neyuki 0.6.2

A small Lua-like scripting language implemented in Rust
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219

//! The `utf8` library. Neyuki strings are always valid UTF-8; these
//! functions work in characters (code points) where `string` works in
//! bytes. Byte positions are 1-based like everywhere else.

use std::cell::RefCell;
use std::rc::Rc;

use crate::vm::libs::args::{int_arg, opt_int_arg, str_arg};
use crate::vm::libs::strlib::{pos_relat_end, pos_relat_start};
use crate::vm::machine::VM;
use crate::vm::value::{Value, VmTable};

pub fn create_utf8_lib() -> Value {
    let mut table = VmTable::new();
    macro_rules! reg {
        ($($name:literal => $func:ident),* $(,)?) => {
            $(table.set_str($name, crate::native!(concat!("utf8.", $name), $func));)*
        };
    }
    reg! {
        "char" => utf8_char,
        "chars" => utf8_chars,
        "codepoint" => utf8_codepoint,
        "codes" => utf8_codes,
        "len" => utf8_len,
        "offset" => utf8_offset,
        "sub" => utf8_sub,
    }
    crate::vm::libs::args::frozen(table)
}

/// `utf8.char(...)`: the string made of the given code points.
fn utf8_char(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let mut out = String::with_capacity(args.len());
    for index in 0..args.len() {
        let code = int_arg(args, index, &format!("code point {}", index + 1))?;
        let c = u32::try_from(code)
            .ok()
            .and_then(char::from_u32)
            .ok_or_else(|| format!("code point {} is not a valid character", code))?;
        out.push(c);
    }
    Ok(vec![Value::string(out)])
}

/// The byte range `i..=j` (1-based, negative from the end) as a 0-based
/// half-open range that starts on a character boundary.
fn byte_range(s: &str, i: i64, j: i64) -> Result<(usize, usize), String> {
    let start = pos_relat_start(i, s.len()) - 1;
    let end = pos_relat_end(j, s.len());
    if start > s.len() {
        return Err(format!("byte position {} is out of bounds", i));
    }
    if !s.is_char_boundary(start) {
        return Err(format!("byte position {} is inside a character", start + 1));
    }
    Ok((start, end.max(start)))
}

/// `utf8.codepoint(s, i?, j?)`: the code points of the characters that
/// start between byte positions `i` and `j` (both default to `i`).
fn utf8_codepoint(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let s = str_arg(args, 0, "s")?;
    let i = opt_int_arg(args, 1, "i", 1)?;
    let j = opt_int_arg(args, 2, "j", i)?;
    let (start, end) = byte_range(s, i, j)?;
    Ok(s[start..]
        .char_indices()
        .take_while(|(offset, _)| start + offset < end)
        .map(|(_, c)| Value::Int(c as i64))
        .collect())
}

/// `utf8.len(s, i?, j?)`: how many characters start between byte positions
/// `i` and `j`.
fn utf8_len(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let s = str_arg(args, 0, "s")?;
    let i = opt_int_arg(args, 1, "i", 1)?;
    let j = opt_int_arg(args, 2, "j", -1)?;
    let (start, end) = byte_range(s, i, j)?;
    let count = s[start..]
        .char_indices()
        .take_while(|(offset, _)| start + offset < end)
        .count();
    Ok(vec![Value::from_usize(count)])
}

