Compare commits

...

63 Commits

Author SHA1 Message Date
8f8f8ffe80 Update README.md 2025-04-07 17:50:32 +02:00
6ed7838a53 [Compiler-Rust]: Vec.get(0) -> Vec.first() 2024-01-02 17:22:36 +01:00
6085eb0d6e Add solution to project euler's problem 3 2023-12-23 22:10:07 +01:00
01c062ce8f Add increment function to the standard library 2023-12-23 22:05:32 +01:00
90d67867cc Fix fasm error when enqueueing ints > 2^32
On x64 the only instruction supporting imm64 is mov rXX,
imm64 so there has to be a special case for those numbers
2023-12-23 22:01:06 +01:00
1f78ad0a7f Progress on Project Euler problems 2023-12-22 22:57:55 +01:00
3440226658 Add newline 2023-12-22 22:19:52 +01:00
1beec6190e Add decrement to the standard library 2023-12-22 22:19:52 +01:00
8478b10210 Simplifications curtesy of clippy 2023-12-22 22:19:52 +01:00
dcfb3147ba Add first project euler problem 2023-07-28 07:05:09 +02:00
b1cb4a0a0e Optimize swp and req with two-item queue 2023-07-28 07:04:39 +02:00
92f7ec405d Add queue overflow check 2023-07-28 04:55:08 +02:00
09ac457b9d Fix names clobbering fasm's "type" keyword 2023-07-28 03:10:57 +02:00
abcbe9d68b Add divmod 2023-07-28 03:01:13 +02:00
95133360ee Make imports work on all OSes 2023-02-06 16:03:48 +01:00
5aa365c681 Improve importing 2023-02-06 15:00:37 +01:00
131386632f Fix tests 2023-01-23 06:08:26 +01:00
476aa8bfb2 Implement importing 2023-01-23 05:28:13 +01:00
4286fb4424 Start working on importing 2023-01-23 03:58:53 +01:00
4064309b26 Make oob check write to stderr instead of stdout 2023-01-11 19:45:38 +01:00
296ee6c5c8 Add icons for VSCode 2023-01-11 19:43:03 +01:00
d7c6cada32 remove decrease intrinsic+update editor support 2023-01-11 03:02:10 +01:00
34f52ae864 Add array oob checks + improve test handling 2023-01-11 02:33:30 +01:00
0448b56adf Improve diagnostics 2023-01-09 12:35:48 +01:00
83537737f5 Merge branch 'main' of https://gittea.dev/0x4261756D/kurz 2023-01-09 12:23:43 +01:00
df3e1b027f Time interpretation 2023-01-09 12:23:01 +01:00
2ce863c899 Add asm compilation and running 2023-01-05 00:19:29 +01:00
ffdc54947b Improve queue runaway problems
By changing the order of "pointer" increases it is possible to check and move back the queue if it is empty before function calls and req.
With the function call change in particular it is now possible to run conway_diehard.qbl a queue of length 1024 (down from ~16384)
2023-01-04 00:45:34 +01:00
3d2cd64c6d Implement enough asm generation to compile rule110 2023-01-03 21:52:48 +01:00
870dcb105d Readd queue runaway optimizations in comparisons 2023-01-03 20:55:05 +01:00
a754602bde Readd queue runaway optimization in intToStr 2023-01-03 20:44:07 +01:00
84442e5eb9 Finish enough asm generation to compile tests/while.qbl 2023-01-03 20:39:08 +01:00
5bae80e9aa Assembly generation, WIP 2023-01-03 17:07:57 +01:00
fbccc9dd15 Remove any from syntax highlighting 2022-12-31 16:14:52 +01:00
cdfa5b1310 Fix tests failing on Windows due to EOL shenanigans 2022-12-31 16:12:53 +01:00
41667ff4a3 Remove the "any" type 2022-12-31 16:09:55 +01:00
e3a1223ac9 WIP assembly gen 2022-12-23 14:49:09 +01:00
97bd50fa55 Fix Conway test output 2022-12-22 11:53:13 +01:00
fb37c4f58b Update conways 2022-12-22 11:43:17 +01:00
6d31d5da98 Add Conways Game of Life 2022-12-22 10:29:57 +01:00
23c6ec8dad Add more intrinsics 2022-12-22 10:29:25 +01:00
5682ee708c Add syntax highlighting for vscode 2022-12-22 05:44:08 +01:00
53cfd52dcd Improve rule110 2022-12-22 00:48:24 +01:00
8397d32756 Add array tests 2022-12-22 00:48:02 +01:00
bf4a08e484 Make interpret fallible, fix applications at end of file, add more checks 2022-12-22 00:47:41 +01:00
0abc74f043 Add editor syntax file 2022-12-22 00:45:59 +01:00
dedb601a9b Small updates to celebrate turing-completeness 2022-12-22 00:08:13 +01:00
e4a7bcccc0 Add arrays, making the language turing complete
See tests/rule110.qbl for turing completeness
2022-12-21 21:48:52 +01:00
2255aaff8b Fix sending detection 2022-12-19 12:13:40 +01:00
f60441ef9d Improved the test system 2022-12-18 04:05:50 +01:00
35f65101ad Add testing functionality and add error reporting 2022-12-15 21:53:01 +01:00
0de0838284 Add new operations and intrinsics 2022-12-15 20:34:56 +01:00
d1883ff3ab Introduce boolean types 2022-12-14 20:39:51 +01:00
2a560cfcef Add and update tests 2022-12-14 11:46:47 +01:00
c2800dfcc7 Make while swallow the input 2022-12-14 11:46:39 +01:00
89a7780d10 Add some tests 2022-12-14 09:34:49 +01:00
48cb618d5a Add interpreter 2022-12-14 09:34:41 +01:00
abfb3d2f3d Make requeueing an operation instead of an intrinsic 2022-12-14 08:14:27 +01:00
d9e7f18049 Implemented function validation and type checking 2022-12-14 07:57:51 +01:00
ac6a835a64 Began validation 2022-12-14 04:12:03 +01:00
1e8322cc7c Merge branch 'main' of https://gittea.dev/0x4261756D/kurz 2022-12-14 01:48:14 +01:00
6a2bc25eaf Implement function extraction and their removal from the token stream 2022-12-05 00:38:20 +01:00
9976ef9fe9 Initial commit, barebones tokenizer working 2022-11-29 02:04:01 +01:00
46 changed files with 3564 additions and 2 deletions

2
.gitignore vendored Normal file
View File

@ -0,0 +1,2 @@
/target
out*

7
Cargo.lock generated Normal file
View File

@ -0,0 +1,7 @@
# This file is automatically @generated by Cargo.
# It is not intended for manual editing.
version = 3
[[package]]
name = "kurz"
version = "0.1.0"

8
Cargo.toml Normal file
View File

@ -0,0 +1,8 @@
[package]
name = "kurz"
version = "0.1.0"
edition = "2021"
# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
[dependencies]

View File

@ -1,6 +1,6 @@
MIT License
Copyright (c) <year> <copyright holders>
Copyright (c) 2022-2023 0x4261756D
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

View File

@ -1,3 +1,11 @@
**Moved to [Codeberg](https://codeberg.org/0x4261756D/kurz)**
# kurz
Queue based language
## Roadmap
- [x] turing completeness
- [ ] compiled
- [ ] cross-plattform
- [ ] self-hosted

29
editor/micro/kurz.yaml Normal file
View File

@ -0,0 +1,29 @@
filetype: kurz
detect:
filename: "\\.qbl$"
rules:
- comment:
start: "//"
end: "$"
rules:
- todo: "(TODO|FIXME):?"
- constant.string:
start: "\""
end: "\""
skip: "\\\\."
rules:
- constant.specialChar: "\\\\."
- constant.bool: "\\b(true|false)\\b"
# ints
- constant.number: "\\b([0-9]*)\\b"
- symbol.brackets: "[{}]"
# function
- identifier: "function"
- type: "\\b(bool|int|str)\\b"
- symbol.operator: "([-+<>]|==|!=|=>|print(ln)?)"
- identifier: "\\b(if|else|while)\\b"
- special: "\\b(deq|swp|dup|req|depth)\\b"
- special: "\\?\\?\\?"
- statement: "arr"

View File

@ -0,0 +1,17 @@
// A launch configuration that launches the extension inside a new window
// Use IntelliSense to learn about possible attributes.
// Hover to view descriptions of existing attributes.
// For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
{
"version": "0.2.0",
"configurations": [
{
"name": "Extension",
"type": "extensionHost",
"request": "launch",
"args": [
"--extensionDevelopmentPath=${workspaceFolder}"
]
}
]
}

View File

@ -0,0 +1,4 @@
.vscode/**
.vscode-test/**
.gitignore
vsc-extension-quickstart.md

View File

@ -0,0 +1,9 @@
# Change Log
All notable changes to the "kurz-lang" extension will be documented in this file.
Check [Keep a Changelog](http://keepachangelog.com/) for recommendations on how to structure this file.
## [Unreleased]
- Initial release

View File

@ -0,0 +1,5 @@
# kurz-lang README
This is a syntax highlighting extension for the lanuage '[kurz](https://gittea.dev/0x4261756D/kurz)'.
**Enjoy!**

Binary file not shown.

After

Width:  |  Height:  |  Size: 612 B

Binary file not shown.

After

Width:  |  Height:  |  Size: 612 B

Binary file not shown.

View File

@ -0,0 +1,20 @@
{
"comments": {
// symbol used for single line comment. Remove this entry if your language does not support line comments
"lineComment": "//",
},
// symbols used as brackets
"brackets": [
["{", "}"],
],
// symbols that are auto closed when typing
"autoClosingPairs": [
["{", "}"],
["\"", "\""],
],
// symbols that can be used to surround a selection
"surroundingPairs": [
["{", "}"],
["\"", "\""],
]
}

View File

@ -0,0 +1,29 @@
{
"name": "kurz-lang",
"displayName": "kurz",
"description": "Syntax highlighting for the kurz language",
"version": "0.0.1",
"engines": {
"vscode": "^1.74.0"
},
"categories": [
"Programming Languages"
],
"contributes": {
"languages": [{
"id": "kurz",
"aliases": ["kurz", "kurz"],
"extensions": [".qbl"],
"configuration": "./language-configuration.json",
"icon": {
"light": "./icons/kurz-light.png",
"dark": "./icons/kurz-dark.png"
}
}],
"grammars": [{
"language": "kurz",
"scopeName": "source.kurz",
"path": "./syntaxes/kurz.tmLanguage.json"
}]
}
}

