cito-users

[CI] Ć now on GitHub

[Piotr F. <pi...@fu...>]

Hi,

cito is now available at https://github.com/pfusik/cito

It is a complete rewrite of cito version 0, which was just a prototype. There are many changes in the language, so please take a look at the language reference on GitHub. The language is not finished yet, but I expect many more additions than incompatible changes.

It's much easier now to add a target language because of the test suite (currently 200+ tests). Let me know which language you are interested in. Would you like to contribute this backend?

Do you have questions? Suggestions?

--Piotr

[CI] alive?

[Raoul D. <ra...@gm...>]

hi,

To what degree is CI alive? E.g. last release was in 2013.

thank you.

Re: [CI] How to: Type alias

[P. F. <pi...@fu...>]

Dnia 2015-01-19 13:46 Valen napisał(a):

>Im thinking about to use C pre processor (with cito source files), to
>define a macro for replacing a string
>'FIXED' with sting 'int'.
>
>Maybe there are some better options ?

Another option is sed. Worth trying unless you have string literals that 
contain "FIXED". :)

Piotr

Re: [CI] How to: Type alias

[Valen <val...@gm...>]

On Mon, 12 Jan 2015 12:48:14 +0100
"Piotr Fusik" <pi...@fu...> wrote:

> >In C:
> >typedef int fixed;  
> 
> It's not supported.

Hi, 

Im thinking about to use C pre processor (with cito source files), to
define a macro for replacing a string
'FIXED' with sting 'int'.

Maybe there are some better options ?

---
Valen

Re: [CI] How to: Type alias

[Valen <val...@gm...>]

On Mon, 12 Jan 2015 12:48:14 +0100
"Piotr Fusik" <pi...@fu...> wrote:

> You can only use a macro to define local variables:

Thanks!


---
Valen

Re: [CI] How to: Type alias

[P. F. <pi...@fu...>]

Hello,

>Im want to write 'fixed' instead of 'int', for some class fields.
>Is there a away to do so ?
>
>In C:
>typedef int fixed;

It's not supported.

You can only use a macro to define local variables:

public class Test
{
	macro fixed(x) { int x; }
	public static int Foo()
	{
		fixed(x);
		fixed(y);
		x = 1;
		y = 2;
		return x + y;
	}
}

Piotr

[CI] How to: Type alias

[Valen <val...@gm...>]

Hi,

Im want to write 'fixed' instead of 'int', for some class fields.
Is there a away to do so ?

In C:
typedef int fixed;

Thanks,
Valen

Re: [CI] Only one instance of class

[Valen <val...@gm...>]

On Fri, 9 Jan 2015 20:21:40 +0100
"Piotr Fusik" <pi...@fu...> wrote:

I see.
Thanks!

> There are several problems with singletons. "Modern C++ Design" has a
> whole chapter about singletons and yet it doesn't cover new
> approaches such as Dependency Injection.
> 
> If you really need a singleton, you can implement it in native code,
> e.g.


---
Valen

Re: [CI] Only one instance of class

[Piotr F. <pi...@fu...>]

Hello,

There are several problems with singletons. "Modern C++ Design" has a whole 
chapter about singletons and yet it doesn't cover new approaches such as 
Dependency Injection.

If you really need a singleton, you can implement it in native code, e.g.

static Foo GetInstance()
{
#if C
    native {
        ...
    }
#elif JAVA
    native {
        ...
    }
...
#endif
}

Hope that helps,
Piotr

-----Oryginalna wiadomość----- 
From: Valen
Sent: Friday, January 09, 2015 1:46 PM
To: cit...@li...
Subject: [CI] Only one instance of class

Hi,

I want to create only one instance of my class
and i want methods of this instance, will be accessible from any line of
my code (without need to pass 'this' pointer).


So, the static methods are just fine for this.
But i cant share vars (fields) between this methods, because i cannot
declare 'static' for vars (fields).

How can i share vars, between some static methods ?

Thanks,
Valen

------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net
_______________________________________________
cito-users mailing list
cit...@li...
https://lists.sourceforge.net/lists/listinfo/cito-users 

[CI] Only one instance of class

[Valen <val...@gm...>]

Hi,

I want to create only one instance of my class
and i want methods of this instance, will be accessible from any line of
my code (without need to pass 'this' pointer).


So, the static methods are just fine for this.
But i cant share vars (fields) between this methods, because i cannot
declare 'static' for vars (fields).

How can i share vars, between some static methods ?

Thanks,
Valen

Re: [CI] 1 Year since last update

[Mihai D. <mih...@gm...>]

How would one go about using floats? How about int16? virtual methods?


