Repo Archiving Tool

A small archiving tool that helps speed up the process of archiving multiple repos via the GitHub API. A second script helps manage adding a collaborator to multiple repos (for example setting a robot user to be a maintainer)

Installation

All from within your terminal --->

1. Clone this repo to your local computer
2. Type cd repo-archiving-tool and press enter to go into the new folder
3. Type npm install and hit enter.
4. Type touch .env and press enter to create a .env file.

Personal Access Token

The GitHub API requires a PAT (Personal Access Token) to allow you to interact with it, as well as to ensure security. You must edit the .env file with you own GitHub PAT.

You can do this by adding pat= followed by your PAT.

It should look similar to pat=7g9h9uhui9dh1d8hd012h01 for a classic PAT or pat=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 for a fine-grained PAT. (In both cases, these are random alphanumeric strings, not real PATs!)

How to create a Personal Access Token

Instructions

Archiving repos

1. Ensure that every repo has had all their PRs closed and the readme file states that the repo is now archived, as per Verify ADR #0039, if you have access.
2. List the names of each repo by adding an array called reposToBeArchived to the .env file, one repo name per element. Make sure that you have the exact name (it is case-sensitive) or else it will throw an error. It should look like reposToBeArchived=["repo1", "repo2", "repo3"] etc.
3. Now you can run the program by typing node batch-archive.js into the terminal. As it is running, you will get a readout of each repo as it is archived.
4. Any errors that occur will be flagged by the program, with a message to help you resolve the issue.
5. Check the repos to ensure the job it completed.

Adding collaborators

The intended use-case here is to add a robot user to be maintainer to several repos at once.

1. As with archival, the repo list is contained in the .env file in a property reposList=["repo1", "repo2", "repo3"] etc.

2. Adding a collaborator also needs that user to be included in the .env file. Add another property like userToEscalate=AhmedFulan

3. Adding a collaborator also needs you to specify the privilege level for the user. As it is expected that this script will be used to add maintainers, this privilege level is hard-coded in the script, as is the organisation owner for the repo:

   const repoOwner = 'alphagov';
   const escalationPrivilege = 'maintain';

   The acceptable values for the privilege level are pull, triage, push, maintain, admin and any custom repository role name defined by the owning organization. See Add a repository collaborator in the GitHub API docs for more information.

4. Now you can run the program by typing node batch-maintainer.js into the terminal. As it is running, you will get a readout of each repo as it is archived.

5. Any errors that occur will be flagged by the program, with a message to help you resolve the issue.

6. Check the repos to ensure the job it completed.

Possibilities

Further to being able to batch archive, there is a wealth of other elements of a repo that you can adjust using the API via a request call. There is an exhaustive list of possible functionality here --->

List of Repository update parameters