Comment by jcon321 | Hacker News Reader

Comment by jcon321 | Hacker News Reader

Gitlab. Just use Issues you can do everything with the free tier. (It's called "Issues workflow" - gitlab goes a little overboard though, but I'd look at pictures of peoples issues list to get examples).

My opinion would be to not use all the fancy features that automatically tie issues to merge requests, releases, epics, pipelines etc... it's way to much for a small team that is not doing any type of management.

Just use some basic labels, like "bug" or "feature" and then use labels to denote where they are in the cycle such as "sprinted", "needs testing" etc. Can use the Boards feature if you want something nice to look at. Can even assign weights and estimates.

You can tie all the issues of a current sprint to a milestone, call the milestone a version or w/e and set a date. Now you have history of features/bugs worked on for a version.

In terms of testing, obviously automated tests are best and should just be a requirement built into every requirement. Some times though tests must be done manually, and in that case attach a word doc or use the comments feature on an issue for the "test plan".

If possible, could I get your opinion on a specific example? In my current situation, I was asked to add a feature which required a few (java) classes. So -

* It seems like this would have been a milestone?

* So then maybe a few issues for the different classes or requirements?

* For each issue, after/during development I would note what tests are needed, maybe in the comments section of the issue? Maybe in the description?

* And then automated tests using junit?

Each issue can be from a 1 line fix to a week or even 2 week thing. But by that point, probably a metaissue that corresponds to other issues or issue with inline subtasks (GitHub has check boxes).

Features cut across code, so no 1-1 mapping with classes. Tests are generally self-documenting and land alongside the feature they are for. You can document them, but likely either a comment in the issue/PR if technically interesting, or in a separate ~wiki doc as part of a broader specification.

Ideally each commit is valid & passes tests (see "conventional commits") and each issue/PR has accompanying tests whether around a new feature or bugfix. Particular test frameworks change every year.

I don't know your deployment schedule or rules. I represent milestones as groups of independent issues (bug fixes or new features) that would all go together as a release. I don't use milestones as a group of multiple issues that represent one requirement (that would be referred to as an epic). However epics are part of the paid version, however there's no reason why you couldn't use milestones this way.

If you have a requirement (doesn't matter how big or small) I'd treat that as 1 issue (regardless of how many java classes or lines of codes need modifying). If the issue is complex then within the issue's description you can use markdown (like checkboxes or bullet points) to identify subset requirements. However, if you can break that large requirement into functional changes that could exist/be deployed separately then I'd probably do multiple independent issues with some type of common identifier in the issue's name (or use your interpretation of milestones and put all those issues into 1).

If you use gitlab as your git repository then tying an issue to a merge request is easy and it would then show you the diff (aka all the changes to source code) that the issue required for implementation.

In terms of tests, same kind of answer - I don't know your rules. Every issue should have a test plan, perhaps using markdown in the issues description would convey that test plan the easiest. If you automate the test using junit then not sure the test plan is anything more than "make sure test xyz from junit passes", if it's a manual test then the issue's description can have a list of steps using markdown.

Reply on news.ycombinator.com