Files
pfm-ocr/backend/cleanup_specific_images.py
T

53 lines
1.7 KiB
Python

import os
import json
def main():
to_delete = [
"IMG_20260701_134635.jpg",
"IMG_20260701_134636.jpg"
]
base_dir = "backend"
if not os.path.exists(base_dir):
base_dir = "."
uploads_dir = os.path.join(base_dir, "uploads")
public_dir = os.path.join(base_dir, "pfm-web-app/public")
for filename in to_delete:
# Delete manual label json
ml_path = os.path.join(uploads_dir, f"manual_label_{filename}.json")
if os.path.exists(ml_path):
os.remove(ml_path)
print(f"Deleted: {ml_path}")
# Delete public image
img_public_path = os.path.join(public_dir, "test-images", filename)
if os.path.exists(img_public_path):
os.remove(img_public_path)
print(f"Deleted: {img_public_path}")
# Filter test_images_results.jsonl
jsonl_path = os.path.join(uploads_dir, "test_images_results.jsonl")
if os.path.exists(jsonl_path):
remaining_lines = []
with open(jsonl_path, "r", encoding="utf-8") as f:
for line in f:
if not line.strip():
continue
try:
data = json.loads(line)
fn = data.get("filename")
if fn not in to_delete:
remaining_lines.append(line)
except Exception as e:
print(f"Error parsing line: {e}")
remaining_lines.append(line)
with open(jsonl_path, "w", encoding="utf-8") as f:
f.writelines(remaining_lines)
print(f"Filtered results JSONL. Remaining entries: {len(remaining_lines)}")
if __name__ == "__main__":
main()