3 ms·

Ask HN: How do you Track new Versions of Dependencies?

I have been thinking a lot about pro-active dependency management. I would like to have a tool or a way to gather latest release versions of various tools and then ideally get alerted if a new version gets released. For example Gradle is at 7.3.1, if 7.3.2 becomes available I would like to know about it. Doing it for a single tool is simple, but having to track latest Git version, NPM, Gradle and all the other tools I work with doesn't scale very well. Is there an existing solution to this problem?

Jugurtha 5y ago
Append `/releases.atom` to any GitHub repository, and you'll get an RSS feed. Example: https://github.com/gradle/gradle/releases.atom https://github.com/gradle/gradle/releases.atom You can add that to your email client, such as Thunderbird. You get the releases and the release notes. I use that for our dependencies and I read the release notes. I also pull the repos and read many commit logs and diffs to predict the direction of these dependencies: if I see a lot of "movement" and "activity" in a category, I can fairly accurately predict that there will be more and I can shape our engineering effort that way. Example in Thunderbird on Ubuntu: - Click on the menu - Click New - Click "Feed account" - Enter a name "My project's dependencies" - Click "Next" - Click "Finish" You will have one "My project's dependencies" element appear on the left sidebar in Thunderbird. Click on that, then: - Click "Manage feed subscriptions", from the right pane. - Enter: https://github.com/gradle/gradle/releases.atom https://github.com/gradle/gradle/releases.atom in the "Feed URL" field on the window that appears - Click "Add" - Click "Close". You will see "Release notes from gradle" appear on the left sidebar, as a child element of "My project's dependencies" 'feed account'.
- 100011_100001 5y ago
  This is awesome and I love how simple it is.
- 100011_100001 5y ago
  This works like a charm for me. Now that I am using it, I can't believe how easy it is.
  - Jugurtha 5y ago
    It's nice, right? The team also uses that now and we can catch a lot of things this way. Like, a lot. The product we make is heavy on integration and orchestration of a bunch of tools, technologies, systems, and librarires. It takes complexity away from people involved in machine learning, so we implicitly inherit that complexity and, therefore, there are a lot of things we must track. This is effective.

Jugurtha 5y ago

Append `/releases.atom` to any GitHub repository, and you'll get an RSS feed. Example: https://github.com/gradle/gradle/releases.atom https://github.com/gradle/gradle/releases.atom You can add that to your email client, such as Thunderbird. You get the releases and the release notes. I use that for our dependencies and I read the release notes. I also pull the repos and read many commit logs and diffs to predict the direction of these dependencies: if I see a lot of "movement" and "activity" in a category, I can fairly accurately predict that there will be more and I can shape our engineering effort that way. Example in Thunderbird on Ubuntu: - Click on the menu - Click New - Click "Feed account" - Enter a name "My project's dependencies" - Click "Next" - Click "Finish" You will have one "My project's dependencies" element appear on the left sidebar in Thunderbird. Click on that, then: - Click "Manage feed subscriptions", from the right pane. - Enter: https://github.com/gradle/gradle/releases.atom https://github.com/gradle/gradle/releases.atom in the "Feed URL" field on the window that appears - Click "Add" - Click "Close". You will see "Release notes from gradle" appear on the left sidebar, as a child element of "My project's dependencies" 'feed account'.

100011_100001 5y ago
This is awesome and I love how simple it is.
100011_100001 5y ago
This works like a charm for me. Now that I am using it, I can't believe how easy it is.
- Jugurtha 5y ago
  It's nice, right? The team also uses that now and we can catch a lot of things this way. Like, a lot. The product we make is heavy on integration and orchestration of a bunch of tools, technologies, systems, and librarires. It takes complexity away from people involved in machine learning, so we implicitly inherit that complexity and, therefore, there are a lot of things we must track. This is effective.