2023-03-27 21:22:29 -07:00
|
|
|
use std::collections::VecDeque;
|
|
|
|
|
|
|
|
|
2023-03-27 09:47:02 -07:00
|
|
|
mod tokenize;
|
|
|
|
mod treeify;
|
|
|
|
mod groupify;
|
|
|
|
mod find_subs;
|
|
|
|
|
|
|
|
|
2023-03-27 21:22:29 -07:00
|
|
|
use crate::parser::tokenize::tokenize;
|
|
|
|
use crate::parser::groupify::groupify;
|
|
|
|
use crate::parser::treeify::treeify;
|
|
|
|
use crate::parser::find_subs::find_subs;
|
|
|
|
|
2023-04-09 08:39:36 -07:00
|
|
|
use crate::quantity::Quantity;
|
2023-04-13 08:05:52 -07:00
|
|
|
use crate::quantity::Unit;
|
2023-04-01 13:50:52 -07:00
|
|
|
|
2023-03-27 21:22:29 -07:00
|
|
|
use crate::tokens::Token;
|
|
|
|
|
2023-03-28 10:46:29 -07:00
|
|
|
/// Specifies the location of a token in an input string.
|
|
|
|
/// Used to locate ParserErrors.
|
|
|
|
#[derive(Debug)]
|
|
|
|
#[derive(Copy, Clone)]
|
|
|
|
pub struct LineLocation {
|
|
|
|
pub pos: usize,
|
|
|
|
pub len: usize
|
|
|
|
}
|
|
|
|
|
2023-03-27 21:22:29 -07:00
|
|
|
#[derive(Debug)]
|
|
|
|
enum PreToken {
|
|
|
|
PreNumber(LineLocation, String),
|
|
|
|
PreWord(LineLocation, String),
|
|
|
|
PreOperator(LineLocation, String),
|
|
|
|
|
|
|
|
PreGroupStart(LineLocation),
|
|
|
|
PreGroupEnd(LineLocation),
|
|
|
|
PreGroup(LineLocation, VecDeque<PreToken>),
|
|
|
|
|
|
|
|
Container(Token)
|
|
|
|
}
|
|
|
|
|
|
|
|
impl PreToken {
|
|
|
|
#[inline(always)]
|
|
|
|
pub fn get_line_location(&self) -> &LineLocation {
|
|
|
|
match self {
|
|
|
|
PreToken::PreNumber(l, _)
|
|
|
|
| PreToken::PreWord(l, _)
|
|
|
|
| PreToken::PreOperator(l, _)
|
|
|
|
| PreToken::PreGroupStart(l)
|
|
|
|
| PreToken::PreGroupEnd(l)
|
|
|
|
| PreToken::PreGroup(l, _)
|
|
|
|
=> l,
|
|
|
|
|
|
|
|
_ => panic!()
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#[inline(always)]
|
|
|
|
pub fn get_mut_line_location(&mut self) -> &mut LineLocation {
|
|
|
|
match self {
|
|
|
|
PreToken::PreNumber(l, _)
|
|
|
|
| PreToken::PreWord(l, _)
|
|
|
|
| PreToken::PreOperator(l, _)
|
|
|
|
| PreToken::PreGroupStart(l)
|
|
|
|
| PreToken::PreGroupEnd(l)
|
|
|
|
| PreToken::PreGroup(l, _)
|
|
|
|
=> l,
|
|
|
|
|
|
|
|
_ => panic!()
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#[inline(always)]
|
|
|
|
pub fn to_token(self) -> Result<Token, (LineLocation, ParserError)>{
|
|
|
|
match self {
|
2023-04-07 09:33:55 -07:00
|
|
|
PreToken::PreNumber(l, mut s) => {
|
|
|
|
|
|
|
|
// The length check here ensures that
|
|
|
|
// `.` is not parsed as `0.`
|
|
|
|
// That should be a syntax error.
