Python library for handling chunk uploads (based on flow-php-server). Library contains helper methods for:
- Testing if uploaded file chunk exists.
- Validating file chunk
- Creating separate chunks folder
- Validating uploaded chunks
- Merging all chunks to a single file
The library currently only supports Tornado, but it is trivial to implement interfaces for other web servers (flask, etc).
This library is compatible with HTML5 file upload library: https://github.com/flowjs/flow.js
Install the pip module:
pip install flowjs
Create a new py file named upload.py
:
# Import flowjs
import uuid
import os
import flowjs
config = flowjs.Config()
request = flowjs.requests.TornadoRequest(tornado_http_request)
upload_directory = "/path/to/final/upload/directory"
upload_file_name = uuid.uuid1().hex + "_" + request.get_file_name()
upload_path = os.path.join(upload_directory, upload_file_name)
try:
result = flowjs.save(upload_path, config, request)
if result:
# file was saved to path
pass
else:
# not final chunk or invalid request
pass
except flowjs.NoContent:
# Set status to 204 No Content
# Do _not_ return any content in the response body
except flowjs.BadRequest:
# Set status to 400 Bad Request
Temporary directory where chunks are stored (will default to your system temp dir).
Make sure that the temp path exists and is writable. All chunks will be saved in this folder.
Function applied to the request to generate a unique identifier for the file.
Defaults to a SHA1 hash of the flowIdentifier
Function applied before each chunk is merged into the resultant file.
Should the server delete the file chunks after it saves the final file.
Defaults to True
To use this module with other webservers, implement the IRequest
and IFile
interfaces with your webserver's specific implementation details.
Your participation in development is very welcome!