On Sun, May 18, 2014 at 7:44 PM, Piotr Fusik <pi...@fu...> wrote:

> Hello,
>
> You mean the release?
>
> It doesn't mean that this project is dead.
> The current version works very well within its limits. Some minor bugs have
> been fixed in Git.
>
> And there's too early for the next, more powerful version.
>
> Piotr
>
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform
> available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> cito-users mailing list
> cit...@li...
> https://lists.sourceforge.net/lists/listinfo/cito-users
>

Re: [CI] 1 Year since last update

[P. F. <pi...@fu...>]

Hello,

You mean the release?

It doesn't mean that this project is dead.
The current version works very well within its limits. Some minor bugs have 
been fixed in Git.

And there's too early for the next, more powerful version.

Piotr

[CI] 1 Year since last update

[Mihai D. <mih...@gm...>]

Happy anniversary?

Re: [CI] Is there a way to call a virtual func of base calss, from the overrided virtual func of a derived class ?

[Valen <val...@gm...>]

On Thu, 06 Feb 2014 16:06:57 +0100
"Piotr Fusik" <pi...@fu...> wrote:

> Temporary workaround:
The workaround seems reasonable,
thanks !

---
Valen

Re: [CI] Is there a way to call a virtual func of base calss, from the overrided virtual func of a derived class ?

[P. F. <pi...@fu...>]

Hello,

Unfortunately not yet, but it's on my todo-list. The syntax will be C#-ish 
(base.foo()).

Temporary workaround:

public class A
{
    void fooA()
    {
        // do stuff
    }

    public virtual void foo()
    {
        fooA();
    }
}

public class B : A
{
    public override void foo()
    {
        fooA();
        // do stuff
    }
}

Piotr

Dnia 2014-02-06 15:12 Valen napisał(a):

>Hi!
>
>Another question...
>
>Is there a way to call a virtual func of base calss,
>from the overrided virtual func of a derived class ?
>
>
>public class A
>{
>    public virtual void foo()
>    {
>        
>    }
>}
>
>
>public class B : A
>{
>    public override void foo()
>    {
>      // how to call A::foo()  ?
>      // from here
>    }
>}
>
>
>
>
>Thanks,
>Valen
>
>---------------------------------------------------------------------------
---
>Managing the Performance of Cloud-Based Applications
>Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
>Read the Whitepaper.
>http://pubads.g.doubleclick.net/gampad/clk?id=121051231&iu=/4140/ostg.clktrk
>_______________________________________________
>cito-users mailing list
>cit...@li...
>https://lists.sourceforge.net/lists/listinfo/cito-users
>

Re: [CI] Is there a way to call a constructor of base class, from the constructor of a derived class ?

[Valen <val...@gm...>]

On Thu, 06 Feb 2014 15:03:26 +0100
"Piotr Fusik" <pi...@fu...> wrote:


Hi, Piotr!

> The base class constructor is called automatically during derived
> class construction.
A-ha !
I see.


Thanks for quick reply.


---
Valen

[CI] Is there a way to call a virtual func of base calss, from the overrided virtual func of a derived class ?

[Valen <val...@gm...>]

Hi!

Another question...

Is there a way to call a virtual func of base calss,
from the overrided virtual func of a derived class ?


public class A
{
    public virtual void foo()
    {
        
    }
}


public class B : A
{
    public override void foo()
    {
      // how to call A::foo()  ?
      // from here
    }
}




Thanks,
Valen

Re: [CI] Is there a way to call a constructor of base class, from the constructor of a derived class ?

[P. F. <pi...@fu...>]

Hello,

The base class constructor is called automatically during derived class 
construction.
It is not possible to call it again.

Piotr

Dnia 2014-02-06 14:47 Valen napisał(a):

>Hi!
>
>Is there a way to call a constructor of base class,
>from the constructor of a derived class ?
>
>
>public class A
>{
>    /// Constructor
>    A()
>    {
>      // vars init for class A
>    }
>}
>
>
>public class B : A
>{
>
>    /// Constructor
>    B()
>    {
>      // how to call A::A()  ?
>      // here
>    }
>}
>
>
>Thanks,
>Valen
>
>---------------------------------------------------------------------------
---
>Managing the Performance of Cloud-Based Applications
>Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
>Read the Whitepaper.
>http://pubads.g.doubleclick.net/gampad/clk?id=121051231&iu=/4140/ostg.clktrk
>_______________________________________________
>cito-users mailing list
>cit...@li...
>https://lists.sourceforge.net/lists/listinfo/cito-users
>

[CI] Is there a way to call a constructor of base class, from the constructor of a derived class ?

[Valen <val...@gm...>]

Hi!

Is there a way to call a constructor of base class,
from the constructor of a derived class ?


public class A
{
    /// Constructor
    A()
    {
      // vars init for class A
    }
}


public class B : A
{

    /// Constructor
    B()
    {
      // how to call A::A()  ?
      // here
    }
}


Thanks,
Valen

Re: [CI] Any chance to add 16 bit math ?

