magic_lobster_party

I didn’t read everything, but I mostly agree with the author, especially on this point:

While you can definitely abuse exceptions, functional-style error values are not a one-size-fits-all solution.

There are time and place for both. I think exceptions are good for bigger errors. Like database connection errors. Things that shouldn’t happen without any easy backup plan. Those errors might need to be escalated as high as possible where proper action can be made (like resetting the database connection and everything relying on it).

Functional style is great for smaller stuff. Like key not found in hash maps. In many cases there might be good defaults that can be used instead.

It’s a concept of a plan

Haven’t read through this, but this sounds like what C++ is to C. I’m not sure adding more complexity and features to an already complex language is the right way forward. What is needed is a language that cuts down all the burden that has accumulated in C++ over 3 decades.

Something like Zig sounds like the better path forward to me. A completely new language from scratch with cross interoperability to C++. I’m surprised it’s not mentioned even once in the page.

The great thing about Slack is how easy it is to make automations. I guess this one just reads RSS feeds.

At my work we have automations notifying us about production errors for example.

I like to mix between OOP and FP for different levels. OOP is great for higher architectural problems. FP is great for everything under it.

And yes, inheritance was a huge mistake. Just use composition and interfaces instead.

I agree, and I count that as “key information that’s difficult to understand from the code”.

IMO, comments should be used to provide value to the code. If they’re used too much, then readers of the code will more likely stop reading them altogether. They already got what they need from the code itself and the comments usually don’t add much value.

If they’re sparse, then that’s a good indication they’re important and shouldn’t be missed.

I think JSON is more robust than XML by now. Mostly due to its simplicity. There are few reasons why anyone would pick XML over JSON these days.

It makes you want to die on a hill

I think comments are good as a last resort when it’s difficult to communicate the intention of the code with other means.

If I find code that’s hard to understand, I’ll first try to find better variable or function names. Often this is enough.

If it’s still too difficult to understand, I try to restructure the code to better communicate the flow of the code.

If that doesn’t help (or is too difficult), then I might add a comment explaining key information that’s difficult to understand from the code.

Yes, I think so. The downside with Python comes when refactoring the code. There’s always this double checking if the code is correctly indented after the refactor. Sometimes small mistakes creep in.

It’s really hard to tell when Python code is incorrectly indented. It’s often still valid Python code, but you can’t tell if it’s wrong unless you know the intention of the code.

In order languages it’s always obvious when code is incorrectly indented. There’s no ambiguity.

I don’t like YAML because it’s overly complicated. The specification is like 80 pages long. How the hell did they think that was a good idea?

JSON on the other hand is super simple. It doesn’t do more than it needs to.

Just compare this: https://yaml.org/spec/1.2.2/

With this: https://www.json.org/json-en.html

The entire JSON specification is shorter than just the table of contents of the YAML specification!

Another thing I like about JSON is that you can format it however you want with the whitespace. Want everything on one line? Just write everything on one line!

If data can be represented as a JSON, then there’s generally only one way to represent it in JSON (apart from whitespace). In YAML the same data can be represented in 1000s of different ways. You pick one.

Turn on a lamp

The only problem is to ensure the entire team agrees to only use it like an interface and nothing else. But I guess that’s the only proper way to do it in C++, for now.

In your example, the declaration of ArrayList look like:

public class ArrayList extends AbstractList implements List {
}

The dependence on AbstractList is public. Any public method in AbstractList is also accessible from the outside. It opens up for tricky dependencies that can be difficult to unravel.

Compare it with my solution:

public class ArrayList implements List {
    private AbstractList = new AbstractList();
}

Nothing about the internals of ArrayList is exposed. You’re free to change the internals however you want. There’s no chance any outside code will depend on this implementation detail.

If the lists have shared components then that can be solved with composition. It’s semantically the same as using abstract classes, but with the difference that this code dependency doesn’t need to be exposed to the outside. This makes the dependency more loosely coupled.

I usually break it out using composition if that’s ever needed. Either by wrapping around all the implementations, or as a separate component that is injected into each implementation.

Ask Bjarne to add interfaces enough many times until he gives in.

On a more serious note, I’m not exactly sure what the best C++ practice is. I guess you just have to live with abstract classes if you really want interfaces.

In 99% of the cases, inheritance can easily be replaced with composition and/or interfaces. Abstract classes tend to cause hard dependencies that are tough to work with.

I’m not sure why you would use abstract classes without data. Just use interfaces.

So things like abstract classes are mostly absent from my codebase.

I believe the consensus nowadays is that abstract classes should be avoided like the plague even in languages like Java and C#.

It doesn’t matter, why the present is garbage, it’s garbage and we should address that.

The problem is fixing it without inadvertently breaking for someone else. Changing the default behavior isn’t easy.

There’s probably some critical systems that relies on old outdated practices because that’s the way it worked when it was written 20 years ago. Why should they go back and fix their code when it has worked perfectly fine for the past two decades?