Add --data-dir CLI argument to replace symlink-based data persistence

The Docker entrypoint previously created symlinks from /app/ to /app/data/
so that scripts writing relative paths would persist to the mounted volume.
This caused symlink loops in production when stale symlinks leaked into the
host data directory.

Instead, all scrapers, merge_and_map.py, regen_map.py, and run_all.sh now
accept a --data-dir argument (default: ".") that controls where data files
are read from and written to. The entrypoint and crontab pass
--data-dir /app/data, eliminating the need for symlinks entirely.

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

build/crontab

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

build/entrypoint.sh

@@ -3,20 +3,11 @@ set -euo pipefail
 
 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 \
-         byty_idnes.json byty_psn.json byty_cityhome.json byty_merged.json \
-         mapa_bytu.html; do
-    # Remove real file if it exists (e.g. baked into image)
-    [ -f "/app/$f" ] && [ ! -L "/app/$f" ] && rm -f "/app/$f"
-    ln -sf "$DATA_DIR/$f" "/app/$f"
-done
-
 echo "[entrypoint] Starting crond..."
 crond -b -l 2
 
 echo "[entrypoint] Starting initial scrape in background..."
-bash /app/run_all.sh &
+bash /app/run_all.sh --data-dir "$DATA_DIR" &
 
 echo "[entrypoint] Starting HTTP server on port 8080..."
 exec python3 -m http.server 8080 --directory "$DATA_DIR"

build/run.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+docker rm -f maru-hleda-byt
+
+# gitea registry login with kacerr / token
+docker run -d --name maru-hleda-byt \
+  -p 8080:8080 \
+  -v /srv/maru-hleda-byt/data:/app/data \
+  gitea.home.hrajfrisbee.cz/littlemeat/maru-hleda-byt:0.01

merge_and_map.py

@@ -7,6 +7,7 @@ PSN a CityHome mají při deduplikaci prioritu (načtou se první).
 """
 from __future__ import annotations
 
+import argparse
 import json
 import re
 from pathlib import Path
@@ -40,7 +41,7 @@ def dedup_key(estate: dict) -> str:
     return f"{street}_{price}_{area}"
 
 
-def main():
+def main(data_dir: str = "."):
     # Definice zdrojů — PSN a CityHome jako první (mají prioritu při deduplikaci)
     sources = [
         ("PSN",         "byty_psn.json"),
@@ -51,10 +52,11 @@ def main():
         ("iDNES",       "byty_idnes.json"),
     ]
 
+    data_path = Path(data_dir)
     all_estates = []
 
     for label, filename in sources:
-        path = Path(filename)
+        path = data_path / filename
         if path.exists():
             data = json.loads(path.read_text(encoding="utf-8"))
             # Ensure source is set (Sreality legacy)
@@ -111,7 +113,7 @@ def main():
         print(f"  {src}: {count}")
 
     # Save merged data
-    merged_path = Path("byty_merged.json")
+    merged_path = data_path / "byty_merged.json"
     merged_path.write_text(
         json.dumps(deduplicated, ensure_ascii=False, indent=2),
         encoding="utf-8",
@@ -119,8 +121,12 @@ def main():
     print(f"\n✓ Sloučená data: {merged_path.resolve()}")
 
     # Generate map
-    generate_map(deduplicated)
+    generate_map(deduplicated, output_path=str(data_path / "mapa_bytu.html"))
 
 
 if __name__ == "__main__":
-    main()
+    parser = argparse.ArgumentParser(description="Merge scraped data and generate map")
+    parser.add_argument("--data-dir", type=str, default=".",
+                        help="Directory for reading/writing data files (default: current dir)")
+    args = parser.parse_args()
+    main(data_dir=args.data_dir)

regen_map.py

@@ -5,6 +5,7 @@ Doplní chybějící plochy ze Sreality API, opraví URL, aplikuje filtry.
 """
 from __future__ import annotations
 
+import argparse
 import json
 import time
 import urllib.request
@@ -57,8 +58,9 @@ def fetch_area(hash_id: int) -> int | None:
     return None
 
 
-def main():
-    json_path = Path("byty_sreality.json")
+def main(data_dir: str = "."):
+    data_path = Path(data_dir)
+    json_path = data_path / "byty_sreality.json"
     if not json_path.exists():
         print("Soubor byty_sreality.json nenalezen. Nejprve spusť scrape_and_map.py")
         return
@@ -100,15 +102,19 @@ def main():
     print(f"Zbývá: {len(filtered)} bytů")
 
     # Save updated data
-    filtered_path = Path("byty_sreality.json")
+    filtered_path = data_path / "byty_sreality.json"
     filtered_path.write_text(
         json.dumps(filtered, ensure_ascii=False, indent=2),
         encoding="utf-8",
     )
 
     # Generate map
-    generate_map(filtered)
+    generate_map(filtered, output_path=str(data_path / "mapa_bytu.html"))
 
 
 if __name__ == "__main__":
-    main()
+    parser = argparse.ArgumentParser(description="Regenerate map from existing data")
+    parser.add_argument("--data-dir", type=str, default=".",
+                        help="Directory for reading/writing data files (default: current dir)")
+    args = parser.parse_args()
+    main(data_dir=args.data_dir)

run_all.sh

@@ -4,6 +4,7 @@
 # Použití:  ./run_all.sh
 # Nebo s limity:  ./run_all.sh --max-pages 1 --max-properties 10
 # Nebo s logováním:  ./run_all.sh --log-level DEBUG
+# S vlastním adresářem: ./run_all.sh --data-dir /app/data
 # ============================================================
 set -euo pipefail
 cd "$(dirname "$0")"
@@ -26,16 +27,19 @@ 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 "  --data-dir DIR        Adresář pro čtení/zápis datových souborů (default: .)"
     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 --data-dir /app/data             # Docker produkce"
 }
 
 # Parse arguments
 SCRAPER_ARGS=""
+DATA_DIR="."
 while [[ $# -gt 0 ]]; do
     case $1 in
         -h|--help)
@@ -46,6 +50,10 @@ while [[ $# -gt 0 ]]; do
             SCRAPER_ARGS="$SCRAPER_ARGS $1 $2"
             shift 2
             ;;
+        --data-dir)
+            DATA_DIR="$2"
+            shift 2
+            ;;
         *)
             echo "Unknown argument: $1"
             echo ""
@@ -55,6 +63,8 @@ while [[ $# -gt 0 ]]; do
     esac
 done
 
+SCRAPER_ARGS="$SCRAPER_ARGS --data-dir $DATA_DIR"
+
 step() {
     CURRENT=$((CURRENT + 1))
     echo ""
@@ -87,7 +97,7 @@ wait $PID_CH  || { echo -e "${RED}✗ CityHome selhalo${NC}"; FAILED=$((FAILED +
 # ── Sloučení + mapa ──────────────────────────────────────────
 
 step "Sloučení dat a generování mapy"
-python3 merge_and_map.py || { echo -e "${RED}✗ Merge selhal${NC}"; FAILED=$((FAILED + 1)); }
+python3 merge_and_map.py --data-dir "$DATA_DIR" || { echo -e "${RED}✗ Merge selhal${NC}"; FAILED=$((FAILED + 1)); }
 
 # ── Otevření mapy ────────────────────────────────────────────
 
@@ -100,4 +110,4 @@ else
 fi
 echo "============================================================"
 
-command -v open &>/dev/null && open mapa_bytu.html || true
+command -v open &>/dev/null && open "$DATA_DIR/mapa_bytu.html" || true

scrape_and_map.py

@@ -207,10 +207,10 @@ def load_cache(json_path: str = "byty_sreality.json") -> dict[int, dict]:
         return {}
 