View File

@ -0,0 +1,89 @@
{
"$schema": "https://raw.githubusercontent.com/martinring/tmlanguage/master/tmlanguage.json",
"name": "kurz",
"patterns":
[
{
"include": "#keywords"
},
{
"include": "#strings"
},
{
"include": "#comment"
}
],
"repository":
{
"keywords":
{
"patterns":
[
{
"name": "keyword.control.kurz",
"match": "\\b(if|else|while)\\b"
},
{
"name": "keyword.operator",
"match": "[-+<>]|==|!=|=>|print(ln)?"
},
{
"name": "support.function.kurz",
"match": "function"
},
{
"name": "support.type.kurz",
"match": "\\b(bool|int|str)\\b"
},
{
"name": "constant.numeric.kurz",
"match": "[0-9]"
},
{
"name": "constant.language.kurz",
"match": "true|false"
},
{
"name": "support.function.kurz",
"match": "\\b(deq|swp|dup|req|depth)\\b"
},
{
"name": "keyword.control",
"match": "arr"
}
]
},
"comment":
{
"patterns":
[
{
"name": "comment.line.kurz",
"begin": "//",
"end": "$",
"patterns":
[
{
"name": "markup.bold.kurz",
"match": "(TODO|FIXME):?"
}
]
}
]
},
"strings":
{
"name": "string.quoted.double.kurz",
"begin": "\"",
"end": "\"",
"patterns":
[
{
"name": "constant.character.escape.kurz",
"match": "\\\\."
}
]
}
},
"scopeName": "source.kurz"
}

View File

@ -0,0 +1,29 @@
# Welcome to your VS Code Extension
## What's in the folder
* This folder contains all of the files necessary for your extension.
* `package.json` - this is the manifest file in which you declare your language support and define the location of the grammar file that has been copied into your extension.
* `syntaxes/kurz.tmLanguage.json` - this is the Text mate grammar file that is used for tokenization.
* `language-configuration.json` - this is the language configuration, defining the tokens that are used for comments and brackets.
## Get up and running straight away
* Make sure the language configuration settings in `language-configuration.json` are accurate.
* Press `F5` to open a new window with your extension loaded.
* Create a new file with a file name suffix matching your language.
* Verify that syntax highlighting works and that the language configuration settings are working.
## Make changes
* You can relaunch the extension from the debug toolbar after making changes to the files listed above.
* You can also reload (`Ctrl+R` or `Cmd+R` on Mac) the VS Code window with your extension to load your changes.
## Add more language features
* To add features such as IntelliSense, hovers and validators check out the VS Code extenders documentation at https://code.visualstudio.com/docs
## Install your extension
* To start using your extension with Visual Studio Code copy it into the `<user home>/.vscode/extensions` folder and restart Code.
* To share your extension with the world, read on https://code.visualstudio.com/docs about publishing an extension.

View File

@ -0,0 +1,55 @@
//valid,4613732
//:END:
import "../std.qbl"
true 1 2 0 0 0
while
{
// a b even_count 0 sum
fib
// even_count 0 sum b a+b
dup ==
// sum b a+b even_count even?
req req req req
if
// sum b a+b even_count
{
req dup
// b a+b even_count sum b
req req deq 3 +
// b a+b even_count sum+b
req req
}
else
{
req req req
}
// even_count sum+b b a+b
decrement 0 req req
// a+b even_count-1 0 sum+b b
check
// even_count-1 0 sum+b b a+b big?
req req req req swp
// even_count-1 0 sum+b b big? a+b
req req req swp
// a+b even_count-1 0 sum+b big? b
req req req req
// big? b a+b even_count-1 0 sum+b
}
deq deq deq deq intToStr println
function int int => int int fib
{
req dup +
}
function int => int nPrint
{
dup intToStr req print
}
function int => int bool check
{
4000000 dup <
}

View File

@ -0,0 +1,56 @@
//valid,6857
//:END:
import "../std.qbl"
600851475143 2 divHasLargerPF 3
while
{
// n p
req dup 2 req req +
// p n p+2
swp req
// n p p+2
divHasLargerPF
// p+2 hasLPF newN
req
}
deq 2 - intToStr println
function int int => bool int divHasLargerPF
{
req dup req divIfDivisible
// n newN
1 req dup
// newN 1 n newN
== req req
// is1 n newN
if
{
false req deq
}
else
{
true deq req
}
}
function int int => int divIfDivisible
{
// p n
dup req dup swp
// p n p n
divmod 0
// p n p/n p%n 0
req req req ==
// p n p/n isDiv
req req req
if
{
deq req divIfDivisible
}
else
{
req deq deq
}
}

View File

@ -0,0 +1,32 @@
//valid,233168
//:END:
import "../std.qbl"
true 3 0
while
{
// i sum
dup check req req +
incAndCheck req
}
deq intToStr println
function int => int check
{
3 dup % 0 req != req
// not_divisible start
if
{
5 dup % 0 req != req
if
{
deq 0
}
}
}
function int => bool int incAndCheck
{
1 + 1000 dup < req
}

