Creating and Managing a Private Fork from a Public Repository

Creating a private fork of a public repository allows you to work on a project privately while still being able to sync with the original public source. This guide outlines the process of setting up a private fork, making changes, syncing with the public repo, and contributing back to the original project. Follow these steps to maintain your own private version of a public repository while staying up-to-date with upstream changes.

Step 1: Create a private fork:

Create a new private repository on GitHub.
Open terminal and run:

git clone --bare https://github.com/original-owner/public-repo.git
cd public-repo.git
git push --mirror https://github.com/your-username/private-repo.git
cd ..
rm -rf public-repo.git

Step 2: Work on your private fork:

Clone your private repo:

git clone https://github.com/your-username/private-repo.git
cd private-repo

Make changes, commit, and push:

git commit -am "Your changes"
git push origin main

Step 3: Sync with the public repo:

Add the public repo as a remote:

git remote add public https://github.com/original-owner/public-repo.git

Pull changes from the public repo:

git pull public main

Push updated code to your private repo:

git push origin main

Step 4: Create a pull request to the public repo:

Fork the public repo on GitHub.
Clone your public fork:

git clone https://github.com/your-username/public-repo-fork.git
cd public-repo-fork

Add your private repo as a remote:

git remote add private https://github.com/your-username/private-repo.git

Create a new branch for the pull request:

git checkout -b pr-branch
git pull private main
git push origin pr-branch

Create a pull request on GitHub from your public fork’s pr-branch to the original public repo.

Remember to repeat steps 3 and 4 as needed to keep your repos in sync and contribute changes back to the public repo.

Miscellaneous

Download from Private Repository

Download the latest release binary from a private GitHub repo. (i.e. a .tar.gz that you have manually uploaded in a GitHub release). Update OAUTH_TOKEN, OWNER, REPO, FILE_NAME with your custom values.

#!/usr/bin/env bash

# Authorize to GitHub to get the latest release tar.gz
# Requires: oauth token, https://help.github.com/articles/creating-an-access-token-for-command-line-use/
# Requires: jq package to parse json

# Your oauth token goes here, see link above
OAUTH_TOKEN="34k234lk234lk2j3lk4j2l3k4j2kj3lk"
# Repo owner (user id)
OWNER="your-user-name"
# Repo name
REPO="the-repo-clean-name"
# The file name expected to download. This is deleted before curl pulls down a new one
FILE_NAME="file-you-are-downloading.tar.gz"

# Concatenate the values together for a
API_URL="https://$OAUTH_TOKEN:@api.github.com/repos/$OWNER/$REPO"

# Gets info on latest release, gets first uploaded asset id of a release,
# More info on jq being used to parse json: https://stedolan.github.io/jq/tutorial/
ASSET_ID=$(curl $API_URL/releases/latest | jq -r '.assets[0].id')
echo "Asset ID: $ASSET_ID"

# curl does not allow overwriting file from -O, nuke
rm -f $FILE_NAME

# curl:
# -O: Use name provided from endpoint
# -J: "Content Disposition" header, in this case "attachment"
# -L: Follow links, we actually get forwarded in this request
# -H "Accept: application/octet-stream": Tells api we want to dl the full binary
curl -O -J -L -H "Accept: application/octet-stream" "$API_URL/releases/assets/$ASSET_ID"