Prioritize older images & make use MM:SS format

extract_rowing_data.py

@@ -12,6 +12,7 @@ import argparse
 import base64
 import json
 import mimetypes
+import re
 import sys
 from pathlib import Path
 
@@ -130,6 +131,29 @@ def extract_rowing_data(image_path: str) -> dict:
     }
 
 
+_VERSION_RE = re.compile(r"^(?P<base>.+?)(?:\s+\((?P<num>\d+)\))?(?P<ext>\.\w+)$")
+
+
+def _keep_latest_versions(paths: list[Path]) -> list[Path]:
+    """Keep only the highest-numbered version of each base filename.
+
+    e.g. given IMG_5454_screen.jpg, IMG_5454_screen (1).jpg, IMG_5454_screen (3).jpg
+    only IMG_5454_screen (3).jpg is kept. No-suffix counts as version 0.
+    """
+    best: dict[str, tuple[int, Path]] = {}
+    for p in paths:
+        m = _VERSION_RE.match(p.name)
+        if not m:
+            best.setdefault(p.name, (0, p))
+            continue
+        base_key = m.group("base") + m.group("ext")
+        num = int(m.group("num")) if m.group("num") else 0
+        prev_num, _ = best.get(base_key, (-1, p))
+        if num > prev_num:
+            best[base_key] = (num, p)
+    return sorted(p for _, p in best.values())
+
+
 def main():
     parser = argparse.ArgumentParser(description="Extract rowing data from display photos")
     parser.add_argument("--image", help="Path to a single image")
@@ -145,12 +169,13 @@ def main():
 
     if args.dir:
         dir_path = Path(args.dir)
-        image_paths = sorted(
+        all_images = sorted(
             p for p in dir_path.iterdir() if p.suffix.lower() in IMAGE_EXTS
         )
-        if not image_paths:
+        if not all_images:
             print(f"No images found in {args.dir}")
             sys.exit(1)
+        image_paths = _keep_latest_versions(all_images)
     else:
         image_paths = [Path(args.image)]