Upload folder using huggingface_hub

run_all_samples.py

@@ -1,40 +1,51 @@
-# run_all_samples.py
+# run_all_samples.py (with logging)
 import os
 import json
 from pathlib import Path
+from datetime import datetime
 
 from fastapi.testclient import TestClient
-from app import app as fastapi_app   # FIXED IMPORT
+from app import app as fastapi_app  # FIXED IMPORT
 
 
-# -----------------------------
-# 1. Paths
-# -----------------------------
+# ----------------------------------------------------
+# Paths
+# ----------------------------------------------------
 BASE_DIR = Path(__file__).parent
 SAMPLES_DIR = BASE_DIR / "samples"
 RESULTS_DIR = BASE_DIR / "results"
+LOG_FILE = BASE_DIR / "run_all_samples.log"
 
 RESULTS_DIR.mkdir(exist_ok=True)
 
+# ----------------------------------------------------
+# Logging helper
+# ----------------------------------------------------
+def log(msg: str):
+    timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+    with open(LOG_FILE, "a", encoding="utf-8") as f:
+        f.write(f"[{timestamp}] {msg}\n")
 
-# -----------------------------
-# 2. File filter
-# -----------------------------
+
+# ----------------------------------------------------
+# File filter
+# ----------------------------------------------------
 def is_bill_file(path: Path):
     return path.suffix.lower() in [".pdf", ".png", ".jpg", ".jpeg"]
 
 
-# -----------------------------
-# 3. MAIN EXTRACTION FUNCTION
-# -----------------------------
+# ----------------------------------------------------
+# Main extraction function
+# ----------------------------------------------------
 def main():
-    print("\n🚀 Running extractor on local sample files...\n")
-    print(f"📁 Using samples from: {SAMPLES_DIR}\n")
+    log("========== RUN STARTED ==========")
+    log(f"Samples directory: {SAMPLES_DIR}")
 
     if not SAMPLES_DIR.exists():
-        raise RuntimeError("❌ ERROR: 'samples/' folder does not exist in your Space.")
+        log("ERROR: samples/ folder does not exist!")
+        return
 
-    client = TestClient(fastapi_app)  # FIXED
+    client = TestClient(fastapi_app)
 
     summary = {}
 
@@ -42,7 +53,7 @@ def main():
         if not fp.is_file() or not is_bill_file(fp):
             continue
 
-        print(f"🔍 Processing: {fp.name}")
+        log(f"Processing file: {fp.name}")
 
         file_url = f"file://{fp.resolve()}"
 
@@ -50,23 +61,28 @@ def main():
             res = client.post(
                 "/extract-bill-data",
                 json={"document": file_url},
-                timeout=120
+                timeout=180
             )
         except Exception as e:
-            print(f"❌ Error calling extractor: {e}")
+            log(f"ERROR calling extractor: {e}")
             summary[fp.name] = "EXTRACTOR_ERROR"
             continue
 
         if res.status_code != 200:
-            print(f"❌ API error {res.status_code}")
+            log(f"ERROR API returned status {res.status_code}")
             summary[fp.name] = "API_ERROR"
             continue
 
         data = res.json()
 
         out_path = RESULTS_DIR / (fp.stem + ".json")
-        with open(out_path, "w", encoding="utf-8") as f:
-            json.dump(data, f, indent=2, ensure_ascii=False)
+        try:
+            with open(out_path, "w", encoding="utf-8") as f:
+                json.dump(data, f, indent=2, ensure_ascii=False)
+            log(f"Saved result JSON: {out_path}")
+        except Exception as e:
+            log(f"ERROR writing output JSON: {e}")
+            continue
 
         try:
             total_items = data["data"]["total_item_count"]
@@ -74,14 +90,16 @@ def main():
             total_items = "N/A"
 
         summary[fp.name] = total_items
-        print(f"   → Saved: {out_path}")
-        print(f"   → Items Extracted: {total_items}\n")
+        log(f"Items Extracted: {total_items}")
 
-    print("\n🎉 DONE! All sample bills processed.\n")
-    print("📊 Summary:")
+    # Summary
+    log("======== SUMMARY ========")
     for fn, items in summary.items():
-        print(f"  {fn}: {items}")
+        log(f"{fn}: {items}")
+
+    log("========== RUN FINISHED ==========\n")
 
 
+# Run manually if executed directly
 if __name__ == "__main__":
     main()