTTP handler ──────────────────────────────────────────────────────────────
+
+class Handler(SimpleHTTPRequestHandler):
+    def log_message(self, format, *args):
+        pass  # suppress default access log; use our own where needed
+
+    def _send_json(self, status: int, body, extra_headers=None):
+        payload = json.dumps(body, ensure_ascii=False).encode("utf-8")
+        self.send_response(status)
+        self.send_header("Content-Type", "application/json; charset=utf-8")
+        self.send_header("Content-Length", str(len(payload)))
+        self.send_header("Access-Control-Allow-Origin", "*")
+        self.send_header("Access-Control-Allow-Methods", "GET, POST, OPTIONS")
+        self.send_header("Access-Control-Allow-Headers", "Content-Type")
+        if extra_headers:
+            for k, v in extra_headers.items():
+                self.send_header(k, v)
+        self.end_headers()
+        self.wfile.write(payload)
+
+    def do_OPTIONS(self):
+        self.send_response(204)
+        self.send_header("Access-Control-Allow-Origin", "*")
+        self.send_header("Access-Control-Allow-Methods", "GET, POST, OPTIONS")
+        self.send_header("Access-Control-Allow-Headers", "Content-Type")
+        self.end_headers()
+
+    def do_GET(self):
+        if self.path.startswith("/api/"):
+            self._handle_api_get()
+        elif self.path.rstrip("/") == "/scrapers-status":
+            self._serve_status_page()
+        else:
+            log.debug("GET %s → static file: %s", self.path, self.translate_path(self.path))
+            super().do_GET()
+
+    def _handle_api_get(self):
+        if self.path in ("/api/ratings", "/api/ratings/export"):
+            ratings = load_ratings()
+            extra = None
+            if self.path == "/api/ratings/export":
+                extra = {"Content-Disposition": 'attachment; filename="ratings.json"'}
+            log.info("GET %s → %d ratings", self.path, len(ratings))
+            self._send_json(200, ratings, extra)
+        elif self.path == "/api/status":
+            data = _load_json(DATA_DIR / "status.json")
+            if data is None:
+                self._send_json(404, {"error": "status not available"})
+                return
+            log.info("GET /api/status → ok")
+            self._send_json(200, data)
+        elif self.path == "/api/status/history":
+            data = _load_json(DATA_DIR / "scraper_history.json", default=[])
+            if not isinstance(data, list):
+                data = []
+            log.info("GET /api/status/history → %d entries", len(data))
+            self._send_json(200, data)
+        else:
+            self._send_json(404, {"error": "not found"})
+
+    def _serve_status_page(self):
+        status = _load_json(DATA_DIR / "status.json")
+        history = _load_json(DATA_DIR / "scraper_history.json", default=[])
+        if not isinstance(history, list):
+            history = []
+        is_running = (DATA_DIR / "scraper_running.json").exists()
+        html = _render_status_html(status, history, is_running)
+        payload = html.encode("utf-8")
+        self.send_response(200)
+        self.send_header("Content-Type", "text/html; charset=utf-8")
+        self.send_header("Content-Length", str(len(payload)))
+        self.end_headers()
+        self.wfile.write(payload)
+
+    def do_POST(self):
+        if self.path == "/api/ratings":
+            length = int(self.headers.get("Content-Length", 0))
+            if length == 0:
+                self._send_json(400, {"error": "empty body"})
+                return
+            try:
+                raw = self.rfile.read(length)
+                data = json.loads(raw.decode("utf-8"))
+            except Exception as e:
+                log.warning("Bad request body: %s", e)
+                self._send_json(400, {"error": "invalid JSON"})
+                return
+            if not isinstance(data, dict):
+                self._send_json(400, {"error": "expected JSON object"})
+                return
+            save_ratings(data)
+            log.info("POST /api/ratings → saved %d ratings", len(data))
+            self._send_json(200, {"ok": True, "count": len(data)})
+        else:
+            self._send_json(404, {"error": "not found"})
+
+
+if __name__ == "__main__":
+    log.info("Server starting on port %d, data dir: %s", PORT, DATA_DIR)
+    handler = functools.partial(Handler, directory=str(DATA_DIR))
+    server = HTTPServer(("0.0.0.0", PORT), handler)
+    try:
+        server.serve_forever()
+    except KeyboardInterrupt:
+        log.info("Stopped.")