[Valen <val...@gm...>]

On Thu, 12 Sep 2013 23:48:20 +0200
Piotr Fusik <pi...@fu...> wrote:

> Yes, short type will be added at some point,
Good news :)


> However, I don't understand how it is going to help your Z80 project.
I just want to use both 32-bit and 16-bit math in my cito project.


> I suppose your C compiler has 16-bit ints, and if that's the case,
> just use int, keeping in mind its limits.
This is an option, but this will limit all my math to 16-bit.

> If your C compiler uses
> 32-bit int on Z80, it's going to be slow even if you use short
> variables all over the place. 


If cito will translate a new 16-bit "short" type to "C short",
then i can use _both_ types in cito files. (32-bit and/or 16-bit.)



---
Valen

Re: [CI] Any chance to add 16 bit math ?

[Jakob E. <ja...@au...>]

Yes, this is the way to go. However, the cito specification should maybe say that its int can be between 16 - 32 bits? 

Piotr Fusik <pi...@fu...> skrev:

>Hello,
>
>Yes, short type will be added at some point, as well as float, double and maybe long (with limited support for JavaScript, as it has no standard 64-bit integers).
>
>However, I don't understand how it is going to help your Z80 project. I suppose your C compiler has 16-bit ints, and if that's the case, just use int, keeping in mind its limits. If your C compiler uses 32-bit int on Z80, it's going to be slow even if you use short variables all over the place. That's because standard C performs integer arithmetic after promoting operands to int.
>
>Piotr
>
>Użytkownik Valen <val...@gm...> napisał:
>
>>Hi,
>>I'm use cito to generate C code for z80 (small 8-bit CPU).
>>
>>Is it possible to add 16 bit math ("short" C type) to cito ?
>>For this CPU, the 16 bit math is  much faster, than 32 bit math.
>>
>>---
>>Thanks,
>>Valen
>>
>>------------------------------------------------------------------------------
>>How ServiceNow helps IT people transform IT departments:
>>1. Consolidate legacy IT systems to a single system of record for IT
>>2. Standardize and globalize service processes across IT
>>3. Implement zero-touch automation to replace manual, redundant tasks
>>http://pubads.g.doubleclick.net/gampad/clk?id=51271111&iu=/4140/ostg.clktrk
>>_______________________________________________
>>cito-users mailing list
>>cit...@li...
>>https://lists.sourceforge.net/lists/listinfo/cito-users
>------------------------------------------------------------------------------
>How ServiceNow helps IT people transform IT departments:
>1. Consolidate legacy IT systems to a single system of record for IT
>2. Standardize and globalize service processes across IT
>3. Implement zero-touch automation to replace manual, redundant tasks
>http://pubads.g.doubleclick.net/gampad/clk?id=51271111&iu=/4140/ostg.clktrk
>_______________________________________________
>cito-users mailing list
>cit...@li...
>https://lists.sourceforge.net/lists/listinfo/cito-users

Re: [CI] Any chance to add 16 bit math ?

[Piotr F. <pi...@fu...>]

Hello,

Yes, short type will be added at some point, as well as float, double and maybe long (with limited support for JavaScript, as it has no standard 64-bit integers).

However, I don't understand how it is going to help your Z80 project. I suppose your C compiler has 16-bit ints, and if that's the case, just use int, keeping in mind its limits. If your C compiler uses 32-bit int on Z80, it's going to be slow even if you use short variables all over the place. That's because standard C performs integer arithmetic after promoting operands to int.

Piotr

Użytkownik Valen <val...@gm...> napisał:

>Hi,
>I'm use cito to generate C code for z80 (small 8-bit CPU).
>
>Is it possible to add 16 bit math ("short" C type) to cito ?
>For this CPU, the 16 bit math is  much faster, than 32 bit math.
>
>---
>Thanks,
>Valen
>
>------------------------------------------------------------------------------
>How ServiceNow helps IT people transform IT departments:
>1. Consolidate legacy IT systems to a single system of record for IT
>2. Standardize and globalize service processes across IT
>3. Implement zero-touch automation to replace manual, redundant tasks
>http://pubads.g.doubleclick.net/gampad/clk?id=51271111&iu=/4140/ostg.clktrk
>_______________________________________________
>cito-users mailing list
>cit...@li...
>https://lists.sourceforge.net/lists/listinfo/cito-users

[CI] Any chance to add 16 bit math ?

[Valen <val...@gm...>]

Hi,
I'm use cito to generate C code for z80 (small 8-bit CPU).

Is it possible to add 16 bit math ("short" C type) to cito ?
For this CPU, the 16 bit math is  much faster, than 32 bit math.

