Mercurial > repos > greg > data_manager_gtdbtk_database_installer
diff data_manager/gtdbtk_database_installer.py @ 0:3ab83cb7e2d2 draft
Uploaded
author | greg |
---|---|
date | Tue, 15 Mar 2022 15:32:31 +0000 |
parents | |
children | 7093598fa300 |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/data_manager/gtdbtk_database_installer.py Tue Mar 15 15:32:31 2022 +0000 @@ -0,0 +1,70 @@ +#!/usr/bin/env python + +import argparse +import json +import os +import sys +import tarfile +from urllib.request import Request, urlopen +from urllib.parse import urlparse + + +def url_download(url, work_dir): + url_parts = urlparse(url) + file_path = os.path.abspath(os.path.join(work_dir, os.path.basename(url_parts.path))) + src = None + dst = None + try: + req = Request(url) + src = urlopen(req) + with open(file_path, 'wb') as dst: + while True: + chunk = src.read(2**10) + if chunk: + dst.write(chunk) + else: + break + except Exception as e: + sys.exit(str(e)) + finally: + if src: + src.close() + if tarfile.is_tarfile(file_path): + fh = tarfile.open(file_path, 'r:*') + else: + return file_path + fh.extractall(work_dir) + os.remove(file_path) + return work_dir + + +def download(database_id, database_name, url, out_file): + + with open(out_file) as fh: + params = json.load(fh) + + work_dir = params['output_data'][0]['extra_files_path'] + os.makedirs(work_dir) + file_path = url_download(url, work_dir) + + data_manager_json = {"data_tables": {}} + data_manager_entry = {} + data_manager_entry['value'] = database_id + data_manager_entry['name'] = database_name + data_manager_entry['path'] = file_path + data_manager_json["data_tables"]["gtdbtk_database"] = data_manager_entry + + with open(out_file, 'w') as fh: + json.dump(data_manager_json, fh, sort_keys=True) + + +parser = argparse.ArgumentParser() + +parser.add_argument('--database_name', dest='database_name', help='GTDB-Tk database display name') +parser.add_argument('--database_id', dest='database_id', help='Unique GTDB-Tk database id') +parser.add_argument('--url', dest='url', help='URL to download GTDB-Tk databse version') +parser.add_argument('--out_file', dest='out_file', help='JSON output file') + +args = parser.parse_args() + +download(args.database_id, args.database_name, args.url, args.out_file)