|
|
|
|
if s.len() != 1 && &s[0..1] == "." {
|
|
|
|
s.insert(0, '0');
|
|
|
|
}
|
2023-04-01 14:20:11 -07:00
|
|
|
|
2023-04-07 18:11:20 -07:00
|
|
|
let r = Quantity::new_rational_from_string(&s);
|
2023-04-01 14:20:11 -07:00
|
|
|
if r.is_none() {
|
|
|
|
return Err((l, ParserError::BadNumber))
|
|
|
|
}
|
|
|
|
return Ok(Token::Number(r.unwrap()));
|
2023-03-27 21:22:29 -07:00
|
|
|
},
|
2023-04-01 14:20:11 -07:00
|
|
|
|
2023-03-27 21:22:29 -07:00
|
|
|
PreToken::PreWord(l, s) => {
|
2023-04-09 08:39:36 -07:00
|
|
|
let c = match &s[..] {
|
2023-03-27 21:22:29 -07:00
|
|
|
// Mathematical constants
|
2023-04-01 13:50:52 -07:00
|
|
|
// 100 digits of each.
|
2023-04-10 21:03:16 -07:00
|
|
|
"π"|"pi" => { Some((Quantity::new_float_from_string(
|
2023-04-09 08:39:36 -07:00
|
|
|
"3.141592653589793238462643383279502884197169399375105820974944592307816406286208998628034825342117067"
|
|
|
|
).unwrap(), String::from("π")))},
|
|
|
|
|
2023-04-10 21:03:16 -07:00
|
|
|
"e" => { Some((Quantity::new_float_from_string(
|
2023-04-09 08:39:36 -07:00
|
|
|
"2.718281828459045235360287471352662497757247093699959574966967627724076630353547594571382178525166427"
|
|
|
|
).unwrap(), String::from("e"))) },
|
|
|
|
|
2023-04-10 21:03:16 -07:00
|
|
|
"phi"|"φ" => { Some((Quantity::new_float_from_string(
|
2023-04-09 08:39:36 -07:00
|
|
|
"1.618033988749894848204586834365638117720309179805762862135448622705260462818902449707207204189391137"
|
|
|
|
).unwrap(), String::from("φ"))) },
|
|
|
|
|
|
|
|
_ => { None }
|
|
|
|
};
|
|
|
|
|
|
|
|
if c.is_some() {
|
2023-04-10 21:03:16 -07:00
|
|
|
let (a, b) = c.unwrap();
|
|
|
|
return Ok(Token::Constant(a, b));
|
2023-04-09 08:39:36 -07:00
|
|
|
}
|
|
|
|
|
2023-04-13 08:05:52 -07:00
|
|
|
let c = Unit::from_string(&s);
|
2023-04-10 21:03:16 -07:00
|
|
|
if c.is_some() { return Ok(Token::Number(c.unwrap())); }
|
|
|
|
|
2023-04-09 08:39:36 -07:00
|
|
|
return Err((l, ParserError::Undefined(s)));
|
2023-03-27 21:22:29 -07:00
|
|
|
}
|
2023-04-01 13:50:52 -07:00
|
|
|
|
2023-03-27 21:22:29 -07:00
|
|
|
PreToken::Container(v) => { return Ok(v); }
|
|
|
|
|
|
|
|
PreToken::PreOperator(_,_)
|
|
|
|
| PreToken::PreGroupStart(_)
|
|
|
|
| PreToken::PreGroupEnd(_)
|
|
|
|
| PreToken::PreGroup(_, _)
|
|
|
|
=> panic!()
|
|
|
|
};
|
|
|
|
}
|
|
|
|
|
|
|
|
}
|
2023-03-27 09:47:02 -07:00
|
|
|
|
|
|
|
|
|
|
|
/// Types of parser errors.
|
|
|
|
/// If we cannot parse a string, one of these is returned.