/// `utf8.offset(s, n, i?)`: the byte position where the `n`th character
/// counting from byte position `i` starts (backwards for a negative `n`;
/// `n = 0` finds the start of the character containing `i`), or nil.
fn utf8_offset(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let s = str_arg(args, 0, "s")?;
    let n = int_arg(args, 1, "n")?;
    let default_i = if n >= 0 { 1 } else { s.len() as i64 + 1 };
    let i = opt_int_arg(args, 2, "i", default_i)?;
    let mut pos = pos_relat_start(i, s.len()) - 1;
    if pos > s.len() {
        return Ok(vec![Value::Nil]);
    }
    if n == 0 {
        while pos > 0 && !s.is_char_boundary(pos) {
            pos -= 1;
        }
        return Ok(vec![Value::from_usize(pos + 1)]);
    }
    if !s.is_char_boundary(pos) {
        return Err(format!("byte position {} is inside a character", pos + 1));
    }
    let found = if n > 0 {
        let skip = (n - 1) as usize;
        s[pos..]
            .char_indices()
            .map(|(offset, _)| pos + offset)
            .chain(std::iter::once(s.len()))
            .nth(skip)
    } else {
        let back = n.unsigned_abs() as usize;
        s[..pos]
            .char_indices()
            .rev()
            .map(|(offset, _)| offset)
            .nth(back - 1)
    };
    Ok(vec![match found {
        Some(p) if p <= s.len() => Value::from_usize(p + 1),
        _ => Value::Nil,
    }])
}

/// `utf8.sub(s, i, j?)`: the characters from the `i`th to the `j`th,
/// counted like `string.sub` counts bytes (negative from the end).
fn utf8_sub(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let s = str_arg(args, 0, "s")?;
    let i = opt_int_arg(args, 1, "i", 1)?;
    let j = opt_int_arg(args, 2, "j", -1)?;
    let count = s.chars().count();
    let first = pos_relat_start(i, count);
    let last = pos_relat_end(j, count);
    if first > last {
        return Ok(vec![Value::str("")]);
    }
    let out: String = s.chars().skip(first - 1).take(last + 1 - first).collect();
    Ok(vec![Value::string(out)])
}

/// An iterator over `s`: a table holding the string and the next byte
/// offset, called through `__call`.
fn iterator(s: &str, step: Value) -> Value {
    let mut meta = VmTable::new();
    meta.set_str("__call", step);
    let mut state = VmTable::new();
    state.set_str("s", Value::str(s));
    state.set_str("pos", Value::Int(0));
    state.metatable = Some(Rc::new(RefCell::new(meta)));
    Value::Table(Rc::new(RefCell::new(state)))
}

/// The string and next byte offset an iterator table carries.
fn iterator_state(
    args: &[Value],
) -> Option<(Rc<RefCell<VmTable>>, crate::vm::value::StrRef, usize)> {
    let Some(Value::Table(state)) = args.first() else {
        return None;
    };
    let st = state.borrow();
    let (Value::String(s), Value::Int(pos)) = (st.get_str("s"), st.get_str("pos")) else {
        return None;
    };
    drop(st);
    Some((state.clone(), s, pos as usize))
}

fn codes_step(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let Some((state, s, pos)) = iterator_state(args) else {
        return Ok(vec![Value::Nil]);
    };
    match s.get(pos..).and_then(|rest| rest.chars().next()) {
        Some(c) => {
            state
                .borrow_mut()
                .set_str("pos", Value::from_usize(pos + c.len_utf8()));
            Ok(vec![Value::from_usize(pos + 1), Value::Int(c as i64)])
        }
        None => Ok(vec![Value::Nil]),
    }
}

fn chars_step(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let Some((state, s, pos)) = iterator_state(args) else {
        return Ok(vec![Value::Nil]);
    };
    match s.get(pos..).and_then(|rest| rest.chars().next()) {
        Some(c) => {
            state
                .borrow_mut()
                .set_str("pos", Value::from_usize(pos + c.len_utf8()));
            Ok(vec![Value::string(c.to_string())])
        }
        None => Ok(vec![Value::Nil]),
    }
}

/// `utf8.codes(s)`: iterates `position, code point` for each character.
fn utf8_codes(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let s = str_arg(args, 0, "s")?;
    Ok(vec![iterator(
        s,
        crate::native!("utf8.codes iterator", codes_step),
    )])
}

/// `utf8.chars(s)`: iterates each character as a string.
fn utf8_chars(_vm: &mut VM, args: &[Value]) -> Result<Vec<Value>, String> {
    let s = str_arg(args, 0, "s")?;
    Ok(vec![iterator(
        s,
        crate::native!("utf8.chars iterator", chars_step),
    )])
}