meta Meta language released

[Kaj de Vos]

Those are advanced features. I use them only to accomplish specific goals for Meta. I will certainly not offer all those options for Meta. Usually, they can be much simplified when you take a step back and consider what it is that you want to accomplish.

[ilmenit]

25 minutes ago, Kaj de Vos said:

Those are advanced features. I use them only to accomplish specific goals for Meta. I will certainly not offer all those options for Meta. Usually, they can be much simplified when you take a step back and consider what it is that you want to accomplish.

I agree, it's nothing that needs to be done early in the language development. I listed the cases just a reminded what everything may be needed to cover in the future - may be useful when designing the low-level part of the language addressing the 8bit Atari needs. Often these needs cannot be simplified because of how the old hardware or OSes are designed.

[ilmenit]

@Kaj de Vos - do you have already arrays/series implemented in Meta? If not, how do you plan to approach this topic? Are they going to be dynamic/expandable like in REBOL or fixed size? From REBOL documentation:

REBOL uses the concept of expandable SERIES for holding and manipulating data, rather than the concept of fixed size arrays. For example, in REBOL you would normally write:

    block: copy []
    repeat n 5 [append block n]

rather than:

    block: array 5
    repeat n 5 [poke block n n]

In other words, REBOL does not require you to specify the size of data arrays (blocks, bytes, or strings) in advance. They are dynamic.

I have tried to compile following - are they already implemented?

    flags: array size - Script error: "array" has no value
    flags: copy [] - Script error: "copy" does not work on parameter type: expressions list block!

[Kaj de Vos]

13 minutes ago, ilmenit said:

Script error: "WORD" has no value

means that WORD is undefined. So, not implemented.

14 minutes ago, ilmenit said:

Script error: "METHOD" does not work on parameter type

means that METHOD is defined, but on other parameter types than you gave it.

16 minutes ago, ilmenit said:

do you have already arrays/series implemented in Meta? If not, how do you plan to approach this topic? Are they going to be dynamic/expandable like in REBOL or fixed size?

Both, but expandable is a dynamic feature, so that will come later:

https://language.metaproject.frl/#state

The current support is quite close to REBOL, but static, and only byte arrays and simple strings so far:

array= make static binary! 10

 

[Kaj de Vos]

Not sure if that is on the build server yet.

[ilmenit]

17 minutes ago, Kaj de Vos said:

The current support is quite close to REBOL, but static, and only byte arrays and simple strings so far:

array= make static binary! 10

 

Thanks, how to make an memset of the array? Neither change nor head, forall seem to be implemented yet (no value) - is there some implemented way of setting an array value? Trying something like this:

    sizepl: 8192
    flags:  make static binary! sizepl
    for i sizepl
    [
        change at flags i 1
    ]
    flags: head flags

EDIT: ok, poke seems to work, will return with more testing

Edited November 16, 2021 by ilmenit

[Kaj de Vos]

Meta has let-word!s for constants in addition to set-word!s, so to define the array as static, you need:

sizepl=

flags=

 

change at is very inefficient at that, that would be poke.

 

head is at-head in Meta.

[ilmenit]

2 minutes ago, Kaj de Vos said:

Meta has let-word!s for constants in addition to set-word!s, so to define the array as static, you need:

sizepl=

flags=

 

change at is very inefficient at that, that would be poke.

 

head is at-head in Meta.

thanks, what about pick? Not implemented yet? I also tried at-pick and pick-at.

[Kaj de Vos]

What does the compiler tell you?

[ilmenit]

3 minutes ago, Kaj de Vos said:

What does the compiler tell you?

Script error: "pick" does not work on types combination

Here is the code I'm trying to compile:

iter: 1
sizepl: 8192 ; becase with "sizepl= 8192" this "for i sizepl" does not compile
size= 8190

while [ iter <= 10 ] 
[
    count: 0
	i: 0
	flags= make static binary! sizepl
	for i sizepl
    [
        poke flags i 1
    ]
    flags: at-head flags

    while [ i <= size ]
    [
        p: pick flags i
        if p = 1 
        [
		    prime: (i*2) + 3
            k: i + prime
            while [ k <= size ] 
            [
                poke flags k 0
                k: k + prime
            ]
			count: count + 1
        ]
		i: i + 1
    ]
    iter: iter + 1
]

 

[Kaj de Vos]

1 minute ago, ilmenit said:

"pick" does not work on types combination

So pick is implemented, but you left out what type combination it doesn't support.

[Kaj de Vos]

42 minutes ago, ilmenit said:

forall

is for-all in Meta.

[ilmenit]

3 hours ago, Kaj de Vos said:

So pick is implemented, but you left out what type combination it doesn't support.

it doesn't work neither for flags= make static binary! sizepl nor flags: make static binary! sizepl

** Script error: "pick" does not work on types combination:
Parameter types:
1: static-binary! [flags i
    if p = 1
...
2: integer! [i
    if p = 1
    [
...
At: [pick flags i
    if p = 1
    [
        prime: (i*2) + 3...
In: [
    p: pick flags i
    if p = 1
    [
        prime: (...
Near: [
    while [i <= size]
    [
        p: pick flags i
   ...

also when I set:

sizepl= 8192

then

** Script error: "for" does not work on types combination:
Parameter types:
1: integer! reference [i sizepl
    [
      ...
2: natural16! [sizepl
    [
        ...
3: literal block! [
    [
        poke f...
At: [
    for i sizepl
    [
        poke flags i 1
    ]
   ...
In: [
    count: 0
    i: 0
    flags= make static binary! si...
Near: [
    while [iter <= 10]
    [
        count: 0
        i...