1990
src/main.rs Normal file
View File

@ -0,0 +1,1990 @@
use core::panic;
use std::collections::HashMap;
use std::env;
use std::fs;
use std::iter::Peekable;
use std::path::PathBuf;
use std::process::Command;
use std::process::Stdio;
use std::process::exit;
use std::time::Instant;
#[derive(Debug, Clone, PartialEq)]
enum Token
{
StringLit(String, i32, i32),
IntLit(String, i32, i32),
BoolLit(String, i32, i32),
Keyword(String, i32, i32),
Apply(String, String, i32, i32),
Import(i32, i32),
}
enum TokenizerState
{
Whitespace,
Quote,
Keyword,
Comment,
}
#[derive(Debug,Clone,Copy, PartialEq)]
enum Datatype
{
Int,
String,
Bool,
//Pointer,
// Any,
}
// impl PartialEq for Datatype
// {
// fn eq(&self, other: &Self) -> bool
// {
// core::mem::discriminant(self) == core::mem::discriminant(&Datatype::Any) ||
// core::mem::discriminant(other) == core::mem::discriminant(&Datatype::Any) ||
// core::mem::discriminant(self) == core::mem::discriminant(other)
// }
// }
#[derive(Debug)]
struct Function
{
name: String,
ins: Vec<Datatype>,
outs: Vec<Datatype>,
content: Vec<Operation>,
}
#[derive(Debug)]
struct Arr
{
name: String,
datatype: Datatype,
length: i64,
data: Vec<String>,
}
#[derive(Debug)]
enum Operation
{
Enqueue(Datatype, String, i32, i32),
Dequeue(i32, i32),
// TODO: req can be implemented in terms of dup and dequeue
Requeue(i32, i32),
Swap(i32, i32),
Dup(i32, i32),
Intrinsic(String, i32, i32),
FunctionCall(String, i32, i32),
If(Vec<Operation>, Option<Vec<Operation>>, i32, i32),
While(Vec<Operation>, i32, i32),
Apply(String, String, i32, i32),
Depth(i32, i32),
QueueDiagnostic(i32, i32),
Interrupt(i32, i32),
}
fn main()
{
let intrinsics: HashMap<&str, (Vec<Datatype>, Vec<Datatype>)> = HashMap::from(
[
("print", (Vec::from([Datatype::String]), Vec::new())),
("println", (Vec::from([Datatype::String]), Vec::new())),
("intToStr", (Vec::from([Datatype::Int]), Vec::from([Datatype::String]))),
("-", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Int]))),
("+", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Int]))),
("*", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Int]))),
("divmod", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Int, Datatype::Int]))),
("<", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Bool]))),
(">", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Bool]))),
(">=", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Bool]))),
("<=", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Bool]))),
("==", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Bool]))),
("!=", (Vec::from([Datatype::Int, Datatype::Int]), Vec::from([Datatype::Bool]))),
("&&", (Vec::from([Datatype::Bool, Datatype::Bool]), Vec::from([Datatype::Bool]))),
]);
let args: Vec<String> = env::args().collect();
if args.len() < 2
{
usage()
}
let mut debug = false;
let mut interpret = false;
let mut run = false;
for arg in &args[3..]
{
match arg.as_str()
{
"-d" | "--debug" => debug = true,
"-i" | "--interpret" => interpret = true,
"-r" | "--run" => run = true,
_ => panic!("Unknown option {}", arg),
}
}
match args[1].as_str()
{
"-t" | "--test" =>
{
let mut count = 0;
for f in fs::read_dir(&args[2]).unwrap()
{
let f = f.unwrap();
let file_content = fs::read_to_string(f.path()).unwrap().replace("\r\n", "\n");
println!("========NOW TESTING {:?}========", f.path());
match compile(&file_content, f.path().to_str().unwrap(), &intrinsics, interpret, run, debug)
{
Ok(maybe_msg) =>
{
println!("---Successfully parsed {:?}---", f.path());
if let Some(msg) = &maybe_msg
{
print!("---Output---\n'{}'\n", msg);
}
let expected = &format!("//valid,{}:END:", maybe_msg.unwrap_or(String::new()).replace('\n', "\n//"));
if file_content.starts_with(expected)
{
println!("===PASSED===");
count += 1;
}
else if let Some(index) = file_content.find(":END:")
{
let expected_output = file_content[8..index].replace("\n//", "\n");
println!("\n===FAILED===\nExpected the output to be\n'{}'\n({})", expected_output, expected);
}
else
{
panic!("Could not find an ending marker (:END:) for the expected output in {:?}", f.file_name());
}
}
Err(msg) =>
{
println!("ERROR: {}", msg);
if file_content.starts_with(&format!("//invalid,{}:END:", msg.replace('\n', "\n//")))
{
println!("===PASSED===");
count += 1;
}
else if file_content.starts_with("//invalid,")
{
if let Some(index) = file_content.find(":END:")
{
let expected = &format!("//invalid,{}:END:", msg.replace('\n', "\n//"));
let expected_output = file_content[10..index].replace("\n//", "\n");
println!("\n===FAILED===\nExpected the output to be\n'{}'\n({})", expected_output, expected);
}
else
{
panic!("Could not find an ending marker (:END:) for the expected output in {:?}", f.file_name());
}
}
else
{
println!("Unexpected error");
}
}
}
}
println!("\n\n=========RESULT=========\n{}/{}", count, fs::read_dir(&args[2]).unwrap().count());
}
"-c" | "--compile" =>
{
let file_content = fs::read_to_string(&args[2]).expect("Could not read the source file");
match compile(&file_content, &args[2], &intrinsics, interpret, run, debug)
{
Ok(maybe_msg) =>
{
if let Some(msg) = maybe_msg
{
print!("---Output---\n\n{}", msg);
}
}
Err(msg) => println!("ERROR: {}", msg),
}
}
_ => panic!("Unknown option {}", args[1])
}
}
fn compile(file_content: &str, file_path: &str, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, interpret: bool, run: bool, debug: bool) -> Result<Option<String>, String>
{
let mut tokens: Vec<Token> = tokenize(file_content)?;
println!("---Done tokenizing, got {} tokens---", tokens.len());
let mut functions: Vec<Function> = Vec::new();
extract_functions(&mut tokens, &mut functions, intrinsics, debug)?;
println!("---Done extracting functions, got {} functions and reduced the token count to {}---", functions.len(), tokens.len());
resolve_imports(&mut tokens, &mut functions, file_path, &mut Vec::from([fs::canonicalize(file_path).unwrap()]), intrinsics, debug)?;
println!("---Done importing files---");
let mut arrays: Vec<Arr> = extract_arrays(&mut tokens, intrinsics, &functions, debug)?;
println!("---Done extracting arrays, got {} arrays and reduced the token count to {}---", arrays.len(), tokens.len());
let operations = parse_until_delimiter(&mut tokens.iter().peekable(), intrinsics, None, debug)?;
println!("---Done parsing tokens into {} operations---", operations.len());
validate_function_calls(&operations, &functions, debug)?;
println!("---Done validating function calls---");
typecheck(&operations, &functions, intrinsics, &arrays, debug)?;
println!("---Done typechecking---");
let start = Instant::now();
let output = if interpret
{
println!("---Starting to interpret the program---");
Some(interpret_program(&operations, &mut Vec::new(), &functions, &mut arrays, debug)?)
}
else
{
None
};
if !interpret
{
if let Err(err) = generate_assembly_linux_x64(&operations, &functions, &arrays)
{
return Err(err.to_string());
}
let mut fasm_process = match Command::new("fasm").arg("out.asm").spawn()
{
Ok(process) => process,
Err(err) => return Err(format!("Fasm process error: {}", err)),
};
match fasm_process.wait()
{
Ok(status) =>
{
if !status.success()
{
return Err(format!("fasm exited with an error: {}", status));
}
}
Err(err) => return Err(err.to_string()),
}
}
if run
{
let process = match Command::new("./out").stdout(Stdio::piped()).stderr(Stdio::piped()).spawn()
{
Ok(process) => process,
Err(err) => return Err(err.to_string()),
};
return match process.wait_with_output()
{
Ok(output) =>
{
match String::from_utf8(output.stdout)
{
Ok(stdout) =>
{
match String::from_utf8(output.stderr)
{
Ok(stderr) =>
{
let text = format!("{}{}", stdout, stderr);
match output.status.code()
{
Some(0) => Ok(Some(text)),
_ => Err(text),
}
}
Err(err) => Err(err.to_string()),
}
}
Err(err) => Err(err.to_string()),
}
}
Err(err) => Err(err.to_string()),
};
}
println!("---Done after {:?}---", start.elapsed());
Ok(output)
}
fn resolve_imports(tokens: &mut Vec<Token>, functions: &mut Vec<Function>, file_path: &str, visited_paths: &mut Vec<PathBuf>, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, debug: bool) -> Result<(), String>
{
let mut tokens_iter = tokens.iter();
let mut new_tokens: Vec<Token> = Vec::new();
while let Some(token) = tokens_iter.next()
{
if let Token::Import(line, col) = token
{
if let Some(Token::StringLit(import_path, _, _)) = tokens_iter.next()
{
println!("Resolving {}", import_path);
let full_import_path = fs::canonicalize(file_path).map_err(|e| e.to_string())?.parent().unwrap().join(import_path);
if visited_paths.contains(&full_import_path)
{
println!("--Already visited {}--", full_import_path.display());
}
else
{
visited_paths.push(full_import_path.clone());
let maybe_file_content = fs::read_to_string(full_import_path);
match maybe_file_content
{
Ok(file_content) =>
{
let mut import_tokens: Vec<Token> = tokenize(&file_content)?;
println!("--Done tokenizing the imported file at {}:{}, got {} tokens--", line, col, tokens.len());
extract_functions(&mut import_tokens, functions, intrinsics, debug)?;
resolve_imports(&mut import_tokens, functions, file_path, visited_paths, intrinsics, debug)?;
println!("--Now totalling {} functions--", functions.len());
}
Err(e) => return Err(e.to_string()),
}
}
}
else
{
return Err(format!("Expected an import location at {}:{}", line, col));
}
}
else
{
new_tokens.push(token.clone());
}
}
tokens.clear();
tokens.extend_from_slice(&new_tokens);
Ok(())
}
struct AssemblyData
{
strings: String,
code: String,
arrays: String,
}
fn merge_assemblies(data: &mut AssemblyData, data2: AssemblyData)
{
data.arrays += data2.arrays.as_str();
data.code += data2.code.as_str();
data.strings += data2.strings.as_str();
}
const ASSEMBLY_LINUX_X64_QUEUE_LENGTH: u32 = 8192;
const ASSEMBLY_LINUX_X64_HEADER: &str = "format ELF64 executable 3\n";
const ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE: &str = "\tcmp r12, r13\n\tcmove r12, r14\n\tcmove r13, r14\n";
const ASSEMBLY_LINUX_X64_EXIT: &str = "\tmov rax, 60\n\tmov rdi, 0\n\tsyscall\n";
const ASSEMBLY_LINUX_X64_DYNAMIC_DATA_LENGTH: u32 = 16384;
// r12: head
// r13: tail
// r14: base
// r15: dynamic end
fn generate_assembly_linux_x64(operations: &Vec<Operation>, functions: &Vec<Function>, arrays: &Vec<Arr>) -> Result<(), std::io::Error>
{
let mut data = AssemblyData
{
arrays: format!("segment readable writeable\n\tqueue: rq {}\n\tdynamic: rb {}\n", ASSEMBLY_LINUX_X64_QUEUE_LENGTH, ASSEMBLY_LINUX_X64_DYNAMIC_DATA_LENGTH),
strings: String::from("segment readable\n\tnewline: db 10\n"),
code: String::from("segment executable\n"),
};
for array in arrays
{
data.arrays += format!("\tarr_{}: rq {}\n", array.name, array.length).as_str();
}
data.code += "_start:\n";
merge_assemblies(&mut data, generate_assembly_linux_x64_block(operations, functions, arrays));
data.code += ASSEMBLY_LINUX_X64_EXIT;
for function in functions
{
merge_assemblies(&mut data, generate_assembly_linux_x64_function(function.name.as_str(), &function.content, functions, arrays));
}
if data.code.contains("call intToStr")
{
data.code += "intToStr:\n";
data.code += "\tmov rax, rdi\n";
data.code += "\tmov rsi, 10\n";
data.code += "\txor rdi, rdi\n";
data.code += "\txor rdx, rdx\n";
data.code += "\tintToStringLoop:\n";
data.code += "\t\tdiv rsi\n";
data.code += "\t\tadd rdx, 48\n";
data.code += "\t\tpush rdx\n";
data.code += "\t\txor rdx, rdx\n";
data.code += "\t\tinc rdi\n";
data.code += "\t\tcmp rax, 0\n";
data.code += "\t\tjne intToStringLoop\n";
data.code += "\tmov rsi, r15\n";
data.code += "\tmov qword [dynamic+r15], rdi\n";
data.code += "\tadd r15, 8\n";
data.code += "\tintToStringBuildLoop:\n";
data.code += "\t\tcmp rdi, 0\n";
data.code += "\t\tje intToStringBuildLoopEnd\n";
data.code += "\t\tpop rax\n";
data.code += "\t\tmov byte [dynamic+r15], byte al\n";
data.code += "\t\tinc r15\n";
data.code += "\t\tdec rdi\n";
data.code += "\t\tjmp intToStringBuildLoop\n";
data.code += "\tintToStringBuildLoopEnd:\n";
data.code += "\tmov byte [dynamic+r15], 0\n";
data.code += "\tinc r15\n";
data.code += "\tlea rax, [dynamic+rsi]\n";
data.code += "\tret\n";
}
if data.code.contains("exception_array_read_out_of_bounds")
{
data.strings += "\texception_array_oob_msg db \"Attempted array out-of-bounds access\", 10\n";
data.code += "exception_array_read_out_of_bounds:\n";
//TODO: report the passed sizes
data.code += "\tmov rax, 1\n";
data.code += "\tmov rdi, 2\n";
// size
data.code += "\tmov rdx, 37\n";
// data
data.code += "\tmov rsi, exception_array_oob_msg\n";
data.code += "\tsyscall\n";
data.code += "\tmov rax, 60\n";
data.code += "\tmov rdi, -1\n";
data.code += "\tsyscall\n";
}
if data.code.contains("exception_queue_read_out_of_bounds")
{
data.strings += "\texception_queue_oob_msg db \"Queue overflow\", 10\n";
data.code += "exception_queue_read_out_of_bounds:\n";
//TODO: report the passed sizes
data.code += "\tmov rax, 1\n";
data.code += "\tmov rdi, 2\n";
// size
data.code += "\tmov rdx, 37\n";
// data
data.code += "\tmov rsi, exception_queue_oob_msg\n";
data.code += "\tsyscall\n";
data.code += "\tmov rax, 60\n";
data.code += "\tmov rdi, -1\n";
data.code += "\tsyscall\n";
}
fs::write("out.asm", format!("{}{}{}{}", ASSEMBLY_LINUX_X64_HEADER, data.code, data.arrays, data.strings))
}
