Action! on Rosetta Code

[Tigerofdarkness]

6 hours ago, ilmenit said:

Can be a great test set for Effectus

Exactly and for any other implementation of the Action! language that might appear.

 

2 hours ago, zbyti said:

So little people programming in anything other than ASM on 8-bit at all that almost every time you start typing, you find a compiler/parser flaw or chance to improvement.

 

I been there done that.

If that's not an exageration, it's a sad reflection on the state of those compilers.

 

[zbyti]

25 minutes ago, Tigerofdarkness said:

If that's not an exageration, it's a sad reflection on the state of those compilers.

This is of course a bit of an exaggeration, but quite often when I try something longer then few lines I find 1 or 2 bugs.

 

It's hard for compilers on 6502 make optimizations without brek something.

 

On the other hand games like FloB shows that you can develop a big project successfully.

 

There are always bugs and there will be, the problem is that only few people are looking for it.

 

Encouraging to write in Action! not help, but its hobby - everyone do that they like  

[danwinslow]

19 minutes ago, zbyti said:

Encouraging to write in Action! not help

Doesn't help *your* goals, maybe. What's wrong with Action? It's a great language, and look, no errors, no bugs that you can find in 'longer than a few lines'. What is it exactly you think we should do? Encourage use of languages you like?

[zbyti]

I know the capabilities of the Action! compiler quite well. His bugs are decades old now.

 

Not for a reason it is about 2x slower than the ones I mentioned, this one-pass compiler practically does not do them.

 

I postulate to appreciate the work of modern and active A8 developers rather then write in Latin of A8 world like Action!.

 

Action! is a great and simple language but poor compiler - best of his past times of course.

 

--- EDIT ---

 

of course, every bug I found was reported and fixed. in the case of Mad Pascal it was done in no time.

Edited December 7, 2021 by zbyti

[amarok]

On the main page of Rosetta Code you can read:

Quote

Rosetta Code is a programming chrestomathy site. The idea is to present solutions to the same task in as many different languages as possible, to demonstrate how languages are similar and different, and to aid a person with a grounding in one approach to a problem in learning another.

 

For me, Rosetta Code is also a kind of time capsule for programming languages similar to Rosetta Stone for natural languages. Assuming that the website will somehow survive it might be a great source of information about the beginning of software development for future researchers of ancient programming languages.

 

I know limitations of Action! but it doesn't matter for me in terms of the main idea of Rosetta Code initiative. I like your discussion about the Action! itself but one may ask what is the point of creating software for 40 years old computers at all?

 

Concerning modern languages - I like MadPascal a lot. For me, it is the best compiler for 8-bit Atari computer. Honestly speaking, MadPascal encouraged me to come back to make programs for Atari. Please keep in mind that gr9Lab and SortViz have been implemented in MadPascal already.

 

If you don't mind, I would suggest to stick here to the main topic which is "Action! on Rosetta Code".

[zbyti]

I'm going to use your Rosetta Code for my own development and I would like to have binary not to have to do it myself, but if it's too much work for you, I will do it myself.

[Alfred]

The Rosetta site seems broken. For example when looking at unimplemented tasks, Action! is not shown on the list. There’s another page that I found that also doesn’t display Action! as a language. So it’s difficult to see what is missing.

[Gury]

22 hours ago, ilmenit said:

Can be a great test set for Effectus

True, such a huge list of examples will surely help to improve Effectus. I just have to find some more time.

 

[zbyti]

Effectus will inherit all Mad Pascal's bugs, so it's worth stabilizing Mad Pascal first  

[Gury]

I will not blame Mad Pascal and Tebe for syntax errors in Effectus

 

But..., I will include most recent version of MP in future version.

 

Edited December 8, 2021 by Gury

[amarok]

13 hours ago, Alfred said:

The Rosetta site seems broken. For example when looking at unimplemented tasks, Action! is not shown on the list. There’s another page that I found that also doesn’t display Action! as a language. So it’s difficult to see what is missing.

I've just created page tasks not implemented in Action!. Could you please tell me what else is missing?

[Alfred]

11 hours ago, amarok said:

I've just created page tasks not implemented in Action!. Could you please tell me what else is missing?

If I find the other page missing Action! I'll let you know. The Rosetta tasks seem all over the place from stupidly trivial (Function - Define a function called multiply) to non-trivial (MD5 hash). Many of the uploaded examples seem to me worthless, such as the MD5 hash task. Virtually all of the language examples merely invoke an MD5 library (result := MD5("mystring","hash")) which to me is worthless. I want to see the MD5 function implementation itself, not just the one line invocation. You learn nothing from these one line solutions invoking black box library functions.

 

Many of the tasks are unsuited for the Atari. Things like Fork or Perform a DNS query have no solution since there is neither a multi-tasking OS (Hunt's MTOS isn't really) nor a native TCP/IP implementation. Still, I admire your perseverance.

[Kaj de Vos]

I think FujiNet should be able to do DNS.

[_The Doctor__]

15 hours ago, Alfred said:

If I find the other page missing Action! I'll let you know. The Rosetta tasks seem all over the place from stupidly trivial (Function - Define a function called multiply) to non-trivial (MD5 hash). Many of the uploaded examples seem to me worthless, such as the MD5 hash task. Virtually all of the language examples merely invoke an MD5 library (result := MD5("mystring","hash")) which to me is worthless. I want to see the MD5 function implementation itself, not just the one line invocation. You learn nothing from these one line solutions invoking black box library functions.

 

Many of the tasks are unsuited for the Atari. Things like Fork or Perform a DNS query have no solution since there is neither a multi-tasking OS (Hunt's MTOS isn't really) nor a native TCP/IP implementation. Still, I admire your perseverance.

um cc65 dragon cart.... we have some of this... could be implemented better for sure but that wasn't the point of that exercise... though someone could run with it and make it so...