Unreal Archive Content Submitter

A simple service and web page for accepting content submissions.

• allows user to submit file via HTTP form upload
• virus scan file using clamscan
• scan and index file with unreal-archive
• if a new file, clone/update unreal-archive-data repository and add content data
• push unreal-archive-data to a new branch and create a pull request via Github API
• someone will review the PR and accept it

How

• submission process needs to be asynchronous, as the scan/index/clone/push steps may take some time
• use long polling on client, build up log of events on server and update client as progress happens

Config

Service is configured using environment variables:

• GIT_REPO: URL of the archive data repository to clone and push to
• GIT_USERNAME: username of user used to clone and push repository changes
• GIT_PASSWORD: token for user
• GIT_EMAIL: email address to use on commits
• GH_TOKEN: github personal access token, for opening pull requests
• BIND_HOST: bind web service to this host
• BIND_PORT: web service listens on this port
• ALLOWED_ORIGIN: comma separated list of hosts from which to accept CORS requests for job queries and file uploads
• JOBS_PATH: path to where job history is stored
• STATS_HOST: host name/address of StatsD service
• STATS_PORT: port on STATS_HOST of StatsD service
• UPLOAD_PATH: path to upload temporary files to
• CLAM_SOCKET: path to a ClamD "LocalSocket" file, which can be reused. if this is not set, a new clamd process will be created with its own socket.

Data stores (for content hosting):

• No-Op store:

  • STORE=NOP: no-op storage, used for offline testing

• WebDAV store:

  • STORE=DAV: use WebDav storage, typically only used for testing
  • DAV_IMAGES, DAV_CONTENT: set the URL to PUT image or content files respectively
  • DAV_URL: default DAV URL to PUT files, only used if either of the above are not set

• S3 compatible object store:

  • STORE=S3: use S3 storage
  • S3_KEY: S3 key ID
  • S3_SECRET: S3 access secret
  • S3_BUCKET_IMAGES, BS3_BUCKET_CONTENT: bucket names for image and content files respectively
  • S3_BUCKET: default bucket name to store files, only used if either of the above are not set
  • S3_ENDPOINT: provide the root of your storage API
    • S3_ENDPOINT_IMAGES, BS3_ENDPOINT_CONTENT: optional alternative root endpoints if using different ones per content type
    • example: https://s3.amazonaws.com/
  • S3_URL: provide the public URL of your storage bucket in the appropriate region
    • example: https://__BUCKET__.s3.eu-west-2.amazonaws.com/__NAME__
    • __BUCKET__ and __NAME__ will be replaced by the bucket and uploaded filenames respectively