LightFusion™

I don't think it's a good idea to write the pages in XHTML. I especially don't think that it's a good idea to write them in Transitional. Transitional is, as the name says, for transitioning - for slowly converting legacy pages to modern, lean markup. As such, it should be avoided for new pages - they should be modern and lean from the start. Strict is the way to go.

As for why HTML over XHTML:
http://www.hixie.ch/advocacy/xhtml

In short: Internet Explorer does not support XHTML. Period, no discussion. Oh, and IE7 won't either. So if you want proper code for IE, you have to use HTML, because it simply will pop up a download window for application/xhtml+xml (go here to see the effect; these pages are true XHTML).
You could send the code as text/html to IE (but the SourceForge servers don't let you decide which browser gets what, unless you script it yourself), which means that the browser interprets it as quite horribly broken HTML. Which causes problems.
So go with HTML 4.01 Strict. It's by far the best choice for current interoperable websites.

Quote:

---

Originally Posted by CornedBee

In short: Internet Explorer does not support XHTML. Period, no discussion.

---

That's the problem I had with the last XHTML site I made. It wasn't the server. I designed it and viewed it in Firefox and everything was great. Then I opened it in IE and all I got was code :ehh:.

Grr.. Why does Micosoft want to run everything if they can't do anything right? :mad:

Quote:

---

Originally Posted by chemicalNova

I think what you are talking about isn't a better select case but rather Structured Error Handling. IE, a Try, Catch, Finally type structure.

chem

---

I think both are mentioned in the .doc.

Quote:

---

Originally Posted by Datacide

That's the problem I had with the last XHTML site I made. It wasn't the server. I designed it and viewed it in Firefox and everything was great. Then I opened it in IE and all I got was code :ehh:.

Grr.. Why does Micosoft want to run everything if they can't do anything right? :mad:

---

That's why you don't use Internet Explorer ;)

Datacide, I changed your permissions to admin on SourceForge. I trust that you won't abuse this. Do you know how long you will need admin to use SQL stuff? Or will it be indefinite? Let me know if the changes I made worked or not.

Jake, do you know how long it will be until your computer is done?

My dad and I just got into a big arguement, and he's now saying the computer is his. I hope for this to be a temporary issue. :rolleyes:

...and it's still not done :cry:

No worries Jake. I'm sure it will turn out alright.




I modified the Ground Rules list and removed a few. I also rearranged the categories. If anyone sees one missing that they feel should be included PLEASE feel free to comment on it.

Aims:
• LightFusion will be easy to learn and will produce standard windows applications without lacking in speed of execution.
• LightFusion will make programming enjoyable for anyone who uses it.
• LightFusion will combine the popular features of languages such as Visual Basic and C++ and add features that seem to be lacking from those languages.

General Rules:
1. Always let the programmer know what the compiler is doing, (through the syntax, documentation, features, etc.) as long as doing so does not make programming more difficult.
2. Don’t force the programmer.
3. Support composition of software from separately developed parts.

Syntax Rules:
4. Support program organization and readability.
5. Use features/syntax from other languages except when a clear improvement can be made by doing otherwise.
6. When in doubt, pick a solution that is easiest to teach and that best explains what the compiler is doing.





Also, does anyone have any input on the format for the LightFusion Project fies (*.lfp)? I have 2 ideas in mind:

1) Use INI file format (even though it would be *.lfp). That would make it very readable and since there are plenty of classes/DLLs/whatevers out there to support INI stuff already, it would be easy for other people to understand the format and to write programs/add-ins that access out project file (the top part of the *.lfs might be done this way also for the same reasons).

2) The other option would involve creating our own format which may or may not be easy to read (when viewed in a text editor) depending on how we do. If we do it this way then I would like some input on how we should do it. There are many ways of going about it, but no matter what I am going to have to write a bunch of parsing stuff and error handling (incase the user messes up the file format).


I think INI would be easier and more readable, but I don't know if this format has any downsides (considering it isn't used much any more). I'd like everyones input.

XML not an option? It lends itself very well to things such as projects, unlike INI, which has several shortcomings in this regard. (Mostly hierarchical organization of project components.)

Do you know where I can find a good class for manipulating XML? Its a good idea and I hadn't thought of it.

I don't think there will be that much hierarchical stuff involved. It will just be a list and of files and properties of those files (such as last window size). All other information will be stored in the file itself.

Thanks for the suggestion. I will look around for a good XML drop-in class.

