mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-10 17:44:23 +00:00
98 lines
3.7 KiB
Python
Executable File
98 lines
3.7 KiB
Python
Executable File
#!/usr/bin/env python
|
|
# -*- coding: utf-8 -*-
|
|
import os
|
|
import sys
|
|
import tarfile
|
|
import logging
|
|
import argparse
|
|
import requests
|
|
import tempfile
|
|
|
|
|
|
DEFAULT_URL = 'https://clickhouse-datasets.s3.yandex.net'
|
|
|
|
AVAILABLE_DATASETS = {
|
|
'hits': 'hits_v1.tar',
|
|
'visits': 'visits_v1.tar',
|
|
'hits_100m_single': 'hits_100m_single.tar',
|
|
'hits_1000m_single': 'hits_1000m_single.tar',
|
|
'hits_10m_single': 'hits_10m_single.tar',
|
|
'trips_mergetree': 'trips_mergetree.tar',
|
|
}
|
|
|
|
def _get_temp_file_name():
|
|
return os.path.join(tempfile._get_default_tempdir(), next(tempfile._get_candidate_names()))
|
|
|
|
def build_url(base_url, dataset):
|
|
return os.path.join(base_url, dataset, 'partitions', AVAILABLE_DATASETS[dataset])
|
|
|
|
def dowload_with_progress(url, path):
|
|
logging.info("Downloading from %s to temp path %s", url, path)
|
|
with open(path, 'w') as f:
|
|
response = requests.get(url, stream=True)
|
|
response.raise_for_status()
|
|
total_length = response.headers.get('content-length')
|
|
if total_length is None or int(total_length) == 0:
|
|
logging.info("No content-length, will download file without progress")
|
|
f.write(response.content)
|
|
else:
|
|
dl = 0
|
|
total_length = int(total_length)
|
|
logging.info("Content length is %ld bytes", total_length)
|
|
counter = 0
|
|
for data in response.iter_content(chunk_size=4096):
|
|
dl += len(data)
|
|
counter += 1
|
|
f.write(data)
|
|
done = int(50 * dl / total_length)
|
|
percent = int(100 * float(dl) / total_length)
|
|
if sys.stdout.isatty():
|
|
sys.stdout.write("\r[{}{}] {}%".format('=' * done, ' ' * (50-done), percent))
|
|
sys.stdout.flush()
|
|
elif counter % 1000 == 0:
|
|
sys.stdout.write("{}%".format(percent))
|
|
sys.stdout.flush()
|
|
sys.stdout.write("\n")
|
|
logging.info("Downloading finished")
|
|
|
|
def unpack_to_clickhouse_directory(tar_path, clickhouse_path):
|
|
logging.info("Will unpack data from temp path %s to clickhouse db %s", tar_path, clickhouse_path)
|
|
with tarfile.open(tar_path, 'r') as comp_file:
|
|
comp_file.extractall(path=clickhouse_path)
|
|
logging.info("Unpack finished")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
logging.basicConfig(
|
|
level=logging.INFO,
|
|
format='%(asctime)s %(levelname)s: %(message)s')
|
|
|
|
parser = argparse.ArgumentParser(
|
|
description="Simple tool for dowloading datasets for clickhouse from S3")
|
|
|
|
parser.add_argument('--dataset-names', required=True, nargs='+', choices=AVAILABLE_DATASETS.keys())
|
|
parser.add_argument('--url-prefix', default=DEFAULT_URL)
|
|
parser.add_argument('--clickhouse-data-path', default='/var/lib/clickhouse/')
|
|
|
|
args = parser.parse_args()
|
|
datasets = args.dataset_names
|
|
logging.info("Will fetch following datasets: %s", ', '.join(datasets))
|
|
for dataset in datasets:
|
|
logging.info("Processing %s", dataset)
|
|
temp_archive_path = _get_temp_file_name()
|
|
try:
|
|
download_url_for_dataset = build_url(args.url_prefix, dataset)
|
|
dowload_with_progress(download_url_for_dataset, temp_archive_path)
|
|
unpack_to_clickhouse_directory(temp_archive_path, args.clickhouse_data_path)
|
|
except Exception as ex:
|
|
logging.info("Some exception occured %s", str(ex))
|
|
raise
|
|
finally:
|
|
logging.info("Will remove dowloaded file %s from filesystem if it exists", temp_archive_path)
|
|
if os.path.exists(temp_archive_path):
|
|
os.remove(temp_archive_path)
|
|
logging.info("Processing of %s finished, table placed at", dataset)
|
|
logging.info("Fetch finished, enjoy your tables!")
|
|
|
|
|