|
|
|
|
#[derive(Debug)]
|
|
|
|
pub enum ParserError {
|
|
|
|
//MissingCloseParen,
|
|
|
|
ExtraCloseParen,
|
|
|
|
EmptyGroup,
|
|
|
|
Syntax,
|
|
|
|
Undefined(String),
|
|
|
|
BadNumber
|
|
|
|
}
|
|
|
|
|
|
|
|
impl ParserError {
|
|
|
|
pub fn to_message(&self) -> String {
|
|
|
|
match self {
|
|
|
|
//ParserError::MissingCloseParen => {
|
|
|
|
// String::from("This group is never closed")
|
|
|
|
//},
|
|
|
|
ParserError::ExtraCloseParen => {
|
|
|
|
String::from("Extra close parenthesis")
|
|
|
|
},
|
|
|
|
ParserError::EmptyGroup => {
|
|
|
|
String::from("Groups can't be empty")
|
|
|
|
},
|
|
|
|
ParserError::Syntax => {
|
|
|
|
String::from("Syntax")
|
|
|
|
},
|
|
|
|
ParserError::Undefined(s) => {
|
|
|
|
format!("\"{s}\" isn't defined")
|
|
|
|
},
|
|
|
|
ParserError::BadNumber => {
|
|
|
|
String::from("Invalid number")
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
pub fn parse(
|
|
|
|
s: &String
|
|
|
|
) -> Result<
|
2023-03-27 21:22:29 -07:00
|
|
|
Token,
|
|
|
|
(LineLocation, ParserError)
|
2023-03-27 09:47:02 -07:00
|
|
|
> {
|
|
|
|
|
2023-03-27 21:22:29 -07:00
|
|
|
let tokens = tokenize(s);
|
|
|
|
let (_, tokens) = find_subs(tokens);
|
|
|
|
let g = groupify(tokens)?;
|
2023-03-29 10:39:15 -07:00
|
|
|
let g = treeify(g)?;
|
2023-03-27 09:47:02 -07:00
|
|
|
|
2023-03-29 10:39:15 -07:00
|
|
|
return Ok(g);
|
2023-03-27 09:47:02 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2023-03-27 22:13:14 -07:00
|
|
|
pub fn substitute(
|
2023-04-04 10:22:32 -07:00
|
|
|
s: &String, // The string to substitute
|
2023-03-27 22:13:14 -07:00
|
|
|
c: usize // Location of the cursor right now
|
2023-04-04 10:22:32 -07:00
|
|
|
) -> (
|
|
|
|
usize, // Location of cursor in substituted string
|
|
|
|
String // String with substitutions
|
|
|
|
) {
|
|
|
|
if s == "" { return (c, s.clone()) }
|
2023-03-27 09:47:02 -07:00
|
|
|
let mut new_s = s.clone();
|
|
|
|
|
2023-04-04 10:22:32 -07:00
|
|
|
let l = s.chars().count();
|
2023-03-27 21:22:29 -07:00
|
|
|
let tokens = tokenize(s);
|
|
|
|
let (subs, _) = find_subs(tokens);
|
2023-04-04 10:22:32 -07:00
|
|
|
let mut new_c = l - c;
|
2023-03-27 09:47:02 -07:00
|
|
|
|
|
|
|
for r in subs.iter() {
|
2023-04-04 10:22:32 -07:00
|
|
|
// find_subs gives substitutions in reverse order.
|
|
|
|
|
|
|
|
if { // Don't substitute if our cursor is inside the substitution
|
2023-03-27 22:13:14 -07:00
|
|
|
c >= r.0.pos &&
|
|
|
|
c < r.0.pos+r.0.len
|
|
|
|
} { continue; }
|
|
|
|
|
2023-04-04 10:22:32 -07:00
|
|
|
if c < r.0.pos {
|
|
|
|
let ct = r.1.chars().count();
|
|
|
|
if ct >= r.0.len {
|
|
|
|
if new_c >= ct - r.0.len {
|
|
|
|
new_c += ct - r.0.len
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
new_c -= r.0.len - ct
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2023-03-27 09:47:02 -07:00
|
|
|
new_s.replace_range(
|
|
|
|
r.0.pos..r.0.pos+r.0.len,
|
|
|
|
&r.1[..]
|
|
|
|
)
|
|
|
|
}
|
|
|
|
|
2023-04-04 10:22:32 -07:00
|
|
|
return (new_c, new_s);
|
2023-03-27 09:47:02 -07:00
|
|
|
}
|