fn generate_assembly_linux_x64_array_oob_check(length: i64) -> String
{
let mut data = String::new();
data += "\t\t;;Array bounds check\n";
data += format!("\tcmp qword rax, {}\n", length).as_str();
data += "\tjge exception_array_read_out_of_bounds\n";
data += "\tcmp qword rax, 0\n";
data += "\tjl exception_array_read_out_of_bounds\n";
data += "\t\t;;Array bounds check over\n";
data.clone()
}
fn generate_assembly_linux_x64_queue_oob_check() -> String
{
return "\t\t;;Queue bounds check\n".to_string() +
format!("\tcmp qword r13, {}\n", ASSEMBLY_LINUX_X64_QUEUE_LENGTH).as_str() +
"\tjge exception_queue_read_out_of_bounds\n\t\t;;Queue bounds over\n";
}
fn generate_assembly_linux_x64_block(operations: &Vec<Operation>, functions: &Vec<Function>, arrays: &Vec<Arr>) -> AssemblyData
{
let mut data = AssemblyData
{
arrays: String::new(),
code: String::new(),
strings: String::new(),
};
for operation in operations
{
match operation
{
Operation::Dequeue(line, col) =>
{
data.code += format!("\t;;deq {}:{}\n", line, col).as_str();
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
}
Operation::Enqueue(datatype, value, line, col) =>
{
data.code += format!("\t;;enq {:?} {} {}:{}\n", datatype, value, line, col).as_str();
match datatype
{
Datatype::Int =>
{
if value.parse::<i64>().unwrap() > u32::MAX as i64
{
data.code += format!("\tmov rax, {}\n", value).as_str();
data.code += "\tmov qword [queue+8*r13], rax\n";
}
else
{
data.code += format!("\tmov qword [queue+8*r13], {}\n", value).as_str();
}
}
Datatype::Bool =>
{
data.code += format!("\tmov qword [queue+8*r13], {}\n", if value == "true" { 1 } else { 0 }).as_str();
}
Datatype::String =>
{
data.strings += format!("\tstr_{}_{}: db {}, {}, {}, {}, {}, {}, {}, {}, \"{}\", 0\n",
line, col,
value.len() % 256,
(value.len() >> 8) % 256,
(value.len() >> 16) % 256,
(value.len() >> 24) % 256,
(value.len() >> 32) % 256,
(value.len() >> 40) % 256,
(value.len() >> 48) % 256,
(value.len() >> 56) % 256,
value).as_str();
data.code += format!("\tlea rax, [str_{}_{}]\n", line, col).as_str();
data.code += "\tmov [queue+8*r13], rax\n";
}
}
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
Operation::Requeue(line, col) =>
{
data.code += format!("\t;;req {}:{}\n", line, col).as_str();
data.code += "\tmov rax, r13\n";
data.code += "\tsub rax, r12\n";
data.code += "\tcmp rax, 2\n";
data.code += format!("\tje req_{line}_{col}_special\n").as_str();
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
data.code += format!("\tjmp req_{line}_{col}_end\n").as_str();
data.code += format!("req_{line}_{col}_special:\n").as_str();
data.code += "\tmov qword rax, [queue+8*r12]\n";
data.code += "\tmov qword rdi, [queue+8*r12+8]\n";
data.code += "\tmov qword r12, r14\n";
data.code += "\tmov qword r13, r14\n";
data.code += "\tmov qword [queue+8*r12], rdi\n";
data.code += "\tmov qword [queue+8*r12+8], rax\n";
data.code += "\tadd r13, 2\n";
data.code += format!("req_{line}_{col}_end:\n").as_str();
}
Operation::Swap(line, col) =>
{
data.code += format!("\t;;swp {}:{}\n", line, col).as_str();
data.code += "\tmov rax, r13\n";
data.code += "\tsub rax, r12\n";
data.code += "\tcmp rax, 2\n";
data.code += format!("\tje swp_{line}_{col}_special\n").as_str();
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tmov rbx, [queue+8*r12+8]\n";
data.code += "\tadd r12, 2\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov [queue+8*r13], rbx\n";
data.code += "\tmov [queue+8*r13+8], rax\n";
data.code += "\tadd r13, 2\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
data.code += format!("\tjmp swp_{line}_{col}_end\n").as_str();
data.code += format!("swp_{line}_{col}_special:\n").as_str();
data.code += "\tmov qword rax, [queue+8*r12]\n";
data.code += "\tmov qword rdi, [queue+8*r12+8]\n";
data.code += "\tmov qword r12, r14\n";
data.code += "\tmov qword r13, r14\n";
data.code += "\tmov qword [queue+8*r12], rdi\n";
data.code += "\tmov qword [queue+8*r12+8], rax\n";
data.code += "\tadd r13, 2\n";
data.code += format!("swp_{line}_{col}_end:\n").as_str();
}
Operation::While(while_operations, line, col) =>
{
data.code += format!("\t;;while {}:{}\n", line, col).as_str();
data.code += "\tcmp qword [queue+8*r12], 0\n";
data.code += format!("\tje while_{}_{}_end\n", line, col).as_str();
data.code += format!("while_{}_{}:\n", line, col).as_str();
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
merge_assemblies(&mut data, generate_assembly_linux_x64_block(while_operations, functions, arrays));
data.code += "\tcmp qword [queue+8*r12], 0\n";
data.code += format!("\tjne while_{}_{}\n", line, col).as_str();
data.code += format!("while_{}_{}_end:\n", line, col).as_str();
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
}
Operation::If(if_operations, maybe_else_operations, line, col) =>
{
data.code += format!("\t;;if {}:{}\n", line, col).as_str();
data.code += "\tcmp qword [queue+8*r12], 0\n";
data.code += format!("\tje else_{}_{}\n", line, col).as_str();
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
merge_assemblies(&mut data, generate_assembly_linux_x64_block(if_operations, functions, arrays));
data.code += format!("\tjmp if_{}_{}_end\n", line, col).as_str();
data.code += format!("else_{}_{}:\n", line, col).as_str();
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
if let Some(else_operations) = maybe_else_operations
{
merge_assemblies(&mut data, generate_assembly_linux_x64_block(else_operations, functions, arrays));
}
data.code += format!("if_{}_{}_end:\n", line, col).as_str();
}
Operation::Dup(line, col) =>
{
data.code += format!("\t;;dup {}:{}\n", line, col).as_str();
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
Operation::Intrinsic(name, line, col) =>
{
data.code += format!("\t;;intrinsic {} {}:{}\n", name, line, col).as_str();
match name.as_str()
{
"print" =>
{
// For now printing numbers directly is unsupported
data.code += "\tmov rax, 1\n";
data.code += "\tmov rdi, 1\n";
// load address
data.code += "\tmov rsi, [queue+8*r12]\n";
// size
data.code += "\tmov rdx, [rsi]\n";
// data
data.code += "\tlea rsi, [rsi+8]\n";
// incorporate the null byte
//data.code += "\tinc rdx\n";
data.code += "\tsyscall\n";
// TODO: factor this out
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
}
"println" =>
{
// For now printing numbers directly is unsupported
data.code += "\tmov rax, 1\n";
data.code += "\tmov rdi, 1\n";
// load address
data.code += "\tmov rsi, [queue+8*r12]\n";
// size
data.code += "\tmov rdx, [rsi]\n";
// data
data.code += "\tlea rsi, [rsi+8]\n";
// incorporate the null byte
//data.code += "\tinc rdx\n";
data.code += "\tsyscall\n";
// TODO: factor this out
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
// TODO: Don't syscall twice
data.code += "\tmov rax, 1\n";
data.code += "\tlea rsi, [newline]\n";
data.code += "\tmov rdx, 1\n";
data.code += "\tsyscall\n";
}
"intToStr" =>
{
data.code += "\tmov qword rdi, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tcall intToStr\n";
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"-" =>
{
data.code += "\tmov qword rax, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += "\tmov qword rbx, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tsub rax, rbx\n";
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"+" =>
{
data.code += "\tmov qword rax, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += "\tmov qword rbx, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tadd rax, rbx\n";
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"*" =>
{
data.code += "\tmov qword rax, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += "\tmov qword rbx, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmul rbx\n";
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"divmod" =>
{
data.code += "\tmov qword rax, [queue+8*r12]\n";
data.code += "\tmov qword rdx, 0\n";
data.code += "\tinc r12\n";
data.code += "\tmov qword rbx, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tidiv rbx\n";
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
data.code += "\tmov [queue+8*r13], rdx\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
">" =>
{
data.code += "\tmov rbx, 0\n";
data.code += "\tmov rcx, 1\n";
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tcmp qword rax, [queue+8*r12+8]\n";
data.code += "\tcmovg rbx, rcx\n";
data.code += "\tadd r12, 2\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov qword [queue+8*r13], rbx\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"<" =>
{
data.code += "\tmov rbx, 0\n";
data.code += "\tmov rcx, 1\n";
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tcmp qword rax, [queue+8*r12+8]\n";
data.code += "\tcmovl rbx, rcx\n";
data.code += "\tadd r12, 2\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov qword [queue+8*r13], rbx\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
">=" =>
{
data.code += "\tmov rbx, 0\n";
data.code += "\tmov rcx, 1\n";
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tcmp qword rax, [queue+8*r12+8]\n";
data.code += "\tcmovge rbx, rcx\n";
data.code += "\tadd r12, 2\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov qword [queue+8*r13], rbx\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"<=" =>
{
data.code += "\tmov rbx, 0\n";
data.code += "\tmov rcx, 1\n";
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tcmp qword rax, [queue+8*r12+8]\n";
data.code += "\tcmovle rbx, rcx\n";
data.code += "\tadd r12, 2\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov qword [queue+8*r13], rbx\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"==" =>
{
data.code += "\tmov rbx, 0\n";
data.code += "\tmov rcx, 1\n";
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tcmp qword rax, [queue+8*r12+8]\n";
data.code += "\tcmove rbx, rcx\n";
data.code += "\tadd r12, 2\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov qword [queue+8*r13], rbx\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"!=" =>
{
data.code += "\tmov rbx, 0\n";
data.code += "\tmov rcx, 1\n";
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tcmp qword rax, [queue+8*r12+8]\n";
data.code += "\tcmovne rbx, rcx\n";
data.code += "\tadd r12, 2\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov qword [queue+8*r13], rbx\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"&&" =>
{
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tmov rbx, [queue+8*r12+8]\n";
data.code += "\tand rax, rbx\n";
data.code += "\tadd r12, 2\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
_ => todo!("intrinsic {} {}:{}", name, line, col)
}
}
Operation::Apply(name, word, line, col) =>
{
let array = arrays.iter().find(|x| &x.name == name).unwrap();
data.code += format!("\t;;apply {}.{} {}:{}\n", name, word, line, col).as_str();
match word.as_str()
{
"read" =>
{
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += generate_assembly_linux_x64_array_oob_check(array.length).as_str();
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += format!("\tmov qword rbx, [arr_{}+8*rax]\n", name).as_str();
data.code += "\tmov qword [queue+8*r13], rbx\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
"write" =>
{
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += generate_assembly_linux_x64_array_oob_check(array.length).as_str();
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov qword rbx, [queue+8*r12]\n";
data.code += format!("\tmov qword [arr_{}+8*rax], rbx\n", name).as_str();
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
}
"length" =>
{
data.code += format!("\tmov qword [queue+8*r13], {}\n", array.length).as_str();
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
_ => todo!("apply {}", word)
}
}
Operation::FunctionCall(name, line, col) =>
{
data.code += format!("\t;;func call {} {}:{}\n", name, line, col).as_str();
let function = functions.iter().find(|x| &x.name == name).unwrap();
for _ in 0..function.ins.len()
{
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tinc r12\n";
data.code += ASSEMBLY_LINUX_X64_TRY_RESET_QUEUE;
data.code += "\tmov [queue+8*r13], rax\n";
data.code += "\tinc r13\n";
data.code += generate_assembly_linux_x64_queue_oob_check().as_str();
}
data.code += "\t;; move pointers\n";
// save the current base
data.code += "\tpush r14\n";
// save the current head
data.code += "\tpush r12\n";
// prepare the layout
data.code += "\tmov r14, r13\n";
data.code += format!("\tsub r14, {}\n", function.ins.len()).as_str();
data.code += "\tmov r12, r14\n";
// call
data.code += format!("\tcall {}\n", name).as_str();
// move the sub-queue back to the base
for _ in 0..function.outs.len()
{
data.code += "\tmov rax, [queue+8*r12]\n";
data.code += "\tmov [queue+8*r14], rax\n";
data.code += "\tinc r12\n";
data.code += "\tinc r14\n";
}
// restore the tail
data.code += "\tmov r13, r14\n";
// restore the head
data.code += "\tpop r12\n";
// restore the base
data.code += "\tpop r14\n";
}
Operation::Interrupt(line, col) =>
{
data.code += format!("\t;;interrupt {}:{}\n", line, col).as_str();
data.code += "lea r8, [queue]\n";
data.code += format!("mov r9, {}\n", 1000*line + col).as_str();
data.code += "int3\n";
}
_ => todo!("{:?}", operation)
}
}
data
}
fn generate_assembly_linux_x64_function(name: &str, operations: &Vec<Operation>, functions: &Vec<Function>, arrays: &Vec<Arr>) -> AssemblyData
