Commit 8a4b0a14 authored by Markus Scheidgen's avatar Markus Scheidgen
Browse files

Added a script that merges uploads into a single tar with no file dups. [skip ci]

parent acdbca15
Pipeline #51983 skipped
from typing import Dict
import zipfile
import tarfile
import os.path
import sys
from nomad import config
names: Dict[str, bool] = dict()
def add_upload(tf, upload):
files = [
os.path.join(config.fs.public, upload[0:2], upload, base)
for base in ['', '']]
for f in files:
with zipfile.ZipFile(f) as zf:
for zipinfo in zf.infolist():
name = zipinfo.filename
if name not in names:
names[name] = True
with as bf:
tarinfo = tarfile.TarInfo(name)
tarinfo.size = zipinfo.file_size
tf.addfile(tarinfo, bf)
target = sys.argv[1]
with tarfile.TarFile(target, 'x') as tf:
for upload in sys.argv[2:]:
print('adding upload %s' % upload)
add_upload(tf, upload)
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment