Author Topic: The dawn of a new OOPL (Read 25904 times)

BlakPilar · « **Reply #60 on:** August 19, 2011, 12:33:08 pm »

UPDATE It's been 24 hours, so this double post is allowed
My scanner and parser are now done in a simple sense. By that I mean that they can recognize arithmatic operations, some keywords, the four current modifiers, and non-dimensional variable declarations (as well as errors!). I'm going to be changing the way arrays/matrices are setup, though, basing it off of HOMER's suggestion.

One step closer

Binder News · « **Reply #61 on:** August 19, 2011, 12:56:38 pm »

I didn't get much done as I spent the whole of yesterday evening making Mineopoly on my Minecraft server with my friend. Lol.

BlakPilar · « **Reply #62 on:** August 20, 2011, 11:50:44 pm »

UPDATE
Alright, I've been working on this for the last three hours because I dun goof'd a lot and I had many of the destructive "brain-farts", but classes definitions are now parsed correctly!

By that I mean "global class hello {}" is now a valid statement.

Binder News · « **Reply #63 on:** August 21, 2011, 02:59:09 am »

Oh. Yay. Well, I got a Tokenizer done, which I can use, and a Logger framework started, which will llow for variable level of debug information to be displayed, depending on a command line flag probably.

EDIT: Uploaded Finished tokenizer and unfinished Logger framework.

Scipi · « **Reply #64 on:** September 18, 2011, 01:18:56 pm »

What's the progress on this? Is this still alive?

BlakPilar · « **Reply #65 on:** September 18, 2011, 01:28:48 pm »

Yeah, it's still being worked on on my part, but I haven't really had much time to do much what with school starting and everything. Plus we had a movie theater open near me sooo... lol

I ran into a snag with strings not processing correctly the last time I worked on it, but integers work for the most part.

Binder News · « **Reply #66 on:** September 24, 2011, 10:34:12 am »

Here. I've been busy as well, but I could still take a look at it.

BlakPilar · « **Reply #67 on:** September 24, 2011, 01:16:32 pm »

Alright. I've attached what I have so far and a sample code file I've been working with. Strings are fine until you throw in an escape character (i.e. \n, \t, etc.) and integers don't register for some reason. I haven't really gone in-depth with the debugging, so I'm not entirely sure what it is. I might be able to work on it for a bit later today/tonight.

Binder News · « **Reply #68 on:** September 24, 2011, 04:14:32 pm »

BlakPilar, in the file Scanner.cs, line 108. Change input.Read() to input.Peek(). With the read, you were consuming it twice (meaning that you ended up consuming the character after it as well, without reading it).

EDIT: Post 701! I'm an Addict! Oh, that kinda sounds weird, doesn't it? Well, I'm Level 8!

BlakPilar · « **Reply #69 on:** September 24, 2011, 04:19:49 pm »

Oh, duh haha. And maybe it was bytes not getting read correctly... I'll check it later and see if I have any more .Read()/.Peek() errors. (That is, unless you beat me to it

)

EDIT: Yeah, I know the difference between Read and Peek lol. I was probably just not thinking at the time. And congrats!

BlakPilar · « **Reply #70 on:** November 29, 2011, 05:18:54 pm »

Okay, so, I'm starting to work on this again, but it's not going to be getting my full attention. I'm thinking about possible syntax changes to make it a little easier to code in, because I want this to be an easy and powerful language. I'll be going through all of the suggestions mentioned before in this post and will listen to any future ones. I'll also need to brush up on my System.Reflection.Emit, so that will take some time, but hopefully not much. So yeah... this is the announcement of a slow reboot, I guess!

EDIT: In my spare time, and when I get frustrated with BexIDE, I'll be working on an EBNF definition for the language. I'll upload it when it is sufficiently complete for my liking.

BlakPilar · « **Reply #71 on:** December 03, 2011, 06:53:54 pm »

UPDATE
First thing's first: I've decided on a name change. The new name for this OOPL is now Alpha.NET. HOMER-16 and I have also written a complete (well, mostly complete) EBNF which will help with understanding syntax. I might change one or two things, maybe some keywords (I just realized I need to define all reserved words...), but otherwise this will stay the same. I'm going to be rewriting what I had from my previous compiler because I'm pretty sure I did some things wrong, and it was very... static. So I guess this is becoming my second primary project.

Spoiler For EBNF:

<stmt> := [<modifier>] <type> <ident> [ = <expr> | (<expr>)] | <ident> = <expr> | <compound_stmt> | <jump_stmt> | <stmt> ; <stmt> | <ident> (<expr>); | [<ident> = ] <ident>, ‘.’, <method>() [, ‘.’ <method>() ...] | <ident> = <ident>[, ‘.’, <ident>] | <ident_block_call> <expr> := <type.value> | <arith_expr> | <ident> <type> := str | int | dbl | flt | byte | bool | char | class | enum | struct | void | auto <arith_expr> := <expr> <arith_op> <expr> <arith_op> := + | - | * | / | % | ^ | ++ | -- | += | -= | *= | /= | %= | ^= | >> | << | <bool> ? <stmt> : <stmt> <ident> := <letter> [,<alpha_numeric>*] <alpha_numeric> := <letter> | <digit> | ‘_’ <int> := <digit>+ <digit> := 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 <dbl> := <digit>.<digit>*[d | D] - 15 max <flt> := <digit>.<digit>*[f | F] - 7 max <byte> := 0x<hex> | $<hex> | <hex>h | <binary>b <hex> := ‘0’ | ‘1’ | ‘2’ | ‘3’ | ‘4’ | ‘5’ | ‘6’ | ‘7’ | ‘8’ | ‘9’ | ‘A’ | ‘B’ | ‘C’ | ‘D’ | ‘E’ | ‘F’ | ‘a’ | ‘b’ | ‘c’ | ‘d’ | ‘e’ | ‘f’ | <binary>:= ‘0’ | ‘1’ <bool> := true | !0 | false | 0 <char> := <any ASCII equivalent: 0 - 255> <str> := “ <string_elem>* “ <string_elem> := <any char including escape chars> <escape_chars> := \a | \b | \f | \n | \r | \t | \x | \\ | \” | \’ | \0 <class> := [<modifier>] class <name> [: <class>* | <struct>*, <interface>*] <compound_stmt> <struct>:= [<modifier>] struct <name> [: <class>* | <struct>*, <interface>*] <compound_stmt> <modifier> := public | private | static | sealed | abstract | virtual <loop> := for [<ident>] (<numeric_ident>; <numeric_ident> <comparison> - '==' <numeric_ident>; <numeric_ident> <arith_op>) { <stmt> } | do [<ident>] { <stmt> } while (true) | while [<ident>] (true) { <stmt> } <numeric_ident>:= int | flt | dbl | byte <comparison> := < | > | <= | >= | == | != <enum> := [<modifier>] enum <name> { <enum_list> } <enum_list> := <ident> [‘,’] | <ident> = <expr> [‘,’] <name> := ident <letter>:= ‘A’ | ‘a’ | ‘B’ | ‘b’ | ‘C’ | ‘c’ | ‘D’ | ‘d’ | ‘E’ | ‘e’ | ‘F’ | ‘f’ | ‘G’ | ‘g’ | ‘H’ | ‘h’ | ‘I’ | ‘i’ | ‘J’ | ‘j’ | ‘K’ | ‘k’ | ‘L’ | ‘l’ | ‘M’ | ‘m’ | ‘N’ | ‘n’ | ‘O’ | ‘o’ | ‘P’ | ‘p’ | ‘Q’ | ‘q’ | ‘R’ | ‘r’ | ‘S’ | ‘s’ | ‘T’ | ‘t’ | ‘U’ | ‘u’ | ‘V’ | ‘v’ | ‘W’ | ‘w’ | ‘X’ | ‘x’ | ‘Y’ | ‘y’ | ‘Z’ | ‘z’ <compound_stmt>:= ‘{’, <stmt>*, ‘}’ <control_line> := ‘#’, ‘def ’, <ident> | ‘#’, ‘undef ’, <ident> | ‘#’, ‘if ’, <ident> (* ident must have been declared using #def *) | ‘#’, ‘if ’, <const_expr> | ‘#’, ‘elif ’ <ident> | ‘#’, ‘elif ’ <const_expr> | ‘#’, ‘else ’ <ident> | ‘#’, ‘endif’ | ‘#’, ‘bldt ’, <build_type> <build_type> := console | form | lib <const_expr> := any “hard coded” expression resulting in <bool> <jump_stmt> := break | cont | ret | lbl | goto <switch_block> := case <ident> {<type>: <stmt>} <if> := if [<ident>] (true) {<stmt>} <logic_expr> := && | ‘||’ | ‘!|’ | ! <self> := this <void> := NULL <ident_block_call> := <ident>,'<',<numeric_ident>,'>' <method>:= [<ident>,'.' | <self>,'.'] <name>(<params>) <params>:= <type> <ident>*

Project SummaryAST.............90% Scanner.........10% Validator........0% Parser...........0% IL Generator.....0% IDE..............0% =-=-=-=-=-=-=-=-=-= Overall.........17%

BlakPilar · « **Reply #72 on:** December 06, 2011, 04:35:20 pm »

Well, nothing really to update as of yet, but I have a question. Would it help to have something like $428F automatically converted to an array like {$42, $8F} or an integer/double (17,039 in this case)? So you could do something like byte[] arr = $428F; //arr == {$42, $8F} or int b = $428F; //b == 17039. It's not a guarantee that it'll work, but I figured it'd be helpful. Also, possible minor syntax changes so I can call the language my own.

But thanks to anyone still following this

I have yet to go through this topic again, but if you have one, please share suggestions with me!

EDIT: Also, when I have everything but the IL Generator done, I'll be uploading my source. Hopefully by then I'll have my website all setup and running.

Project SummaryAST............100% Scanner........100% Validator........5% Parser...........0% IL Generator.....0% IDE..............0% =-=-=-=-=-=-=-=-=-= Overall.........34%

Scipi · « **Reply #73 on:** December 06, 2011, 07:21:32 pm »

For '{}' could they be "[[]]" instead? So:

Code: [Select]

while(Arg)
[[
//Code
]]

BlakPilar · « **Reply #74 on:** December 06, 2011, 07:25:51 pm »

Of course! Easily done!

Author Topic: The dawn of a new OOPL (Read 25904 times)

BlakPilar

Re: The dawn of a new OOPL

Binder News

Re: The dawn of a new OOPL

BlakPilar

Re: The dawn of a new OOPL

Binder News

Re: The dawn of a new OOPL

Scipi

Re: The dawn of a new OOPL

BlakPilar

Re: The dawn of a new OOPL

Binder News

Re: The dawn of a new OOPL

BlakPilar

Re: The dawn of a new OOPL

Binder News

Re: The dawn of a new OOPL

BlakPilar

Re: The dawn of a new OOPL

BlakPilar

Re: The dawn of a new OOPL

BlakPilar

Re: The dawn of a new OOPL

BlakPilar

Re: The dawn of a new OOPL

Scipi

Re: The dawn of a new OOPL

BlakPilar

Re: The dawn of a new OOPL