{
let mut data = AssemblyData
{
arrays: String::new(),
code: format!("{}:\n", name),
strings: String::new(),
};
merge_assemblies(&mut data, generate_assembly_linux_x64_block(operations, functions, arrays));
data.code += "\tret\n";
data
}
fn interpret_program(operations: &Vec<Operation>, queue: &mut Vec<String>, functions: &Vec<Function>, arrays: &mut Vec<Arr>, debug: bool) -> Result<String,String>
{
let mut output = String::new();
for operation in operations
{
if debug
{
println!("before: {:?}: {:?}, '{}'", operation, queue, output);
}
match operation
{
Operation::Dequeue(_, _) =>
{
queue.remove(0);
}
Operation::Enqueue(_, value, _, _) =>
{
queue.push(value.clone());
}
Operation::Requeue(_, _) =>
{
let val = queue.remove(0);
queue.push(val);
}
Operation::Dup(_, _) =>
{
let val = queue.first().unwrap();
queue.push(val.clone());
}
Operation::Swap(_, _) =>
{
let first = queue.remove(0);
let second = queue.remove(0);
queue.push(second);
queue.push(first);
}
Operation::FunctionCall(function_name, _, _) =>
{
let function = functions.iter().find(|x| &x.name == function_name).unwrap();
let function_context: &mut Vec<String> = &mut Vec::new();
for _ in 0..function.ins.len()
{
let val = queue.remove(0);
function_context.push(val);
}
output += interpret_program(&function.content, function_context, functions, arrays, debug)?.as_str();
for val in function_context
{
queue.push(val.to_string());
}
}
Operation::If(if_block, maybe_else_block, _, _) =>
{
let val = queue.remove(0);
if val == "true"
{
output += interpret_program(if_block, queue, functions, arrays, debug)?.as_str();
}
else if let Some(else_block) = maybe_else_block
{
output += interpret_program(else_block, queue, functions, arrays, debug)?.as_str();
}
}
Operation::Intrinsic(intrinsic_name, line, col) =>
{
match intrinsic_name.as_str()
{
"print" =>
{
output += queue.remove(0).to_string().as_str();
}
"-" =>
{
let minuend = queue.remove(0).parse::<i64>().unwrap();
let subtrahend = queue.remove(0).parse::<i64>().unwrap();
queue.push((minuend - subtrahend).to_string());
}
"+" =>
{
let addend1 = queue.remove(0).parse::<i64>().unwrap();
let addend2 = queue.remove(0).parse::<i64>().unwrap();
queue.push((addend1 + addend2).to_string());
}
"*" =>
{
let multiplicant1 = queue.remove(0).parse::<i64>().unwrap();
let multiplicant2 = queue.remove(0).parse::<i64>().unwrap();
queue.push((multiplicant1 * multiplicant2).to_string());
}
"divmod" =>
{
let dividend = queue.remove(0).parse::<i64>().unwrap();
let divisor = queue.remove(0).parse::<i64>().unwrap();
queue.push((dividend / divisor).to_string());
queue.push((dividend % divisor).to_string());
}
">" =>
{
let first = queue.remove(0).parse::<i64>().unwrap();
let second = queue.remove(0).parse::<i64>().unwrap();
queue.push((first > second).to_string());
}
"<" =>
{
let first = queue.remove(0).parse::<i64>().unwrap();
let second = queue.remove(0).parse::<i64>().unwrap();
queue.push((first < second).to_string());
}
">=" =>
{
let first = queue.remove(0).parse::<i64>().unwrap();
let second = queue.remove(0).parse::<i64>().unwrap();
queue.push((first >= second).to_string());
}
"<=" =>
{
let first = queue.remove(0).parse::<i64>().unwrap();
let second = queue.remove(0).parse::<i64>().unwrap();
queue.push((first <= second).to_string());
}
"==" =>
{
let first = queue.remove(0).parse::<i64>().unwrap();
let second = queue.remove(0).parse::<i64>().unwrap();
queue.push((first == second).to_string());
}
"!=" =>
{
let first = queue.remove(0).parse::<i64>().unwrap();
let second = queue.remove(0).parse::<i64>().unwrap();
queue.push((first != second).to_string());
}
"&&" =>
{
let first = queue.remove(0).parse::<bool>().unwrap();
let second = queue.remove(0).parse::<bool>().unwrap();
queue.push((first && second).to_string());
}
"println" =>
{
output += format!("{}\n", queue.remove(0)).as_str();
}
"intToStr" =>
{
let val = queue.remove(0).clone();
queue.push(val);
}
_ =>
{
return Err(format!("Unexpected intrinsic '{}' at {}:{}", intrinsic_name, line, col));
}
}
}
Operation::Apply(name, word, line, col) =>
{
let arr: &mut Arr = arrays.iter_mut().find(|x| &x.name == name).unwrap();
match word.as_str()
{
"write" =>
{
let position: i64 = queue.remove(0).parse::<i64>().unwrap();
if position >= arr.length
{
//return Err(format!("Attempted an out of bounds write for array {} ({} >= {}) at {}:{}", arr.name, position, arr.length, line, col));
return Err(String::from("Attempted array out-of-bounds access\n"));
}
if position < 0
{
//return Err(format!("Attempted an out of bounds write for array {} ({} < 0) at {}:{}", arr.name, position, line, col));
return Err(String::from("Attempted array out-of-bounds access\n"));
}
let data = queue.remove(0);
if debug
{
println!("write before: {} {} {:?}", position, data, arr);
}
arr.data[position as usize] = data;
if debug
{
println!("write after: {:?}", arr);
}
}
"read" =>
{
let position: i64 = queue.remove(0).parse::<i64>().unwrap();
if position >= arr.length
{
//return Err(format!("Attempted an out of bounds read for array {} ({} >= {}) at {}:{}", arr.name, position, arr.length, line, col));
return Err(String::from("Attempted array out-of-bounds access\n"));
}
if position < 0
{
//return Err(format!("Attempted an out of bounds read for array {} ({} < 0) at {}:{}", arr.name, position, line, col));
return Err(String::from("Attempted array out-of-bounds access\n"));
}
queue.push(arr.data[position as usize].clone());
}
"length" =>
{
queue.push(arr.length.to_string());
}
_ => return Err(format!("Unexpected application '{}' at {}:{}", word, line, col))
}
}
Operation::While(while_block, _, _) =>
{
loop
{
let val = queue.remove(0);
if val == "false"
{
break;
}
output += interpret_program(while_block, queue, functions, arrays, debug)?.as_str();
}
}
Operation::Depth(_, _) =>
{
let depth = queue.len();
queue.push(depth.to_string());
}
Operation::QueueDiagnostic(line, col) =>
{
println!("---Queue state at {}:{}---\nlength: {}\n{:?}\n------------------------------", line, col, queue.len(), queue);
}
Operation::Interrupt(_, _) => {}
}
if debug
{
println!("after: {:?}: {:?}, '{}'", operation, queue, output);
}
}
Ok(output)
}
fn typecheck(operations: &Vec<Operation>, functions: &Vec<Function>, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, arrays: &Vec<Arr>, debug: bool) -> Result<(), String>
{
for function in functions
{
if debug
{
println!("Now typechecking function '{}'", function.name);
}
typecheck_block(&function.content, &function.ins, &function.outs, functions, intrinsics, arrays, debug)?;
if debug
{
println!("Successfully typechecked function '{}'", function.name);
}
}
if debug
{
println!("Now typechecking main operations");
}
typecheck_block(operations, &Vec::new(), &Vec::new(), functions, intrinsics, arrays, debug)?;
if debug
{
println!("Successfully typechecked main operations");
}
Ok(())
}
fn typecheck_block(operations: &Vec<Operation>, ins: &[Datatype], outs: &Vec<Datatype>, functions: &Vec<Function>, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, arrays: &Vec<Arr>, debug: bool) -> Result<(), String>
{
let actual_outs = get_return_type(operations, ins, functions, intrinsics, arrays, debug)?;
if &actual_outs != outs
{
let (line, col) = match operations.last()
{
Some(operation) =>
{
match operation
{
Operation::Interrupt(line, col) |
Operation::Enqueue(_, _, line, col) |
Operation::Dequeue(line, col) |
Operation::Requeue(line, col) |
Operation::Dup(line, col) |
Operation::Swap(line, col) |
Operation::FunctionCall(_, line, col) |
Operation::If(_, _, line, col) |
Operation::Intrinsic(_, line, col) |
Operation::While(_, line, col) |
Operation::QueueDiagnostic(line, col) |
Operation::Apply(_, _, line, col) |
Operation::Depth(line, col) => (*line, *col),
}
}
None => (-1, -1)
};
return Err(format!("Wrong queue state at the end of a block, expected {:?} but got {:?} at {}:{}", outs, actual_outs, line, col));
}
Ok(())
}
fn get_return_type(operations: &Vec<Operation>, ins: &[Datatype], functions: &Vec<Function>, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, arrays: &Vec<Arr>, debug: bool) -> Result<Vec<Datatype>, String>
{
let type_queue: &mut Vec<Datatype> = &mut Vec::new();
type_queue.extend_from_slice(ins);
let mut debug_string = String::from("");
for operation in operations
{
if debug
{
debug_string = format!("operation: {:?}: {:?}", operation, type_queue);
}
match operation
{
Operation::Dequeue(line, col) =>
{
if type_queue.is_empty()
{
return Err(format!("Attempted to dequeue an element while the queue was empty at {}:{}", line, col));
}
type_queue.remove(0);
}
Operation::Enqueue(datatype, _, _, _) =>
{
type_queue.push(*datatype);
}
Operation::Dup(line, col) =>
{
if let Some(typ) = type_queue.first()
{
type_queue.push(*typ);
}
else
{
return Err(format!("Attempted to dup an element while the queue was empty at {}:{}", line, col));
}
}
Operation::Requeue(line, col) =>
{
if type_queue.is_empty()
{
return Err(format!("Attempted to requeue an element while the queue was empty at {}:{}", line, col));
}
let typ = type_queue.remove(0);
type_queue.push(typ);
}
Operation::Swap(line, col) =>
{
if type_queue.is_empty()
{
return Err(format!("Attempted to get the first element for a swap while the queue was empty at {}:{}", line, col));
}
let first_typ = type_queue.remove(0);
if type_queue.is_empty()
{
return Err(format!("Attempted to get the second element for a swap while the queue was empty at {}:{}", line, col));
}
let second_typ = type_queue.remove(0);
type_queue.push(second_typ);
type_queue.push(first_typ);
}
Operation::FunctionCall(function_name, line, col) =>
{
let function = functions.iter().find(|x| &x.name == function_name).unwrap();
if function.ins.len() > type_queue.len()
{
return Err(format!("Attempted to call function '{}' at {}:{}, with insufficient elements in the queue, expected {:?} but got {:?}", function.name, line, col, function.ins, type_queue));
}
for in_type in &function.ins
{
let actual_type = type_queue.remove(0);
if in_type != &actual_type
{
return Err(format!("Attempted to call function '{}' at {}:{} with a wrong parameter, expected {:?} but got {:?}", function.name, line, col, in_type, actual_type));
}
}
type_queue.extend_from_slice(&function.outs);
}
Operation::If(if_block, maybe_else_block, line, col) =>
{
if type_queue.is_empty()
{
return Err(format!("Encountered if block with an empty queue at {}:{}", line, col));
}
let comparison_type = type_queue.remove(0);
if comparison_type != Datatype::Bool
{
return Err(format!("Expected a Bool as an if condition but got {:?} instead at {}:{}", comparison_type, line, col));
}
if debug
{
println!("Starting to typecheck if block");
}
let if_ret = get_return_type(if_block, type_queue, functions, intrinsics, arrays, debug)?;
let else_ret =
if let Some(else_block) = maybe_else_block
{
if debug
{
println!("Starting to typecheck else block");
}
get_return_type(else_block, type_queue, functions, intrinsics, arrays, debug)?
}
else
{
type_queue.clone()
};
if if_ret != else_ret
{
return Err(format!("Incompatible queue states after if/else construction, expected {:?} but got {:?}", if_ret, else_ret));
}
type_queue.clear();
type_queue.extend_from_slice(&if_ret);
}
Operation::Intrinsic(intrinsic_name, line, col) =>
{
let io = intrinsics.get(intrinsic_name.as_str()).unwrap();
if io.0.len() > type_queue.len()
{
return Err(format!("Attempted to call intrinsic '{}' at {}:{}, with insufficient elements in the queue, expected {:?} but got {:?}", intrinsic_name, line, col, io.0, type_queue));
}
for in_type in &io.0
{
let actual_type = type_queue.remove(0);
if in_type != &actual_type
{
return Err(format!("Attempted to call intrinsic '{}' at {}:{} with a wrong parameter, expected {:?} but got {:?}", intrinsic_name, line, col, in_type, actual_type));
}
}
type_queue.extend_from_slice(&io.1);
}
Operation::While(while_block, line, col) =>
{
if type_queue.is_empty()
{
return Err(format!("Encountered while block with an empty queue at {}:{}", line, col));
}
let comparison_type = type_queue.remove(0);
if comparison_type != Datatype::Bool
{
return Err(format!("Expected a Bool as a while condition but got {:?} instead at {}:{}", comparison_type, line, col));
}
if debug
{
println!("Starting to typecheck while block");
}
let mut outs = type_queue.clone();
outs.insert(0, Datatype::Bool);
typecheck_block(while_block, type_queue, &outs, functions, intrinsics, arrays, debug)?;
}
Operation::Depth(_, _) =>
{
type_queue.push(Datatype::Int);
}
Operation::QueueDiagnostic(line, col) =>
{
println!("---Type queue state at {}:{}---\nlength: {}\n{:?}\n------------------------------", line, col, type_queue.len(), type_queue);
}
Operation::Interrupt(_, _) => {}
Operation::Apply(name, word, line, col) =>
{
match word.as_str()
{
"write" =>
{
if type_queue.is_empty() || type_queue.remove(0) != Datatype::Int
{
return Err(format!("Expected a position for a write application at {}:{}", line, col));
}
let expected_type = arrays.iter().find(|x| &x.name == name).unwrap().datatype;
if type_queue.is_empty()
{
return Err(format!("Expected data for a write application at {}:{}", line, col));
}
let actual_type = type_queue.remove(0);
if actual_type != expected_type
{
return Err(format!("Expected a {:?} value but got a {:?} value at {}:{}", expected_type, actual_type, line, col));
}
}
"read" =>
{
if type_queue.is_empty() || type_queue.remove(0) != Datatype::Int
{
return Err(format!("Expected a position for a read application at {}:{}", line, col));
}
let typ = arrays.iter().find(|x| &x.name == name).unwrap().datatype;
type_queue.push(typ);
}
"length" =>
{
type_queue.push(Datatype::Int);
}
_ => return Err(format!("Encountered unknown application '{}' at {}:{}", word, line, col))
}
}
}
if debug
{
println!("{} => {:?}", debug_string, type_queue);
}
}
Ok(type_queue.clone())
}
fn validate_function_calls(operations: &Vec<Operation>, functions: &Vec<Function>, debug: bool) -> Result<(), String>
{
for function in functions
{
validate_function_calls_in_block(&function.content, functions)?;
if debug
{
println!("Successfully validated function calls in function '{}'", function.name);
}
}
validate_function_calls_in_block(operations, functions)?;
if debug
{
println!("Successfully validated function calls in main operations");
}
Ok(())
}
fn validate_function_calls_in_block(block: &Vec<Operation>, functions: &Vec<Function>) -> Result<(), String>
{
for operation in block
{
match operation
{
Operation::Depth(_, _) | Operation::QueueDiagnostic(_, _) | Operation::Intrinsic(_, _, _) | Operation::Enqueue(_, _, _, _) | Operation::Dequeue(_, _) |
Operation::Requeue(_, _) | Operation::Dup(_, _) | Operation::Swap(_, _) | Operation::Apply(_, _, _, _) | Operation::Interrupt(_, _) => {},
Operation::FunctionCall(function_name, line, col) =>
{
if !functions.iter().any(|x| &x.name == function_name)
{
return Err(format!("Call to unknown function '{}' at {}:{}", function_name, line, col));
}
}
Operation::If(if_block, maybe_else_block, _, _) =>
{
validate_function_calls_in_block(if_block, functions)?;
if let Some(else_block) = maybe_else_block
{
validate_function_calls_in_block(else_block, functions)?;
}
}
Operation::While(while_block, _, _) =>
{
validate_function_calls_in_block(while_block, functions)?;
}
}
}
Ok(())
}
fn extract_arrays(tokens: &mut Vec<Token>, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, functions: &[Function], debug: bool) -> Result<Vec<Arr>, String>
{
let mut tokens_iter = tokens.iter().peekable();
let mut arrays: Vec<Arr> = Vec::new();
let mut new_tokens: Vec<Token> = Vec::new();
while let Some(token) = tokens_iter.next()
{
if let Token::Keyword(word, line, col) = token
{
if word == "arr"
{
if debug
{
println!("Found an array at {}:{}", line, col);
}
if let Some(Token::Keyword(name, _, _)) = tokens_iter.next()
{
if functions.iter().any(|x| &x.name == name)
{
return Err(format!("Cannot redeclare an array with the same name as a function {}:{}", line, col));
}
if arrays.iter().any(|x| &x.name == name)
{
return Err(format!("Cannot redeclare an array with the same name as an array {}:{}", line, col));
}
if intrinsics.contains_key(name.as_str())
{
return Err(format!("An array cannot have the same name as an intrinsic ({}) at {}:{}", name, line, col));
}
if let Some(Token::Keyword(open_curly, _, _)) = tokens_iter.next()
{
if open_curly != "{"
{
return Err(format!("Expected '{{' in array declaration at {}:{}", line, col));
}
}
else
{
return Err(format!("Reached the end of the file while parsing an array at {}:{}", line, col));
}
if let Some(Token::Keyword(typ, _, _)) = tokens_iter.next()
{
let datatype = str_to_datatype(typ, *line, *col)?;
if let Some(Token::IntLit(size_str, _, _)) = tokens_iter.next()
{
let size = size_str.parse::<i64>().unwrap();
if let Some(Token::Keyword(close_curly, _, _)) = tokens_iter.next()
{
if close_curly != "}"
{
return Err(format!("Expected '}}' in array declaration at {}:{}", line, col));
}
}
else
{
return Err(format!("Reached the end of the file while parsing an array at {}:{}", line, col));
}
let mut data: Vec<String> = Vec::new();
let default_val = match datatype
{
Datatype::String => String::new(),
Datatype::Bool => String::from("false"),
Datatype::Int => String::from("0"),
};
for _ in 0..size
{
data.push(default_val.clone());
}
arrays.push(Arr { name: sanitize_name(name.clone()), datatype, length: size , data });
}
}
else
{
return Err(format!("Reached the end of the file while parsing an array at {}:{}", line, col))
}
}
else
{
return Err(format!("Expected array name, at {}:{}", line, col));
}
}
else
{
new_tokens.push(token.clone());
}
}
else
{
new_tokens.push(token.clone());
}
}
tokens.clear();
tokens.extend_from_slice(&new_tokens);
Ok(arrays)
}
fn sanitize_name(name: String) -> String
{
if name == "test"
{
return "test_".to_string();
}
name.replace(['-', '+'], "_").replace('%', "percent").replace('/', "slash")
}
fn str_to_datatype(s: &str, line: i32, col: i32) -> Result<Datatype, String>
{
match s
{
//"any" => Ok(Datatype::Any),
"bool" => Ok(Datatype::Bool),
"int" => Ok(Datatype::Int),
"str" => Ok(Datatype::String),
_ => Err(format!("Expected a datatype for the array, got {} instead at {}:{}", s, line, col))
}
}
fn extract_functions(tokens: &mut Vec<Token>, functions: &mut Vec<Function>, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, debug: bool) -> Result<(), String>
{
let mut tokens_iter = tokens.iter().peekable();
let mut new_tokens: Vec<Token> = Vec::new();
while let Some(token) = tokens_iter.next()
{
if let Token::Keyword(word, line, col) = token
{
if word == "function"
{
if debug
{
println!("Found a function at {}:{}", line, col);
}
let mut ins: Vec<Datatype> = Vec::new();
loop
{
let maybe_token = tokens_iter.next();
match maybe_token
{
Some(token) =>
{
match token
{
Token::IntLit(_, line, col) | Token::StringLit(_, line, col) | Token::BoolLit(_, line, col) | Token::Import(line, col) |
Token::Apply(_, _, line, col) =>
{
return Err(format!("Expected input parameters for a function but got {:?} instead at {}:{}", token, line, col));
}
Token::Keyword(word, line, col) =>
{
if word == "=>"
{
break;
}
match word.as_str()
{
//"any" => ins.push(Datatype::Any),
"str" => ins.push(Datatype::String),
"int" => ins.push(Datatype::Int),
"bool" => ins.push(Datatype::Bool),
_ => return Err(format!("Expected input parameters for a function but got {} instead at {}:{}", word, line, col))
}
}
}
}
None => return Err("Unexpected end of file while extracting a function".to_string())
}
}
if debug
{
println!("ins: {:?}", ins);
}
let mut outs: Vec<Datatype> = Vec::new();
loop
{
let maybe_token = tokens_iter.next();
match maybe_token
{
Some(token) =>
{
match token
{
Token::IntLit(_, line, col) | Token::StringLit(_, line, col) | Token::BoolLit(_, line, col) | Token::Import(line, col) |
Token::Apply(_, _, line, col) =>
{
return Err(format!("Expected input parameters for a function but got {:?} instead at {}:{}", token, line, col));
}
Token::Keyword(word, line, col) =>
{
match word.as_str()
{
//"any" => outs.push(Datatype::Any),
"str" => outs.push(Datatype::String),
"int" => outs.push(Datatype::Int),
"bool" => outs.push(Datatype::Bool),
"{" | "}" | "deq" | "req" | "dup" | "swp" | "true" | "false" | "depth" | "???" | "import" => return Err(format!("Expected function name but got {} at {}:{}", word, line, col)),
_ =>
{
if functions.iter().any(|x| &x.name == word)
{
return Err(format!("Redeclaration of function '{}' at {}:{}", word, line, col));
}
if intrinsics.contains_key(word.as_str())
{
return Err(format!("Function name {} at {}:{} is already an intrinsic", word, line, col));
}
if debug
{
println!("outs: {:?}", outs);
}
let block = parse_block(&mut tokens_iter, intrinsics, debug)?;
functions.push(Function {name: sanitize_name(word.clone()), ins, outs, content: block});
break;
}
}
}
}
}
None => return Err("Unexpected end of file while extracting a function".to_string())
}
}
}
else
{
new_tokens.push(token.clone());
}
}
else
{
new_tokens.push(token.clone());
}
}
tokens.clear();
tokens.extend_from_slice(&new_tokens);
Ok(())
}
fn parse_block(tokens_iter: &mut Peekable<std::slice::Iter<Token>>, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, debug: bool) -> Result<Vec<Operation>, String>
{
if let Some(Token::Keyword(word, line, col)) = tokens_iter.next()
{
if word != "{"
{
return Err(format!("Expected '{{' to open a block but got {} at {}:{}", word, line, col));
}
}
else
{
return Err("Expected '{' to open a block".to_string());
}
parse_until_delimiter(tokens_iter, intrinsics, Some("}"), debug)
}
fn parse_until_delimiter(tokens_iter: &mut Peekable<std::slice::Iter<Token>>, intrinsics: &HashMap<&str, (Vec<Datatype>, Vec<Datatype>)>, delimiter: Option<&str>, debug: bool) -> Result<Vec<Operation>, String>
{
let mut operations: Vec<Operation> = Vec::new();
loop
{
let maybe_token = tokens_iter.next();
match maybe_token
{
Some(token) =>
{
match token
{
Token::IntLit(value, line, col) =>
{
operations.push(Operation::Enqueue(Datatype::Int, value.clone(), *line, *col));
}
Token::StringLit(value, line, col) =>
{
operations.push(Operation::Enqueue(Datatype::String, value.clone(), *line, *col));
}
Token::BoolLit(value, line, col) =>
{
operations.push(Operation::Enqueue(Datatype::Bool, value.clone(), *line, *col));
}
Token::Apply(name, word, line, col) =>
{
operations.push(Operation::Apply(sanitize_name(name.clone()), word.clone(), *line, *col));
}
Token::Import(line, col) =>
{
return Err(format!("Unexpected import token at {}:{}, should have been resolved before, probably a compiler bug", line, col));
}
Token::Keyword(word, line, col) =>
{
if intrinsics.contains_key(word.as_str())
{
operations.push(Operation::Intrinsic(word.clone(), *line, *col));
}
else if word == "if"
{
let block = parse_block(tokens_iter, intrinsics, debug)?;
let else_block =
if let Some(Token::Keyword(maybe_else, _, _)) = tokens_iter.peek()
{
if maybe_else == "else"
{
tokens_iter.next();
Some(parse_block(tokens_iter, intrinsics, debug)?)
}
else
{
None
}
}
else
{
None
};
operations.push(Operation::If(block, else_block, *line, *col));
}
else if word == "while"
{
operations.push(Operation::While(parse_block(tokens_iter, intrinsics, debug)?, *line, *col));
}
else if word == "deq"
{
operations.push(Operation::Dequeue(*line, *col));
}
else if word == "req"
{
operations.push(Operation::Requeue(*line, *col));
}
else if word == "dup"
{
operations.push(Operation::Dup(*line, *col));
}
else if word == "swp"
{
operations.push(Operation::Swap(*line, *col));
}
else if word == "depth"
{
operations.push(Operation::Depth(*line, *col));
}
else if word == "???"
{
operations.push(Operation::QueueDiagnostic(*line, *col));
}
else if word == "interrupt"
{
operations.push(Operation::Interrupt(*line, *col));
}
else if Some(word.as_str()) == delimiter
{
return Ok(operations);
}
else if word == "{" || word == "function"
{
return Err(format!("Unexpected keyword {} at {}:{}", word, line, col));
}
else
{
operations.push(Operation::FunctionCall(sanitize_name(word.clone()), *line, *col));
}
}
}
}
None =>
{
if delimiter.is_some()
{
return Err("Reached the end of the file while parsing a block".to_string());
}
else
{
return Ok(operations);
}
}
}
}
}
fn usage()
{
println!("Usage: kurz -c path/to/file");
exit(0);
}
fn tokenize(text: &str) -> Result<Vec<Token>, String>
{
let mut tokens: Vec<Token> = Vec::new();
let mut line = 1;
let mut col = 1;
let mut state = TokenizerState::Whitespace;
let mut word = String::new();
let mut iter = text.chars().peekable();
let mut application_name = String::new();
while let Some(ch) = iter.next()
{
if ch == '/' && iter.peek() == Some(&'/')
{
state = TokenizerState::Comment;
}
match state
{
TokenizerState::Comment =>
{
if ch == '\n'
{
state = TokenizerState::Whitespace;
}
}
TokenizerState::Whitespace =>
{
// If ch is whitespace, do nothing
if !ch.is_whitespace()
{
match ch
{
'"' =>
{
state = TokenizerState::Quote;
}
_ =>
{
state = TokenizerState::Keyword;
word.push(ch);
}
}
}
}
TokenizerState::Quote =>
{
if ch == '"'
{
state = TokenizerState::Whitespace;
tokens.push(Token::StringLit(word.clone().replace("\\n", "\n"), line, col));
word.clear();
}
else
{
word.push(ch);
}
}
TokenizerState::Keyword =>
{
if ch.is_whitespace()
{
state = TokenizerState::Whitespace;
if application_name.is_empty()
{
if word.parse::<i64>().is_ok()
{
tokens.push(Token::IntLit(word.clone(), line, col));
}
else if word == "true" || word == "false"
{
tokens.push(Token::BoolLit(word.clone(), line, col));
}
else if word == "import"
{
tokens.push(Token::Import(line, col));
}
else
{
tokens.push(Token::Keyword(word.clone(), line, col));
}
}
else
{
tokens.push(Token::Apply(sanitize_name(application_name.clone()), word.clone(), line, col));
application_name.clear();
}
word.clear();
}
else
{
match ch
{
'"' => return Err("Having '\"' in the middle of a word is not allowed".to_string()),
'.' =>
{
application_name = word.clone();
word.clear();
}
_ =>
{
word.push(ch);
}
}
}
}
}
col += 1;
if ch == '\n'
{
col = 1;
line += 1;
}
}
match state
{
TokenizerState::Quote =>
{
return Err("Encountered EOF before closing string".to_string());
}
TokenizerState::Whitespace | TokenizerState::Comment => {},
TokenizerState::Keyword =>
{
if application_name.is_empty()
{
tokens.push(Token::Keyword(word.clone(), line, col));
}
else
{
tokens.push(Token::Apply(sanitize_name(application_name.clone()), word.clone(), line, col));
}
}
}
Ok(tokens)
}

30
std.qbl Normal file
View File

@ -0,0 +1,30 @@
function bool => str boolToStr
{
if
{
"true"
}
else
{
"false"
}
}
function int int => int %
{
divmod deq
}
function int int => int /
{
divmod req deq
}
function int => int decrement
{
1 -
}
function int => int increment
{
1 +
}

View File

@ -0,0 +1,6 @@
//invalid,Attempted array out-of-bounds access
//:END:
arr test { bool 4 }
5 test.read deq

View File

@ -0,0 +1,6 @@
//invalid,Attempted array out-of-bounds access
//:END:
arr test { bool 4 }
5 false test.write

View File

@ -0,0 +1,6 @@
//invalid,Attempted array out-of-bounds access
//:END:
arr test { bool 4 }
-1 test.read deq

View File

@ -0,0 +1,6 @@
//invalid,Attempted array out-of-bounds access
//:END:
arr test { bool 4 }
-1 true test.write

26
tests/array_tests.qbl Normal file
View File

@ -0,0 +1,26 @@
//valid,0000
//0100
//2100
//2103
//:END:
arr test { int 4 }
function => dump
{
true 0
while
{
// i
dup test.read req intToStr req print
1 + test.length dup < req
}
deq "" println
}
dump
1 1 test.write
dump
0 2 test.write
dump
3 3 test.write
dump

7
tests/basic_import.qbl Normal file
View File

@ -0,0 +1,7 @@
//valid,true
//:END:
import "../std.qbl"
true boolToStr println

26
tests/comparisons.qbl Normal file
View File

@ -0,0 +1,26 @@
//valid,true
//false
//true
//false
//false
//
//false
//false
//true
//true
//true
//:END:
import "../std.qbl"
1 0 > boolToStr println
1 0 < boolToStr println
1 0 >= boolToStr println
1 0 <= boolToStr println
1 0 == boolToStr println
"" println
1 1 > boolToStr println
1 1 < boolToStr println
1 1 >= boolToStr println
1 1 <= boolToStr println
1 1 == boolToStr println

355
tests/conway.qbl Normal file
View File

@ -0,0 +1,355 @@
//valid,.#......
//..#.....
//###.....
//........
//........
//........
//........
//........
//--------
//........
//#.#.....
//.##.....
//.#......
//........
//........
//........
//........
//--------
//........
//..#.....
//#.#.....
//.##.....
//........
//........
//........
//........
//--------
//........
//.#......
//..##....
//.##.....
//........
//........
//........
//........
//--------
//........
//..#.....
//...#....
//.###....
//........
//........
//........
//........
//--------
//:END:
arr field { int 64 }
arr field2 { int 64 }
1 0 coordToIndex 1 field.write
2 1 coordToIndex 1 field.write
0 2 coordToIndex 1 field.write
1 2 coordToIndex 1 field.write
2 2 coordToIndex 1 field.write
printArray
true 3
while
{
// i
iteration
0 dup 1 > -
}
deq
function => iteration
{
true 0
while
{
// y
true 0 req
// true x y
while
{
// x y
dup swp req
// y x x
dup swp
// x y x y
neighbourhood
// x y c
dup swp req
// x y x c
req dup swp req
// x y x y c
coordToIndex req req req
// i x y c
dup field.read
// x y c i v
req req req swp
// x y c v i
req req
// c v i x y
update
// x y
check
req
// lt x+1 y
}
// x y
deq check
// lt y+1
}
deq
copyArrays
printArray
}
function => copyArrays
{
true 0
while
{
// 0
dup field2.read
dup field.write
checkArr
}
deq
}
function => printArray
{
true 0
while
{
// y
true 0 req
// true x y
while
{
// x y
dup req dup swp
// x y x y
coordToIndex
// x y i
req req field.read
// x y v
req req printSym
// x y
check req
// lt x+1 y
}
// x y
"" deq req println check
// lt y+1
}
deq
"--------" println
}
function int => printSym
{
// x
1 ==
// eq
if
{
"#"
}
else
{
"."
}
print
}
function int => bool int checkArr
{
// i
1 + field.length dup
// i+1 l i+1
< req
// lt i+1
}
function int => bool int check
{
//???
// i
1 + 8 dup < req
//???
// i+1 8 i+1
}
function int int int => update
{
// count value index
3 dup req req req
// 3 c c v i
== req req req
// eq c v i
if
{
1 deq deq field2.write
}
else
{
// c v i
2 req req req
// 2 c v i
== 1 req req req
// 1 v i eq
== req && req
// teq i
if
{
1
}
else
{
0
}
field2.write
}
}
function int int => int neighbourhood
{
// x y
dup 1 req req -
// x y x-1
0 req req dup >=
// x y x-1 ge
req req req
// ge x y x-1
if
{
req dup swp
// x y x-1 y
req req
// x-1 y x y
row
// x y abc
}
else
{
// x y x-1
req req
// x-1 x y
deq 0
// x y 0
}
dup swp req req
// y x abc x
dup swp req row
// x y abc def
req req +
// x y abcdef
dup swp req req
// y x abcdef x
dup swp req
// x y x y abcdef
coordToIndex
// x y abcdef i
req req req
// i x y abcdef
field.read
// x y abcdef v
req req -
// x y abcdef-v
1 req req req
// 1 x y abcdef
+ req req
// x+1 y abcdef
dup 8 req req req
// x+1 l x+1 y abcdef
< req req req
// lt x+1 y abcdef
if
{
row
// abcdef ghi
}
else
{
deq deq 0
// abcdef 0
}
+
// abcdefghi
}
function int int => int row
{
// x y
1 req dup -
// x y y-1
dup req req
// y-1 x x y
0 dup swp req req
// 0 y-1 x y-1 x y
> req req req req
// gt x y-1 x y
if
{
deq deq 0
// x y 0
}
else
{
// x y-1 x y
coordToIndex req req
// i x y
field.read
// x y a
}
// x y a
dup swp req req
// y x a x
dup swp req
// x y x y a
coordToIndex
// x y a i
req req req field.read
// x y a b
req req +
// x y a+b
req 1 req
// a+b x 1 y
req req + req
// x y+1 a+b
req req 8
// a+b x y+1 l
req req dup
// y+1 l a+b x y+1
< req req req
// lt a+b x y+1
if
{
// a+b x y+1
req coordToIndex
// a+b i
req field.read
// a+b c
}
else
{
req deq deq 0
// a+b 0
}
+
// a+b+c
}
function int int => int coordToIndex
{
req 8 req * +
}

374
tests/conway_diehard.qbl Normal file
View File

@ -0,0 +1,374 @@
//valid,.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//...................#.........
//.............##..............
//..............#...###........
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//--------
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//.............................
//--------
//:END:
arr field { int 841 }
arr field2 { int 841 }
13 9 coordToIndex 1 field.write
14 9 coordToIndex 1 field.write
14 10 coordToIndex 1 field.write
19 8 coordToIndex 1 field.write
18 10 coordToIndex 1 field.write
19 10 coordToIndex 1 field.write
20 10 coordToIndex 1 field.write
printArray
true 129
while
{
// i
iteration
0 dup 1 > -
}
deq
printArray
function => iteration
{
true 0
while
{
// y
true 0 req
// true x y
while
{
// x y
dup swp req
// y x x
dup swp
// x y x y
neighbourhood
// x y c
dup swp req
// x y x c
req dup swp req
// x y x y c
coordToIndex req req req
// i x y c
dup field.read
// x y c i v
req req req swp
// x y c v i
req req
// c v i x y
update
// x y
check
req
// lt x+1 y
}
// x y
deq check
// lt y+1
}
deq
copyArrays
}
function => copyArrays
{
true 0
while
{
// 0
dup field2.read
dup field.write
checkArr
}
deq
}
function => printArray
{
true 0
while
{
// y
true 0 req
// true x y
while
{
// x y
dup req dup swp
// x y x y
coordToIndex
// x y i
req req field.read
// x y v
req req printSym
// x y
check req
// lt x+1 y
}
// x y
"" deq req println check
// lt y+1
}
deq
"--------" println
}
function int => printSym
{
// x
1 ==
// eq
if
{
"#"
}
else
{
"."
}
print
}
function int => bool int checkArr
{
// i
1 + field.length dup
// i+1 l i+1
< req
// lt i+1
}
function int => bool int check
{
//???
// i
1 + 29 dup < req
//???
// i+1 29 i+1
}
function int int int => update
{
// count value index
3 dup req req req
// 3 c c v i
== req req req
// eq c v i
if
{
1 deq deq field2.write
}
else
{
// c v i
2 req req req
// 2 c v i
== 1 req req req
// 1 v i eq
== req && req
// teq i
if
{
1
}
else
{
0
}
field2.write
}
}
function int int => int neighbourhood
{
// x y
dup 1 req req -
// x y x-1
0 req req dup >=
// x y x-1 ge
req req req
// ge x y x-1
if
{
req dup swp
// x y x-1 y
req req
// x-1 y x y
row
// x y abc
}
else
{
// x y x-1
req req
// x-1 x y
deq 0
// x y 0
}
dup swp req req
// y x abc x
dup swp req row
// x y abc def
req req +
// x y abcdef
dup swp req req
// y x abcdef x
dup swp req
// x y x y abcdef
coordToIndex
// x y abcdef i
req req req
// i x y abcdef
field.read
// x y abcdef v
req req -
// x y abcdef-v
1 req req req
// 1 x y abcdef
+ req req
// x+1 y abcdef
dup 29 req req req
// x+1 l x+1 y abcdef
< req req req
// lt x+1 y abcdef
if
{
row
// abcdef ghi
}
else
{
deq deq 0
// abcdef 0
}
+
// abcdefghi
}
function int int => int row
{
// x y
1 req dup -
// x y y-1
dup req req
// y-1 x x y
0 dup swp req req
// 0 y-1 x y-1 x y
> req req req req
// gt x y-1 x y
if
{
deq deq 0
// x y 0
}
else
{
// x y-1 x y
coordToIndex req req
// i x y
field.read
// x y a
}
// x y a
dup swp req req
// y x a x
dup swp req
// x y x y a
coordToIndex
// x y a i
req req req field.read
// x y a b
req req +
// x y a+b
req 1 req
// a+b x 1 y
req req + req
// x y+1 a+b
req req 29
// a+b x y+1 l
req req dup
// y+1 l a+b x y+1
< req req req
// lt a+b x y+1
if
{
// a+b x y+1
req coordToIndex
// a+b i
req field.read
// a+b c
}
else
{
req deq deq 0
// a+b 0
}
+
// a+b+c
}
function int int => int coordToIndex
{
// y x
req 29 req * +
}

View File

@ -0,0 +1,8 @@
//invalid,Function name print at 3:22 is already an intrinsic:END:
function str => print
{
deq
}
"42" print

View File

@ -0,0 +1,8 @@
//invalid,Expected function name but got deq at 3:20:END:
function str => deq
{
deq
}
"42" print

View File

@ -0,0 +1,8 @@
//invalid,Expected function name but got { at 4:2:END:
function str =>
{
deq
}
"42" print

17
tests/recursion.qbl Normal file
View File

@ -0,0 +1,17 @@
//valid,10946
//:END:
function int int int int => int fibonacci
{
dup 0 == req req req req
if
{
req deq deq deq
}
else
{
dup + req 1 req swp - req req req fibonacci
}
}
20 0 1 0 fibonacci intToStr println

View File

@ -0,0 +1,8 @@
//valid,:END:
import "recursive_import_2.qbl"
import "recursive_import.qbl"
function => test
{
"test" println
}

View File

@ -0,0 +1,5 @@
//valid,test
//:END:
import "recursive_import.qbl"
test

8
tests/req_impl.qbl Normal file
View File

@ -0,0 +1,8 @@
//valid,231
//231
//:END:
function int => int req_impl { }
1 2 3 req_impl intToStr intToStr intToStr print print println
1 2 3 req intToStr intToStr intToStr print print println

159
tests/rule110.qbl Normal file
View File

@ -0,0 +1,159 @@
//valid,---------------------------------#
//--------------------------------##
//-------------------------------###
//------------------------------##-#
//-----------------------------#####
//----------------------------##---#
//---------------------------###--##
//--------------------------##-#-###
//-------------------------#######-#
//------------------------##-----###
//-----------------------###----##-#
//----------------------##-#---#####
//---------------------#####--##---#
//--------------------##---#-###--##
//-------------------###--####-#-###
//------------------##-#-##--#####-#
//-----------------########-##---###
//----------------##------####--##-#
//---------------###-----##--#-#####
//--------------##-#----###-####---#
//-------------#####---##-###--#--##
//------------##---#--#####-#-##-###
//-----------###--##-##---########-#
//----------##-#-######--##------###
//---------#######----#-###-----##-#
//--------##-----#---####-#----#####
//-------###----##--##--###---##---#
//------##-#---###-###-##-#--###--##
//-----#####--##-###-######-##-#-###
//----##---#-#####-###----########-#
//---###--####---###-#---##------###
//--##-#-##--#--##-###--###-----##-#
//-########-##-#####-#-##-#----#####
//##------######---########---##---#
//##-----##----#--##------#--###--##
//:END:
function bool bool bool => bool rule110
{
if
{
if
{
if
{
false
}
else
{
true
}
}
}
else
{
if
{
deq true
}
}
}
arr val { bool 34 }
33 true val.write
printArrays
function int => bool int check
{
0 dup 1 > -
}
function => fullApply
{
// 0 b c
0 1 false val.read val.read
0 rule110 val2.write
val.length 1 - 1 dup - req
val.read val.read false
rule110 val.length 1
req - req val2.write
true 1
while
{
// i-1 i
1 dup - req
// i a | i a i | a i b | i b a i 1
val.read dup val.read req dup 1
// i a b i+1 | i+1 i a b | i a b c
req swp + req req req val.read
// i d
req rule110
// i d i | i
dup val2.write
checkUp-1
}
deq
}
function int => bool int checkUp
{
// i+1 l i+1
1 + val.length dup < req
}
function int => bool int checkUp-1
{
// i+1 l 1 | l 1 i+1 | i+1 l-1 i+1
1 + val.length 1 req - dup
// i+1 b
< req
}
arr val2 { bool 34 }
function => copyArrays
{
true 0
while
{
dup val2.read
dup val.write
checkUp
}
deq
}
function bool => str boolToSym
{
if
{
"#"
}
else
{
"-"
}
}
function => printArrays
{
true 0
while
{
dup val.read req boolToSym req print
checkUp
}
deq
"" println
}
true 33
while
{
fullApply
copyArrays
printArrays
check
}
deq

42
tests/test.qbl Normal file
View File

@ -0,0 +1,42 @@
//valid,Hello, World!
//4242test2Falsetesttesttest:END:
"Hello, World!" println 43 foo foo deq
// Dequeues, enqueues 42 and 17, prints the head
function int => int foo
{
deq 42 17 intToStr req print
}
"test2" print false
check
print
function bool => str check
{
if
{
"True"
}
else
{
"False"
}
}
function int => whileFunction
{
dup
0 req >
req while
{
1 - 0 dup >
"test" req req print req
}
deq
}
3 whileFunction

View File

@ -0,0 +1,5 @@
//valid,false
//:END:
import "basic_import.qbl"
false boolToStr println

View File

@ -0,0 +1,9 @@
//valid,42footesttest2stuff
//:END:
function int str str => str str str foo
{
intToStr req req print req deq "test" "test2" "stuff" print
}
42 "foo" "bar" foo print print println

View File

@ -0,0 +1,2 @@
//invalid,Call to unknown function 'foo' at 2:4:END:
foo

View File

@ -0,0 +1,19 @@
//invalid,Call to unknown function 'bar' at 9:7:END:
function int => foo
{
while
{
if
{
bar
}
else
{
1 -
}
}
deq
}
42 foo

View File

@ -0,0 +1,8 @@
//invalid,Call to unknown function 'bar' at 5:5:END:
function => foo
{
bar
}
foo

25
tests/while.qbl Normal file
View File

@ -0,0 +1,25 @@
//valid,10987654321falsefalse
//:END:
import "../std.qbl"
true while
{
false
}
10 0 dup > req
// true 10
while
{
// i
dup intToStr req print
1 - 0 dup > req
}
deq
true true true while
{
false
}
boolToStr boolToStr print println