Skip to content

Generic HTTP #183

Closed
Closed

Description

@mxey

The README mentions plans for a “Generic” registry. Would that be a generic cache for HTTP resources? I'd be interested in files from GitHub releases, for example.

Activity

  1. mxey commented on Jul 7, 2026

    @mxey
    Author

    I thought a bit about how this might look:

    1. An allowlist of upstream domains/URLs, lest this dependency proxy becomes an open proxy for any HTTP GET requests. That would also provide a space in the configuration file to pick HTTP vs HTTPS and set authentication or other headers for the upstream request.
    2. Probably default to caching an upstream response forever, but make it clear that's how it works, and optionally have another endpoint that only returns a cached response if the upstream is unavailable?
    3. The upstream URL has to be part of the path of the proxy URL, not a query string, so it can be used as a prefix in configuration of downloading tools.

    With that in mind:

    The request to the proxy looks like GET http://localhost:8080/generic/immutable/github.com/spinel-coop/rv-ruby/releases/download/20260520/ruby-3.2.1.arm64_linux.tar.gz. This file would be cached forever, hence immutable

    The configuration says github.com is allowed and that it should use HTTPS (or HTTPS would be the default)

  2. added a commit that references this issue on Sep 4, 2026
    30e54a1
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions