Skip to content

hansharhoff/dash-uploader

 
 

Repository files navigation

GitHub tag (latest by date) PyPI PyPI - Downloads GitHub

upload large files with dash-uploader

📤 dash-uploader

The alternative upload component for Dash applications.

Table of contents

Short summary

      💾 Data file size has no limits. (Except the hard disk size)
      ☎ Call easily a callback after uploading is finished.
      📤 Upload files using resumable.js
      📦 All JS and CSS bundled with the package. No need for network calls for CSS/JS.
      ✅ Works with Dash 1.1.0.+ & Python 3.6+. (Possibly with other versions, too)

dash-uploader vs. dcc.Upload

dash-uploader dcc.Upload
Underlying technology resumable.js HTML5 API
File size Unlimited max ~150-200Mb (link)
Uploads to Hard disk (server side) First to browser memory (user side) Then, to server using callbacks.
Data type Uploaded as file; no need to parse at server side. Uploaded as byte64 encoded string -> Needs parsing
See upload progress? Progressbar out of the box No upload indicators out of the box. Generic loading indicator possible. Progressbar not possible.

Installing

pip install dash-uploader

Quickstart

Full documentation here

⚠️Security note: The Upload component allows POST requests and uploads of arbitrary files to the server harddisk and one should take this into account (with user token checking etc.) if used as part of a public website! For this you can utilize the http_request_handler argument of the du.configure_upload. (New in version 0.5.0)

Simple example

import dash
import dash_html_components as html
import dash_uploader as du

app = dash.Dash(__name__)

# 1) configure the upload folder
du.configure_upload(app, r"C:\tmp\Uploads")

# 2) Use the Upload component
app.layout = html.Div([
    du.Upload(),
])

if __name__ == '__main__':
    app.run_server(debug=True)

Example with callback (and other options)

  • New in version 0.3.0: New short callback syntax using @du.callback.
  • New in version 0.2.0: The configure_upload accepts additional parameter use_upload_id, which is a boolean flag (Defaults to True). When True, the uploaded files are put into subfolders <upload_folder>/<upload_id>. This way different users can be forced to upload to different folders.
from pathlib import Path
import uuid

import dash_uploader as du
import dash
import dash_html_components as html
from dash.dependencies import Input, Output, State

app = dash.Dash(__name__)

UPLOAD_FOLDER_ROOT = r"C:\tmp\Uploads"
du.configure_upload(app, UPLOAD_FOLDER_ROOT)

def get_upload_component(id):
    return du.Upload(
        id=id,
        max_file_size=1800,  # 1800 Mb
        filetypes=['csv', 'zip'],
        upload_id=uuid.uuid1(),  # Unique session id
    )


def get_app_layout():

    return html.Div(
        [
            html.H1('Demo'),
            html.Div(
                [
                    get_upload_component(id='dash-uploader'),
                    html.Div(id='callback-output'),
                ],
                style={  # wrapper div style
                    'textAlign': 'center',
                    'width': '600px',
                    'padding': '10px',
                    'display': 'inline-block'
                }),
        ],
        style={
            'textAlign': 'center',
        },
    )


# get_app_layout is a function
# This way we can use unique session id's as upload_id's
app.layout = get_app_layout


@du.callback(
    output=Output('callback-output', 'children'),
    id='dash-uploader',
)
def get_a_list(filenames):
    return html.Ul([html.Li(filenames)])


if __name__ == '__main__':
    app.run_server(debug=True)

Contributing

What? How?
🐞 Found a bug? 🎟 File an Issue
🙋‍♂️ Need help? Ask a question on StackOverflow
📧 Use this thread on community.plotly.com
💡 Want to submit a feature request? 🎭 Discuss about it on community.plotly.com
🎫 File an Issue (feature request)
🧙 Want to write code? 🔥 Here's how you get started!

Documentation

Changelog

Credits

About

Upload large data files using Dash for Python

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 43.7%
  • CSS 33.6%
  • JavaScript 22.5%
  • HTML 0.2%