+        sys.exit(0)

status.html

@@ -1,204 +0,0 @@
-<!DOCTYPE html>
-<html lang="cs">
-<head>
-<meta charset="UTF-8">
-<meta name="viewport" content="width=device-width, initial-scale=1.0">
-<title>Scraper status</title>
-<style>
-  * { margin: 0; padding: 0; box-sizing: border-box; }
-  body {
-    font-family: system-ui, -apple-system, sans-serif;
-    background: #f5f5f5; color: #333;
-    padding: 24px; max-width: 640px; margin: 0 auto;
-  }
-  h1 { font-size: 22px; margin-bottom: 4px; }
-  .subtitle { color: #888; font-size: 13px; margin-bottom: 24px; }
-  .card {
-    background: white; border-radius: 12px; padding: 20px;
-    box-shadow: 0 1px 4px rgba(0,0,0,0.08); margin-bottom: 16px;
-  }
-  .card h2 { font-size: 15px; margin-bottom: 12px; color: #555; }
-  .timestamp {
-    font-size: 28px; font-weight: 700; color: #1976D2;
-  }
-  .timestamp-ago { font-size: 13px; color: #999; margin-top: 2px; }
-
-  /* Source table */
-  .source-table { width: 100%; border-collapse: collapse; }
-  .source-table td { padding: 8px 0; border-bottom: 1px solid #f0f0f0; font-size: 14px; }
-  .source-table tr:last-child td { border-bottom: none; }
-  .source-table .name { font-weight: 600; }
-  .source-table .count { text-align: right; font-variant-numeric: tabular-nums; }
-  .source-table .rejected { text-align: right; color: #999; font-size: 12px; }
-  .badge {
-    display: inline-block; padding: 2px 8px; border-radius: 4px;
-    font-size: 11px; font-weight: 600; color: white;
-  }
-  .badge-ok { background: #4CAF50; }
-  .badge-err { background: #F44336; }
-  .badge-skip { background: #FF9800; }
-
-  /* Summary bar */
-  .summary-row {
-    display: flex; justify-content: space-between; align-items: center;
-    padding: 10px 0; border-bottom: 1px solid #f0f0f0;
-  }
-  .summary-row:last-child { border-bottom: none; }
-  .summary-label { font-size: 13px; color: #666; }
-  .summary-value { font-size: 18px; font-weight: 700; }
-
-  /* Source bar chart */
-  .bar-row { display: flex; align-items: center; gap: 8px; margin: 4px 0; }
-  .bar-label { width: 90px; font-size: 12px; text-align: right; color: #666; }
-  .bar-track { flex: 1; height: 20px; background: #f0f0f0; border-radius: 4px; overflow: hidden; position: relative; }
-  .bar-fill { height: 100%; border-radius: 4px; transition: width 0.5s ease; }
-  .bar-count { font-size: 12px; width: 36px; font-variant-numeric: tabular-nums; }
-
-  /* Loader */
-  .loader-wrap {
-    display: flex; flex-direction: column; align-items: center;
-    justify-content: center; padding: 60px 0;
-  }
-  .spinner {
-    width: 40px; height: 40px; border: 4px solid #e0e0e0;
-    border-top-color: #1976D2; border-radius: 50%;
-    animation: spin 0.8s linear infinite;
-  }
-  @keyframes spin { to { transform: rotate(360deg); } }
-  .loader-text { margin-top: 16px; color: #999; font-size: 14px; }
-
-  .error-msg { color: #F44336; padding: 40px 0; text-align: center; }
-  .link-row { text-align: center; margin-top: 8px; }
-  .link-row a { color: #1976D2; text-decoration: none; font-size: 14px; }
-</style>
-</head>
-<body>
-
-<h1>Scraper status</h1>
-<div class="subtitle">maru-hleda-byt</div>
-
-<div id="content">
-  <div class="loader-wrap">
-    <div class="spinner"></div>
-    <div class="loader-text">Nacitam status...</div>
-  </div>
-</div>
-
-<div class="link-row"><a href="mapa_bytu.html">Otevrit mapu</a></div>
-
-<script>
-var COLORS = {
-  sreality: '#1976D2',
-  realingo: '#7B1FA2',
-  bezrealitky: '#E65100',
-  idnes: '#C62828',
-  psn: '#2E7D32',
-  cityhome: '#00838F',
-};
-
-function timeAgo(dateStr) {
-  var d = new Date(dateStr);
-  var now = new Date();
-  var diff = Math.floor((now - d) / 1000);
-  if (diff < 60) return 'prave ted';
-  if (diff < 3600) return Math.floor(diff / 60) + ' min zpet';
-  if (diff < 86400) return Math.floor(diff / 3600) + ' hod zpet';
