Welcome to the Lounge

30-Jul-19 6:01

that's a fair point.

spider eggs especially.

*shudder*

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

#realJSOP30-Jul-19 6:05

#realJSOP

30-Jul-19 6:05

It's not really our fault that you picked the wrong tool for the job at hand...

If you wanna go native, do it. You can still write un-managed DLLs that can be used in .Net apps.

I don't understand what the fuss is about.

".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013

honey the codewitch30-Jul-19 6:06

30-Jul-19 6:06

i like managed code. i just wish generics did more than they currently do. At least they added covariance with .NET 4 but it has a long way to go.

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

Slacker00730-Jul-19 6:28

30-Jul-19 6:28

honey the monster, codewitch wrote:
i just wish generics did more than they currently do

Um, Generics can do a lot in C# currently. How exactly are generics NOT working for you in C#?

honey the codewitch30-Jul-19 6:32

30-Jul-19 6:32

one example I'm running into right now is template specialization.

I have a finite state machine engine and it works for any transition input type and any accept symbol type.

However, there are additional features that can happen - significant ones that can only exist when the transition type is char - this specialization is effectively a regular expression engine, which means it can parse from a regular expression, and provide regex matching over string inputs. The other kind of FAs it wouldn't even make sense for that.

So because of this I have two separate classes - one generic FA<tinput> class, and one called CharFA where the TInput=char basically.

It means more code to maintain because a lot of it is duplicated. To unduplicate a lot of which i could, I'd have to add another codefile with an interface, and another with static methods to share common functionality, which again, increases the code size.

So it's not even that I can't do it with C#, it's that what is elegantly handled in C++ is clunky in C# to do the same thing, and requires more code.

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

Slacker00730-Jul-19 7:18

30-Jul-19 7:18

Create an interface class called IFa (name it something esle for sure, but this is for ilustration purposes) and implement IFa in a new class CharFA using that interface. or Create an iterface called IFa and a default implementation for IFa and then create another class that implements the default implementation and then just override any methods, etc.

It is object oriented programming, which you hate, but that is how this stuff is usually done, more or less, in C#.

Would that work for you?

Just a suggestion, I am not really trying to solve any problems here. I think your dislike of C# and generics, and object oriented programming may prevent you from seeing how things are done in this language, etc.

Good luck.

honey the codewitch30-Jul-19 7:21

30-Jul-19 7:21

which is what i mentioned in my other reply, after an edit though. it's not that I can't do it. It's that it's clunky and requires more code than the elegant specialization feature in C++

it reminds me of the limitation of lack of multiple inheritance - you can sort of emulate it, but it requires more code.

like i said, i just wish generics could do more.

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

Slacker00730-Jul-19 7:22

30-Jul-19 7:22

I think the "clunky" is relative here, as I don't think it is clunky because I have limited exposure to other techniques.

honey the codewitch30-Jul-19 7:24

30-Jul-19 7:24

that's definitely fair. and I come at C# from a C++ background.

I like C#, don't get me wrong, and it's miles ahead of Java in terms of how it's put together, IMO, but I still miss aspects of C++ development with it, even as it has supplanted C++ as my primary development language and environment.

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

Forogar 30-Jul-19 7:51

Forogar

30-Jul-19 7:51

I miss the multiple inheritance of C++ as well. I know we can do it via Interfaces but it's not quite as straightforward. Perhaps by C#12 it will be there. Sigh | :sigh:

- I would love to change the world, but they won’t give me the source code.

honey the codewitch30-Jul-19 7:54

30-Jul-19 7:54

i'm glad i'm not the only one! =)

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

Fueled By Decaff30-Jul-19 22:09

Fueled By Decaff

30-Jul-19 22:09

There are two ways that I can think of to avoid this code duplication. (Whether these are suitable is up to you.)

1. Implement your byte specific class as a subclass of your generic class?

2. Use dependency injection for the byte specific code.

Good luck

honey the codewitch31-Jul-19 1:24

31-Jul-19 1:24

the latter isn't practical.

the former i already did, and it's sloppy as hell

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

obermd31-Jul-19 4:17

obermd

31-Jul-19 4:17

Take a look at F# or another functional language. State machines and functional languages are made for each other.

honey the codewitch31-Jul-19 4:18

31-Jul-19 4:18

I've strongly considered it. I might eventually move, but I'm familiar with C#.

Maybe if they had Haskell I would have moved already.

Edit: Adding, one of the drawbacks of functional programming is lack of state, and some of these equations are so complicated that state is necessary for optimization and I wonder how a functional language will handle such a thing. Recomputing or doing lazy iteration over these algorithms is grossly impractical even if it's "correct"

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

Rick Shaub31-Jul-19 7:30

Rick Shaub

31-Jul-19 7:30

Memoization might be useful for you. It's basically a way to cache function results with a given input in a dictionary and return the cached value instead of running the calculation again.

honey the codewitch31-Jul-19 7:36

31-Jul-19 7:36

yeah. in fact, I need to explore memoization more anyway for implementing a PEG parser.

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

Slacker00730-Jul-19 6:29

30-Jul-19 6:29

Exactly

PIEBALDconsult30-Jul-19 7:40

PIEBALDconsult

30-Jul-19 7:40

There there.

I agree that a multi-paradigm language is a better idea than one which insists on objects.

One needs to use the right tool for the right job and OOP is not the right tool for a great many jobs.

honey the codewitch30-Jul-19 7:42

30-Jul-19 7:42

it's one of the areas where C++ really shines and I kind of wish other, higher level imperative languages would catch up.

though i'd also like to see C++ have more functional-programming constructs in the future.

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.

PIEBALDconsult30-Jul-19 8:47

PIEBALDconsult

30-Jul-19 8:47

I only ever dabbled in C++ (80s, 90s), so I never became familiar with what it can do.

I went straight from C (mostly on OpenVMS) to C# (and .net) and it was like a Bob-send -- I'm glad I hadn't had to use C++ and the various libraries people talk about. A lot of the hype I heard turned me off of C++ anyway.

But... I want multiple-inheritance and such. There are a number of facets of C# (.net languages) I don't like.

Languages and frameworks should provide features and _allow_ developers to do what their particular task requires rather than dictating what the develop must or must not do.

I may still need to look at D again.

honey the codewitch30-Jul-19 9:02

30-Jul-19 9:02

Well the weird thing about C++ is by itself it's about 2/3 of a language, while the standard template libraries are (usually) the other 1/3

and it's strange to think of it that way, but that's how it ends up baking out - STL is so intrinsic to any significant C++ development that you really don't even want to do it without it, just like you wouldn't want to write an app without an operating system.

It's not just about runtime libraries, although that's most of it. Because of the way templates work, you can use them to basically implement "language features" of a sort. So STL sort of folds itself into the language.

And the cool thing about that is you can potentially make your own "domain specific" language superset from C++ just like STL does - the spirit framework does this, and boost kind of does.

So that's something that's really hard to get used to at first because it's pretty unique to C++

after that, just learning how to use generic programming is the big learning curve, but I swear once you do, you'll fall in love.

When I was growin' up, I was the smartest kid I knew. Maybe that was just because I didn't know that many kids. All I know is now I feel the opposite.