Getting corrupt zips using Python3 ZipStream in Django

Question

I'm using zipstream from here and have a Django view that returns a zip file of all file attachments which are all hosted on Amazon S3. But the zip files are all coming up as corrupt when I download them, that is, I can't open them.

import io
import zipstream
s = io.BytesIO()
with zipstream.ZipFile(s,"w", compression=zipstream.ZIP_DEFLATED) as zf:
    for file_path in file_paths:
        file_dir, file_name = os.path.split(file_path)

        zf.writestr(file_name, urllib.urlopen(file_path).read())

response = StreamingHttpResponse(s.getvalue(), content_type='application/octet-stream')
response['Content-Disposition'] = 'attachment; filename={}'.format('files.zip')
return response

Answer 1

Instead of zipstream package install aiozipstream package. If you've alreday installed the zipstream package uninstall it first. pip uninstall zipstream and then do a pip install aiozipstream

#Do the import in the same way
from zipstream import ZipStream
from django.http import StreamingHttpResponse

def s3_content_generator(key):
    #s3_bucket - your s3 bucket name
    infile_object = s3.get_object(Bucket=s3_bucket, Key= key)
    data = str(infile_object.get('Body', '').read())
    yield bytes(data, 'utf-8')


files = []
#filepath - list of s3 keys
for keyin filepath:
    files.append({'stream': s3_content_generator(key),'name': 'filename'})

#large chunksize fasten the download process
zf = ZipStream(files, chunksize=32768)
response = StreamingHttpResponse(zf.stream(), content_type='application/zip')
response['Content-Disposition'] = 'attachment; filename={}'.format('files.zip')
return response