-  return Math.floor(diff / 86400) + ' dni zpet';
-}
-
-function formatDate(dateStr) {
-  var d = new Date(dateStr);
-  var day = d.getDate();
-  var months = ['ledna','unora','brezna','dubna','kvetna','cervna',
-    'cervence','srpna','zari','rijna','listopadu','prosince'];
-  var hh = String(d.getHours()).padStart(2, '0');
-  var mm = String(d.getMinutes()).padStart(2, '0');
-  return day + '. ' + months[d.getMonth()] + ' ' + d.getFullYear() + ', ' + hh + ':' + mm;
-}
-
-function render(data) {
-  // Check if scrape is currently running
-  if (data.status === 'running') {
-    document.getElementById('content').innerHTML =
-      '<div class="loader-wrap">' +
-      '<div class="spinner"></div>' +
-      '<div class="loader-text">Scraper prave bezi...</div>' +
-      '</div>';
-    setTimeout(loadStatus, 30000);
-    return;
-  }
-
-  var sources = data.sources || [];
-  var totalOk = 0, totalRej = 0;
-  var maxCount = 0;
-  sources.forEach(function(s) {
-    totalOk += s.accepted || 0;
-    totalRej += s.rejected || 0;
-    if (s.accepted > maxCount) maxCount = s.accepted;
-  });
-
-  var html = '';
-
-  // Timestamp card
-  html += '<div class="card">';
-  html += '<h2>Posledni scrape</h2>';
-  html += '<div class="timestamp">' + formatDate(data.timestamp) + '</div>';
-  html += '<div class="timestamp-ago">' + timeAgo(data.timestamp) + '</div>';
-  if (data.duration_sec) {
-    html += '<div class="timestamp-ago">Trvani: ' + Math.round(data.duration_sec) + 's</div>';
-  }
-  html += '</div>';
-
-  // Summary card
-  html += '<div class="card">';
-  html += '<h2>Souhrn</h2>';
-  html += '<div class="summary-row"><span class="summary-label">Vyhovujicich bytu</span><span class="summary-value" style="color:#4CAF50">' + totalOk + '</span></div>';
-  html += '<div class="summary-row"><span class="summary-label">Vyloucenych</span><span class="summary-value" style="color:#999">' + totalRej + '</span></div>';
-  if (data.deduplicated !== undefined) {
-    html += '<div class="summary-row"><span class="summary-label">Po deduplikaci (v mape)</span><span class="summary-value" style="color:#1976D2">' + data.deduplicated + '</span></div>';
-  }
-  html += '</div>';
-
-  // Sources card
-  html += '<div class="card">';
-  html += '<h2>Zdroje</h2>';
-  sources.forEach(function(s) {
-    var color = COLORS[s.name.toLowerCase()] || '#999';
-    var pct = maxCount > 0 ? Math.round((s.accepted / maxCount) * 100) : 0;
-    var badge = s.error
-      ? '<span class="badge badge-err">chyba</span>'
-      : (s.accepted === 0 ? '<span class="badge badge-skip">0</span>' : '<span class="badge badge-ok">OK</span>');
-
-    html += '<div style="margin-bottom:12px;">';
-    html += '<div style="display:flex;justify-content:space-between;align-items:center;margin-bottom:4px;">';
-    html += '<span style="font-weight:600;font-size:14px;">' + s.name + ' ' + badge + '</span>';
-    html += '<span style="font-size:12px;color:#999;">' + (s.rejected || 0) + ' vyloucenych</span>';
-    html += '</div>';
-    html += '<div class="bar-row">';
-    html += '<div class="bar-track"><div class="bar-fill" style="width:' + pct + '%;background:' + color + ';"></div></div>';
-    html += '<span class="bar-count">' + (s.accepted || 0) + '</span>';
-    html += '</div>';
-    html += '</div>';
-  });
-  html += '</div>';
-
-  document.getElementById('content').innerHTML = html;
-}
-
-function loadStatus() {
-  fetch('status.json?t=' + Date.now())
-    .then(function(r) {
-      if (!r.ok) throw new Error(r.status);
-      return r.json();
-    })
-    .then(render)
-    .catch(function(err) {
-      document.getElementById('content').innerHTML =
-        '<div class="error-msg">Status zatim neni k dispozici.<br><small>(' + err.message + ')</small></div>';
-    });
-}
-
-loadStatus();
-</script>
-</body>
-</html>