cv-analysis-service/scripts/manage_minio.py

80 lines
2.4 KiB
Python

import argparse
import gzip
import json
import os
from pathlib import Path
from tqdm import tqdm
from pyinfra.config import get_config
from pyinfra.storage.storage import get_s3_storage
CONFIG = get_config()
def parse_args():
parser = argparse.ArgumentParser()
subparsers = parser.add_subparsers(help="sub-command help", dest="command")
parser_add = subparsers.add_parser("add", help="Add file(s) to the MinIO store")
parser_add.add_argument("dossier_id")
add_group = parser_add.add_mutually_exclusive_group(required=True)
add_group.add_argument("--file", "-f")
add_group.add_argument("--directory", "-d")
subparsers.add_parser("purge", help="Delete all files and buckets in the MinIO store")
parser_show = subparsers.add_parser("show", help="Show compressed json from storage")
parser_show.add_argument("dossier_id")
parser_show.add_argument("file_name")
args = parser.parse_args()
return args
def combine_dossier_id_and_file_id_and_extension(dossier_id, file_id, extension):
return f"{dossier_id}/{file_id}{extension}"
def add_file_compressed(storage, bucket_name, dossier_id, path) -> None:
if Path(path).suffix == ".pdf":
suffix_gz = ".ORIGIN.pdf.gz"
if Path(path).suffix == ".json":
suffix_gz = ".TEXT.json.gz"
path_gz = combine_dossier_id_and_file_id_and_extension(dossier_id, Path(path).stem, suffix_gz)
with open(path, "rb") as f:
data = gzip.compress(f.read())
storage.put_object(bucket_name, path_gz, data)
if __name__ == "__main__":
storage = get_s3_storage(CONFIG)
bucket_name = CONFIG.storage_bucket
if not storage.has_bucket(bucket_name):
storage.make_bucket(bucket_name)
args = parse_args()
if args.command == "add":
if args.file:
add_file_compressed(storage, bucket_name, args.dossier_id, args.file)
elif args.directory:
for fname in tqdm([*os.listdir(args.directory)], desc="Adding files"):
path = Path(args.directory) / fname
add_file_compressed(storage, bucket_name, args.dossier_id, path)
elif args.command == "purge":
storage.clear_bucket(bucket_name)
elif args.command == "show":
json_file = storage.get_object(bucket_name, f"{args.dossier_id}/{args.file_name}")
json_file = gzip.decompress(json_file)
parsed = json.loads(json_file)
print(json.dumps(parsed, indent=2))