 
-def scrape(max_pages: int | None = None, max_properties: int | None = None):
+def scrape(max_pages: int | None = None, max_properties: int | None = None, data_dir: str = "."):
     """Main scraping function. Returns list of filtered estates."""
     all_estates_raw = []
-    cache = load_cache()
+    cache = load_cache(str(Path(data_dir) / "byty_sreality.json"))
 
     logger.info("=" * 60)
     logger.info("Stahuji inzeráty ze Sreality.cz")
@@ -939,6 +939,8 @@ if __name__ == "__main__":
     parser.add_argument("--max-properties", type=int, 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)")
+    parser.add_argument("--data-dir", type=str, default=".",
+                        help="Directory for reading/writing data files (default: current dir)")
     args = parser.parse_args()
 
     # Configure logging
@@ -948,12 +950,13 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    data_dir = Path(args.data_dir)
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties, data_dir=args.data_dir)
 
     if estates:
         # Save raw data as JSON backup
-        json_path = Path("byty_sreality.json")
+        json_path = data_dir / "byty_sreality.json"
         json_path.write_text(
             json.dumps(estates, ensure_ascii=False, indent=2),
             encoding="utf-8",
@@ -961,7 +964,7 @@ if __name__ == "__main__":
         logger.info(f"✓ Data uložena: {json_path.resolve()}")
 
         # Generate map
-        map_path = generate_map(estates)
+        map_path = generate_map(estates, output_path=str(data_dir / "mapa_bytu.html"))
         elapsed = time.time() - start
         logger.info(f"\n⏱  Celkový čas: {elapsed:.0f} s")
         logger.info(f"\nOtevři v prohlížeči:\n  file://{map_path}")

scrape_bezrealitky.py

@@ -170,8 +170,8 @@ def load_cache(json_path: str = "byty_bezrealitky.json") -> dict[int, dict]:
         return {}
 
 
-def scrape(max_pages: int | None = None, max_properties: int | None = None):
-    cache = load_cache()
+def scrape(max_pages: int | None = None, max_properties: int | None = None, data_dir: str = "."):
+    cache = load_cache(str(Path(data_dir) / "byty_bezrealitky.json"))
 
     logger.info("=" * 60)
     logger.info("Stahuji inzeráty z Bezrealitky.cz")
@@ -395,6 +395,8 @@ if __name__ == "__main__":
                         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)")
+    parser.add_argument("--data-dir", type=str, default=".",
+                        help="Directory for reading/writing data files (default: current dir)")
     args = parser.parse_args()
 
     # Configure logging
@@ -404,11 +406,12 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    data_dir = Path(args.data_dir)
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties, data_dir=args.data_dir)
 
     if estates:
-        json_path = Path("byty_bezrealitky.json")
+        json_path = data_dir / "byty_bezrealitky.json"
         json_path.write_text(
             json.dumps(estates, ensure_ascii=False, indent=2),
             encoding="utf-8",

scrape_cityhome.py

@@ -194,8 +194,8 @@ def load_previous(json_path: str = "byty_cityhome.json") -> dict[str, str]:
         return {}
 
 
-def scrape(max_pages: int | None = None, max_properties: int | None = None):
-    previous_first_seen = load_previous()
+def scrape(max_pages: int | None = None, max_properties: int | None = None, data_dir: str = "."):
+    previous_first_seen = load_previous(str(Path(data_dir) / "byty_cityhome.json"))
     logger.info("=" * 60)
     logger.info("Stahuji inzeráty z CityHome (city-home.cz)")
     logger.info(f"Cena: do {format_price(MAX_PRICE)}")
@@ -367,6 +367,8 @@ if __name__ == "__main__":
                         help="Maximum number of properties to include in results")
     parser.add_argument("--log-level", type=str, default="INFO", choices=["DEBUG", "INFO", "WARNING", "ERROR"],
                         help="Logging level (default: INFO)")
+    parser.add_argument("--data-dir", type=str, default=".",
+                        help="Directory for reading/writing data files (default: current dir)")
     args = parser.parse_args()
 
     # Configure logging
@@ -376,11 +378,12 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    data_dir = Path(args.data_dir)
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties, data_dir=args.data_dir)
 
     if estates:
-        json_path = Path("byty_cityhome.json")
+        json_path = data_dir / "byty_cityhome.json"
         json_path.write_text(
             json.dumps(estates, ensure_ascii=False, indent=2),
             encoding="utf-8",

scrape_idnes.py

@@ -278,8 +278,8 @@ def load_cache(json_path: str = "byty_idnes.json") -> dict[str, dict]:
         return {}
 