You don't need a "drop-in" class. There are couple of different XML libraries with Visual Basic. Check References.

chem

I realized that after I posted it. The only XML libraries I found were the MS ones though. They are pretty huge and I think it would be nicer if I could have a simplified class that accessed them, but maybe I will just look around for a tutorial on them.

An XML parser wouldn't be hard. You were writing a HTML parsing class a while back if I'm not mistaken? Its no different, just your own custom tags..

chem

Yeah, but my HTML parser was really slow and I would rather spend my time on more important things. I guess I could use the MS XML DLL, I just have no idea how to use it.

Okay, I started working on an XML format, but it is insanely hard to use. I made a thread about it, but I'm not sure if I will get what I am looking for (an easy to use class instead of the cumbersome msxml.dll). If I can't find anything easier then I might get someone else to do that part while I work on the rest of the IDE. I'm not sure.

Okay, I want to shift the focus back to syntax as many people have suggested. I want everyone (who wants to) to post a small project (small enough that we can post in in [vbcode] tag's and not in a .zip) and we will attempt to convert it to LightFusion.

This will help us make decisions about syntax and as issues come up we can resolve them. I have been looking around for a small project that would be good to use, but I can't really find one. I will keep looking though.

Penagate, how are the logos coming?

Datacide, did me giving you admin permissions solve the SQL problems you were having?

Okay, here is one peice of code from back when (fairly sloppy) that I hope to get the ball rolling with:

VB Code:

Private Sub moSysTray_MouseMove()
    Static ssLastClipText   As String
    Dim sClipText       As String
    Dim sCounts         As String
    Dim lSpellErrCnt    As Long
    Dim lGramErrCnt     As Long
    Dim sShortClipText  As String
    
    lSpellErrCnt = -1
    lGramErrCnt = -1
    If Clipboard.GetFormat(vbCFText) Then
        sClipText = Clipboard.GetText(vbCFText)
        If ssLastClipText <> sClipText Then
            ssLastClipText = sClipText
            sCounts = SpellMe(sClipText, , True)
            If sCounts = "||ERROR||" Then Exit Sub
            If Len(sCounts) > 0 Then
                lSpellErrCnt = Left$(sCounts, InStr(1, sCounts, " "))
                lGramErrCnt = Right$(sCounts, InStr(1, sCounts, " ") - 1)
            End If
            If Len(sClipText) > 39 Then
                sShortClipText = Left$(sClipText, 36) & "..."
            Else
                sShortClipText = Left$(sClipText, 39)
            End If
            moSysTray.ToolTipText = """" & sShortClipText & """" & vbCrLf _
                    & "Spell:" & lSpellErrCnt & Chr$(149) & "Gram:" & lGramErrCnt
        End If
    Else
        moSysTray.ToolTipText = "Clipboard data is not text!"
    End If
End Sub

I'm not sure where we are on Dim-ing things, but here is my attempt at the LightFusion conversion:

Code:

---

Private Sub moSysTray_MouseMove() {
    String sLastClipText
    String sClipText
    String sCounts
    Long lSpellErrCnt
    Long lGramErrCnt
    String sShortClipText
   
    lSpellErrCnt = -1
    lGramErrCnt = -1
    If(Clipboard.GetFormat(lfCFText)) {
        sClipText = Clipboard.GetText(lfCFText)
        If(sLastClipText <> sClipText) {
            sLastClipText = sClipText
            sCounts = SpellMe(sClipText, , True)
            If(sCounts == "||ERROR||") {Exit Sub}
            If(Len(sCounts) >> 0) {
                lSpellErrCnt = Left(sCounts, InStr(1, sCounts, " "))
                lGramErrCnt = Right(sCounts, InStr(1, sCounts, " ") - 1)
            }
            If(Len(sClipText) >> 39) {
                sShortClipText = "{ Left(sClipText, 36) }..."
            } Else {
                sShortClipText = Left(sClipText, 39)
            }
            moSysTray.ToolTipText = "\"{ sShortClipText }\"{ lfCrLf }Spell:{ lSpellErrCnt }•Gram:{ lGramErrCnt }"
        }
    } Else {
        moSysTray.ToolTipText = "Clipboard data is not text!"
    }
}

---

That is just a start. Anyeone see anything they think could be done better or that I did wrong?

The way I did the string at the end was based off a suggestion I made a while back to use { and } mid string instead of & and joining strings together. Also, \ is used as the escape character to get a " to display.