---
Thanks,
Valen

Re: [CI] Cito support for pascal and PHP

[Enrico <eir...@gm...>]

Dear,

> How does your Pascal target relate to Delphi?
Pascal back-end generates object pascal code with the intention to be
compatible with Delphi and FreePascal/Lazarus. Delphi is a little more
restrictive respect to FreePascal and I've trouble to run the test as I
have only licence of some versions (the latest that I've is XE2 where the
newest is XE4).

> I will gladly apply patches that provide PHP and Pascal targets.
The code that I  write up to now is here:
http://sourceforge.net/p/eirocasoftware/code/230/tree/trunk/CSharp/Vario/Cito/
The integration is very easy, just make virtual all the methods in
SourceGenerator.cs needed to compile.
I'm still testing them, as:

> You can also try this small project:
https://github.com/pfusik/datamatrix-ci
As evidenced two "know" problems, the for(;;) and the ternary operator ?:
are a nightmare if are not supported by target language (and in pascal
aren't). I'm working on patching, but a stable long term solution must
involve some changes in Cito architecture (support a sort of two pass
generation in order to be able to make some operation currently unfeasible
- e.g. adding a new variable or a temporary function -).

Let me know how to proceed to keep synced the changes (maybe a git fork is
still a valid option).

>> in order to avoid changes in AST data structures;
>Do you mean a layer that implements OOP in non-OOP languages?
As mentioned also some operators have the need to add some emulated logic,
declaring a priori the variable needs logic too, ... my idea is to keep
CiXXXXX classes related to AST clean form attribute valid only for some
language and move them in the back-end generator.

> Variables that only differ by case should be probably disallowed.
I think is preferred to rename them on target language that does not
support case sensitive (like was done with global constants), the question
is that is "easy" to do in the parsing (where you have the variable scope)
but tricky to do a posteriori.

Ciao,
Enrico



2013/8/3 Piotr Fusik <pi...@fu...>

> Hello,
>
> I am interested in new Cito backends. I was planning to code myself
> backends
> for Perl 6, Python and Ruby.
>
> I will gladly apply patches that provide PHP and Pascal targets. Someone
> asked me for the Delphi backend. How does your Pascal target relate to
> Delphi?
>
> > all the Cito source that I was able to find (hello.ci, FAIL, ASAP)
>
> You can also try this small project:
> https://github.com/pfusik/datamatrix-ci
>
> > - Multi step conversion in order to support declarative languages (C,
> Pascal)
> > in order to avoid changes in AST data structures;
>
> Do you mean a layer that implements OOP in non-OOP languages?
>
> > - support case insensitivity/reserved words;
>
> Renaming identifiers that are reserved in some languages should be done
> generally. We already had this problem with "module" used in the ASAP
> project while being a keyword in D.
> Variables that only differ by case should be probably disallowed.
>
> > - introduce helpers like variable scope tracking, statement stack, ...
> > (e.g. useful to guess the meaning of a break);
>
> Sounds good.
>
> > I have also a question about the roadmap, in particular there is a plan
> to
> support matrix and double?
>
> First I want to simplify extending the library. Then I'll work on floating
> point types.
>
> Piotr
>
> ------------------------------------------------------------------------------
> Get your SQL database under version control now!
> Version control is standard for application code, but databases havent
> caught up. So what steps can you take to put your SQL databases under
> version control? Why should you start doing it? Read more to find out.
> http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
> _______________________________________________
> cito-users mailing list
> cit...@li...
> https://lists.sourceforge.net/lists/listinfo/cito-users
>

Re: [CI] Cito support for pascal and PHP

[P. F. <pi...@fu...>]

Hello,

I am interested in new Cito backends. I was planning to code myself backends 
for Perl 6, Python and Ruby.

I will gladly apply patches that provide PHP and Pascal targets. Someone 
asked me for the Delphi backend. How does your Pascal target relate to 
Delphi?

> all the Cito source that I was able to find (hello.ci, FAIL, ASAP)

You can also try this small project:
https://github.com/pfusik/datamatrix-ci

> - Multi step conversion in order to support declarative languages (C, 
Pascal)
> in order to avoid changes in AST data structures;

Do you mean a layer that implements OOP in non-OOP languages?

> - support case insensitivity/reserved words;

Renaming identifiers that are reserved in some languages should be done 
generally. We already had this problem with "module" used in the ASAP 
project while being a keyword in D.
Variables that only differ by case should be probably disallowed.

> - introduce helpers like variable scope tracking, statement stack, ...
> (e.g. useful to guess the meaning of a break);

Sounds good.

> I have also a question about the roadmap, in particular there is a plan to 
support matrix and double?

First I want to simplify extending the library. Then I'll work on floating 
point types.

Piotr