 
-def scrape(max_pages: int | None = None, max_properties: int | None = None):
-    cache = load_cache()
+def scrape(max_pages: int | None = None, max_properties: int | None = None, data_dir: str = "."):
+    cache = load_cache(str(Path(data_dir) / "byty_idnes.json"))
 
     logger.info("=" * 60)
     logger.info("Stahuji inzeráty z Reality iDNES")
@@ -499,6 +499,8 @@ if __name__ == "__main__":
                         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)")
+    parser.add_argument("--data-dir", type=str, default=".",
+                        help="Directory for reading/writing data files (default: current dir)")
     args = parser.parse_args()
 
     # Configure logging
@@ -508,11 +510,12 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    data_dir = Path(args.data_dir)
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties, data_dir=args.data_dir)
 
     if estates:
-        json_path = Path("byty_idnes.json")
+        json_path = data_dir / "byty_idnes.json"
         json_path.write_text(
             json.dumps(estates, ensure_ascii=False, indent=2),
             encoding="utf-8",

scrape_psn.py

@@ -122,8 +122,8 @@ def load_previous(json_path: str = "byty_psn.json") -> dict[str, str]:
         return {}
 
 
-def scrape(max_pages: int | None = None, max_properties: int | None = None):
-    previous_first_seen = load_previous()
+def scrape(max_pages: int | None = None, max_properties: int | None = None, data_dir: str = "."):
+    previous_first_seen = load_previous(str(Path(data_dir) / "byty_psn.json"))
     logger.info("=" * 60)
     logger.info("Stahuji inzeráty z PSN.cz")
     logger.info(f"Cena: do {format_price(MAX_PRICE)}")
@@ -346,6 +346,8 @@ if __name__ == "__main__":
                         help="Maximum number of properties to include in results")
     parser.add_argument("--log-level", type=str, default="INFO", choices=["DEBUG", "INFO", "WARNING", "ERROR"],
                         help="Logging level (default: INFO)")
+    parser.add_argument("--data-dir", type=str, default=".",
+                        help="Directory for reading/writing data files (default: current dir)")
     args = parser.parse_args()
 
     # Configure logging
@@ -355,11 +357,12 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    data_dir = Path(args.data_dir)
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties, data_dir=args.data_dir)
 
     if estates:
-        json_path = Path("byty_psn.json")
+        json_path = data_dir / "byty_psn.json"
         json_path.write_text(
             json.dumps(estates, ensure_ascii=False, indent=2),
             encoding="utf-8",

scrape_realingo.py

@@ -135,8 +135,8 @@ def load_cache(json_path: str = "byty_realingo.json") -> dict[int, dict]:
         return {}
 
 
-def scrape(max_pages: int | None = None, max_properties: int | None = None):
-    cache = load_cache()
+def scrape(max_pages: int | None = None, max_properties: int | None = None, data_dir: str = "."):
+    cache = load_cache(str(Path(data_dir) / "byty_realingo.json"))
 
     logger.info("=" * 60)
     logger.info("Stahuji inzeráty z Realingo.cz")
@@ -354,6 +354,8 @@ if __name__ == "__main__":
                         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)")
+    parser.add_argument("--data-dir", type=str, default=".",
+                        help="Directory for reading/writing data files (default: current dir)")
     args = parser.parse_args()
 
     # Configure logging
@@ -363,11 +365,12 @@ if __name__ == "__main__":
         handlers=[logging.StreamHandler()]
     )
 
+    data_dir = Path(args.data_dir)
     start = time.time()
-    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties)
+    estates = scrape(max_pages=args.max_pages, max_properties=args.max_properties, data_dir=args.data_dir)
 
     if estates:
-        json_path = Path("byty_realingo.json")
+        json_path = data_dir / "byty_realingo.json"
         json_path.write_text(
             json.dumps(estates, ensure_ascii=False, indent=2),
             encoding="utf-8",