PS - What is the "greater-than" operator? I thought >> was shift right, but I thought all C++ conditional operators had 2 characters...

I LOVE WHAT I DO!!! I GUESS THAT'S WHY I'M SO GOOD AT IT ;)

eyeRmonkey, check it out: lightfusion.sf.net. I sent you the password.

.....!!!!

*Jaw drop*

Nice dude... Very nice.

For everyone, that isn't me and just experienced what I did: Datacide managed to use PHP to setup a way for me to edit the news and other sections of the webpage manually. Its friggin' awesome.

Great work datacide. :thumb: Do you think the RSS feed will still be possible or did you make this because you couldn't get the RSS to work? No hurry, I was just curious what prompted you to add an AWESOME feature. Thanks a MILLION!

Quote:

---

Originally Posted by eyeRmonkey

Do you think the RSS feed will still be possible or did you make this because you couldn't get the RSS to work? No hurry, I was just curious what prompted you to add an AWESOME feature.

---

It just made sense. I mean, obviously the site is going to have lot's of updates So might as well have a way to do it easier and faster. As for the RSS, I haven't looked into it yet but it should still work with this. I mean, no point in having a RSS if you can't change the site, huh?

Greater than operator is > as in a > b.

I made some small changes. Sub -> Void, inline variable declaration, and removed Exit Sub (because I think it is very bad programming anyway)

Code:

---

Private Void moSysTray_MouseMove()
{
    String sShortClipText
   
    Long lSpellErrCnt = -1
    Long lGramErrCnt = -1
   
    If (Clipboard.GetFormat(lfCFText))
    {
        String sClipText = Clipboard.GetText(lfCFText)
       
        If(sLastClipText != sClipText)
        {
            String sLastClipText = sClipText
            String sCounts = SpellMe(sClipText, , True)
            If(sCounts != "||ERROR||") {
                If(Len(sCounts) > 0) {
                    lSpellErrCnt = Left(sCounts, InStr(1, sCounts, " "))
                    lGramErrCnt = Right(sCounts, InStr(1, sCounts, " ") - 1)
                }
                If(Len(sClipText) > 39) {
                    sShortClipText = Left(sClipText, 36) + "..."
                } Else {
                    sShortClipText = Left(sClipText, 39)
                }
                moSysTray.ToolTipText = "\" {sShortClipText} \" {nl}Spell: {lSpellErrCnt} • Gram: {lGramErrCnt}"
            }
        }
    } Else {
        moSysTray.ToolTipText = "Clipboard data is not text!"
    }
}

---

Also, I think the {} trick inside strings should be restricted to variable names only, not function calls like you did with Len. Otherwise it is unnecessarily hard to read.

Quote:

---

Originally Posted by Datacide

It just made sense. I mean, obviously the site is going to have lot's of updates So might as well have a way to do it easier and faster. As for the RSS, I haven't looked into it yet but it should still work with this. I mean, no point in having a RSS if you can't change the site, huh?

---

RSS is good because I can update things from one place and everything changes.

Quote:

---

Originally Posted by eyeRmonkey

I want everyone (who wants to) to post a small project (small enough that we can post in in [vbcode] tag's and not in a .zip) and we will attempt to convert it to LightFusion.

---

starting with the most basic of examples:

Code:

---

REM Basic
10 Dim s As String
20 s = "Hello World"
30 WHILE (TRUE)
40    PRINT s
50 WEND

// C
int main(void) {
    char *s = "Hello World";
    while (1) {
        printf("%s", s);
    }
}

-- haskell
main :: IO()
main =
    do
        (putStr . unlines) ["Hello", "World"]

// C#
string s = "Hello World";
while (1) {
    Console.WriteLine(s);
}

' VB.NET
Dim s As String = "Hello World"
While True
    Console.WriteLine(s)
Wend

---

moving up a bit:

Code:

---

#include <stdio.h>

int main(void)
{
        int x = 1, sum = 0, num = 0, tmp = 0;

        while(x <= 100)
        {
                tmp = 0;
                sum = 0;

                while(tmp <= x)
                {
                        sum = sum + tmp;
                        tmp++;
                }

                num = num + sum;
                printf("%i\n", num);
                x++;
        }

        return(0);
}

---

Code:

---

int isAlpha(char c) {
        if (c < 'A') {return(-1);}
        else if (c > 'z') {return(-1);}
        else if ((c > 'Z') && (c < 'a')) {return(-1);}
        else {return(0);}
}


char toLower(char c) {
        int alpha_offset = 'a' - 'A';
        if (isAlpha(c) == 0) {
                if ((c >= 'A') && (c <= 'Z')) {return(c + alpha_offset);}
        }
        return(c);
}

---

Quote:

---

Originally Posted by penagate

Greater than operator is > as in a > b.

