A workaround I've used sometimes to retrigger CI is to do `git commit --amend --no-edit --reset-author`. This changes the sha by changing the author timestamp. Then you can force push it to the PR branch

You don't even need `--reset-author`; the committer timestamp changes without it too:

    $ git init
    $ git commit -m hello --allow-empty
    [master (root-commit) 1177c71] hello
    $ git commit --amend --no-edit --allow-empty
    [master (root-commit) 7dcb290] hello

There's an item on their roadmap to allow commenting on unchanged lines: https://github.com/github/roadmap/issues/456

I think the person you're replying to is saying that "fully implement every unit test up front" does not describe TDD. With TDD you continuously alternate between writing tests and writing production code. It's not like you write a whole test suite up front.

Yes this was my point.

There must be a typo somewhere in your comment, I don't understand how 350/week for 5 months is better than 500/month.

I guess what GP meant is that you can rent only whenever you need it, as opposed to 500/month for _every_ month you own?

I actually meant “rent the car for X month”, but I made some mistakes here. First, I took the exact numbers for the calculation, but rounded them generously in favor of owning the car in the post. Second I made some sort of mistake, doing the math, so I’m still slightly off. The correct math:

The smallest golf on the list is 579 EUR/Month, that’s 6.948 EUR per year. Renting a Golf at starcar costs 1436 EUR for the month of August. That’s 4.84, just shy of 5.

This seems to be covered by the new ability to extend issues with custom fields.

not really - u can have multiple labels for an issue. but one value for a custom field.

that's why the scoped/hierarchical stuff start becoming powerful for issues.

Gitlab had the right insight here.

I guess you have to be careful not to return inside of these :)

I don't understand why you wouldn't be able to, especially given "a million years" to learn whatever you need to learn. Why does it feel inaccessible to you?

I'm not good at implementing algorithms that directly relate to mathematical equations. I learned this from 1 year banging away at computational photography while in university and 2 years banging away at automated trading systems at one of my first jobs.

Given enough time I can copy other peoples implementations, but its extremely mechanistic to the point where you can't say that I actually 'wrote' it or learned anything. And it would be questionable if all the functionality would fit together as a succinctly as the posters.

I hope you don't think that is goal post moving, but if I do something mathematical in order to get it correct I have to sacrifice everything else: speed of production, efficiency of the end-product, readability of the code, etc. Compared to my output in logic problems or HCI, you'd think two entirely different people were involved and one was significantly smarter.

It's a failing (of the sort where asked "What is your biggest weakness?" at an interview, I can always answer immediately this), but it's what allows me to be impressed with work like an entire RTS game engine in C in just 3 years as a passion project.

It makes sense to me that merge commits would "contain changes" (i.e. that a merge commit's tree would have content that is not in either parent's tree), otherwise how else would you represent the result of resolving merge conflicts?

ripgrep isn't a 1:1 rewrite of grep. It's a different program.

The government decides all the time that some people should be punished. It's the justice system.