 

[Kaj de Vos]

I deployed a new compiler to the server. Should be implemented now.

[ilmenit]

4 minutes ago, Kaj de Vos said:

I deployed a new compiler to the server. Should be implemented now.

Thanks, I'm getting closer to the end of compilation issues. Now for the code:

iter: 1
sizepl= 8192
size= 8190

while [ iter <= 10 ] 
[
    count: 0
    i: 0
    flags= make static binary! sizepl
    for i sizepl
    [
        poke flags i 1
    ]
    flags: at-head flags

    i: 0
    while [ i <= size ]
    [
        p: pick flags i
        if p = 1 
        [
            prime: ( i * 2 ) + 3
            k: i + prime
            while [ k <= size ] 
            [
                poke flags k 0
                k: k + prime
            ]
            count: count + 1
        ]
        i: i + 1
    ]
    iter: iter + 1
]

I receive the following output and error:

code: [none while
    [/block! compile/block
        [logic! is-lesser-or-equal
            [integer! compile/value [global! k]]
            [natural16! compile/value [global! size]]
        ]
    ]
    [/block! compile/block
        [byte! poke
            [/static-binary! compile/reference [global! flags]]
            [integer! compile/value [global! k]]
            [byte! 0]
        ]
        [integer16! set
            [:integer! compile/reference [global! k]]
            [integer16! add
                [integer! compile/value [global! k]]
                [integer! compile/value [global! prime]]
            ]
        ]
    ]
]
** Script error: "while" does not work on types combination:
Parameter types:
1: expressions list block! [[k <= size]
    [
   ...
2: expressions list block! [
    [
        poke f...
At: [
    while [k <= size]
    [
        poke flags k 0
    ...
In: [
    prime: (i * 2) + 3
    k: i + prime
    while [k <=...
Near: [
    if p = 1
    [
        prime: (i * 2) + 3
        k...

 

[Kaj de Vos]

Oops, debug code was enabled. Fixed now.

 

while does not require the condition to be in a block. I changed that back to what it was in Logo, too.

[ilmenit]

Ok, removed the block from the while but now again I'm getting the error in for:

 

iter: 1
sizepl= 8192
size= 8190

while iter <= 10 
[
    count: 0
    i: 0
    flags= make static binary! sizepl
    for i sizepl
    [
        poke flags i 1
    ]
    flags: at-head flags

    i: 0
    while i <= size
    [
        p: pick flags i
        if p = 1 
        [
            prime: ( i * 2 ) + 3
            k: prime + i 
            while k <= size
            [
                poke flags k 0
                k: k + prime
            ]
            count: count + 1
        ]
        i: i + 1
    ]
    iter: iter + 1
]

error:

** Script error: method "for" does not work on types combination:
Return value not used.
Parameter types:
1: integer! reference [i sizepl
    [
      ...
2: natural16! [sizepl
    [
        ...
3: statements sequence block! [
    [
        poke f...
At: [
    for i sizepl
    [
        poke flags i 1
    ]
   ...
In: [
    count: 0
    i: 0
    flags= make static binary! si...
Near: [
    while iter <= 10
    [
        count: 0
        i: ...

[Kaj de Vos]

One step closer: on the back-end instead of the front-end. Updated again.

[ilmenit]

19 minutes ago, Kaj de Vos said:

One step closer: on the back-end instead of the front-end. Updated again.

thanks, now there is another error in the same code:

** Script error: no known implicit method "is-equal" with 2 parameters.
At: [= 1
    [
        prime: (i * 2) + 3
        k: prime + ...
In: [p = 1
    [
        prime: (i * 2) + 3
        k: prime ...
Near: [
    if p = 1
    [
        prime: (i * 2) + 3
        k...

[Kaj de Vos]

Ah, changed that to EQUALS. You can use that as workaround (prefix). Laptop is empty now, more fixes tomorrow.

[ilmenit]

1 minute ago, Kaj de Vos said:

Ah, changed that to EQUALS. You can use that as workaround (prefix). Laptop is empty now, more fixes tomorrow.

If I understand properly what do you mean by (prefix) I got the following error:

 

** Script error: no known implicit method "is-equal" with 2 parameters.
At: [EQUALS p 1
    [
        prime: (i * 2) + 3
        k: p...
In: [
    if EQUALS p 1
    [
        prime: (i * 2) + 3
    ...
Near: [
    p: pick flags i
    if EQUALS p 1
    [
        pri...

 

[Kaj de Vos]

Hm, so I messed that up, too. Moment...

[Kaj de Vos]

Last fix for today.

[ilmenit]

1 hour ago, Kaj de Vos said:

Last fix for today.

as always, do not commit just before leaving the computer

Now it returns:

out.c:45: Error: Invalid left operand for binary operator '='
1 errors and 0 warnings generated.

 

[Kaj de Vos]

That does mean that you're now through the back-end and have started the next level. It may well be an unrelated bug.