I made some small changes. Sub -> Void, inline variable declaration, and removed Exit Sub (because I think it is very bad programming anyway)

. . . .

Also, I think the {} trick inside strings should be restricted to variable names only, not function calls like you did with Len. Otherwise it is unnecessarily hard to read.

---

Awesome work. Good point about { and }.

On a side note (although decision won't be final) what do you guys think about forcing the syntax into a readable format like VB does? For instance the IDE would force a space after { and before } wether it is used in string or as part of an If block or something.

I also think we should force text into the proper case (upper/lowwer). I depend on VB capitilzing things for me so I know wether I misspelled it or not.

What do you guys think about those 2 things?

Datacide, is there any way you could give me access to the whole CSS page (via the CMS)? That way I could add styles and change ones when needed. That would help a lot.

Also, for each of the sections (such as supporters) the CMS doesn't give me access to the title of that section, is that on purpose? If so, its not a big deal.

I'd like it if the language was case-insensitive so that I can type everything in lowercase, which I personally find more readable.

But then, sometimes it is good to have case-sensitivity, such as differentiating classes and objects (e.g. "Toolbar" class, "toolbar" object).

Lies! You're just lazy!

chem

Quote:

---

Originally Posted by penagate

type everything in lowercase, which I personally find more readable.

---

Just do. The C++ standard library has nothing but a few constants in upper case.

Quote:

---

Originally Posted by penagate

I'd like it if the language was case-insensitive so that I can type everything in lowercase, which I personally find more readable.

But then, sometimes it is good to have case-sensitivity, such as differentiating classes and objects (e.g. "Toolbar" class, "toolbar" object).

---

Sorry, I didn't mean it that way. The language should be case insensitive, BUT there should be an option in the IDE to turn on case sensitivity like VB. Where if you type something in all lower case it will captialize it properly for you.

I think that is a useful feature because if I type something in all lower case and it doesn't auto-capitalize in the right places then I know I mis-spelled it. Maybe there would be a better way to impliment a feature that has the same results. Like a syntax spell-checker or something.

eyeRmonkey, the more research I do, the more it seems pointless to have RSS. RSS is just for people to subscribe with an RSS Reader and view updates when it happens. If it's soooo hard to check out the site for updates we can just send out monthly newsletters.

I think you misunderstood what I meant. The only reason I wanted XML on our website was so that I could change the news VIA sourceforge.net and it would automatically update on the website. That way it would update in 3 places: The RSS feed people are subscribed to, the sourceforge page for LightFusion and lightfusion.sourceforge.net.

Its obviously not a huge deal if you can't get it done. I just thought it would be nice. If its too much trouble don't sweat it, but challanges are always a good thing also. Its up to you. You are already doing a ton of work for us our of pure generosity so there is no pressure to finish this minor detail.

Quote:

---

Originally Posted by Datacide

eyeRmonkey, the more research I do, the more it seems pointless to have RSS. RSS is just for people to subscribe with an RSS Reader and view updates when it happens. If it's soooo hard to check out the site for updates we can just send out monthly newsletters.

---

I'm afraid I have to call this opinion ignorant. My RSS reader currently monitors no less than 41 sites. Yes, it WOULD be a lot of trouble to check all these pages daily by hand.
Don't judge the technology before you've used it for a while.

Datacide, The RSS feed is already made for us so there is no reason NOT to use it. It is just a matter of if you can find a way to use it on the site. It will be used either way because it is done for us (with much thanks to SourceForge for that).

Quote:

---

Originally Posted by CornedBee

... My RSS reader currently monitors no less than 41 sites. ...

---

Talk about being plugged in. ;) You are certainly up on the latest technology CornedBee. Not that RSS is all THAT new or anything. I just mean you are definitely nerd (in the best sense of the word :)).

Quote:

---

Originally Posted by tr333

starting with the most basic of examples:

Code:

---

REM Basic
10 Dim s As String
20 s = "Hello World"
30 WHILE (TRUE)
40    PRINT s
50 WEND

// C
int main(void) {
    char *s = "Hello World";
    while (1) {
        printf("%s", s);
    }
}

-- haskell
main :: IO()
main =
    do
        (putStr . unlines) ["Hello", "World"]

// C#
string s = "Hello World";
    while (1) {
    Console.WriteLine(s);
}

' VB.NET
Dim s As String = "Hello World"
    While True
    Console.WriteLine(s)
Wend

---

moving up a bit:

Code:

---

#include <stdio.h>

int main(void)
{
        int x = 1, sum = 0, num = 0, tmp = 0;

        while(x <= 100)
        {
                tmp = 0;
                sum = 0;

                while(tmp <= x)
                {
                        sum = sum + tmp;
                        tmp++;
                }

                num = num + sum;
                printf("%i\n", num);
                x++;
        }

        return(0);
}

---

Code:

---

int isAlpha(char c) {
        if (c < 'A') {return(-1);}
        else if (c > 'z') {return(-1);}
        else if ((c > 'Z') && (c < 'a')) {return(-1);}
        else {return(0);}
}


char toLower(char c) {
        int alpha_offset = 'a' - 'A';
        if (isAlpha(c) == 0) {
                if ((c >= 'A') && (c <= 'Z')) {return(c + alpha_offset);}
        }
        return(c);
}

---

---

Code:

---

// Previously Basic
String s = "Hello World"
While(True) {
    Console.Print(s)
}

// Previously C
Private Void Main() {
String s = "Hello World"
    While(True) {
        Console.Print(s)
    }
}

// Perviously haskell
Private Void Main() {
String s = "Hello{NL}World"
    While(True) {
        Console.Print(s)
    }
}

// Perviously C#
// Same as C

// Perviously VB.NET
// Same as C

---

Code:

---

Private Void Main()
{
        Int x = 1
        Int sum = 0
        Int num = 0
        Int tmp = 0

        While(x <= 100)
        {
                tmp = 0
                sum = 0

                while(tmp <= x)
                {
                        sum += tmp
                        tmp++
                }

                num += sum
                Console.Print("{num}{NL}")
                x++
        }
}

---

Code:

---

Private Bool isAlpha(String c) {
        If(c < "A") {Return(True)}
        Else If (c > "z") {Return(True)}
        Else If ("a" > c > "Z") {Return(True)}
        Else {Return(False)}
}


Private String toLower(String c) {
        Int alpha_offset = "a" - "A"
        If(!isAlpha(c)) {
                if ("Z" >= c >= "A") {Return(c + alpha_offset)}
        }
        Return(c)
}

---

Thanks for the code sample tr333. We need to get more of these. I think I will try to post a small project that we could translate soon.

This brings up a good point. Is it worthwhile to offer a Char type? I am sure it is easy for the compiler to work with it since it won't be variable in length?

What do you guys think? I'm leaning towards yes.

One other thing I have been meaing to mention. I was considering wether or not we should have references like VB or have code-based library calls and includes like in C++.

I think code based would be better because people just learning the language would have a better understanding of whats happening (where as a noob to VB could go months without knowing how to access References in the VB IDE).

Also, it is easier to put the Include in the code when you are trying to transfer the project from one person to another (like on a forum) without the acutal file.

To make a nice compramise the IDE could automatically propmy you with a list of includes and would insert it in the propper place for you. That is the best of both worls IMO.

Converting Haskell to LF is rather pointless. Oh, and the conversion is wrong, because the Haskell code doesn't loop indefinitely, and it prints a newline between Hello and World.

If the String "type" is going to be a class, then of course you need a char type. Of course they won't be variable length as it can only hold 8 bytes (or it should anyway, unless you break the standard and have a char hold 32).

chem

Quote:

---

Originally Posted by CornedBee

Converting Haskell to LF is rather pointless. Oh, and the conversion is wrong, because the Haskell code doesn't loop indefinitely, and it prints a newline between Hello and World.

---

I'll change that.

Quote:

---

Originally Posted by chemicalNova

If the String "type" is going to be a class, then of course you need a char type. Of course they won't be variable length as it can only hold 8 bytes (or it should anyway, unless you break the standard and have a char hold 32).

chem

---

I know we had talked about making Stings into a class. I guess that makes sense though. It would be 8 bytes I'm sure and I know it won't be variable in length. I said that above.

Quote:

---

Originally Posted by eyeRmonkey

What do you guys think? I'm leaning towards yes.

---

:)

chem

Bits, not bytes. 8 bytes per character is a LOT.

Which reminds me: Unicode support? The world is moving towards it.