summaryrefslogtreecommitdiffstats
path: root/third_party/rust/minimal-lexical/src/parse.rs
blob: 9349699eb35e53bb11248ea28282bdce589086d6 (plain)
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
//! Parse byte iterators to float.

#![doc(hidden)]

#[cfg(feature = "compact")]
use crate::bellerophon::bellerophon;
use crate::extended_float::{extended_to_float, ExtendedFloat};
#[cfg(not(feature = "compact"))]
use crate::lemire::lemire;
use crate::num::Float;
use crate::number::Number;
use crate::slow::slow;

/// Try to parse the significant digits quickly.
///
/// This attempts a very quick parse, to deal with common cases.
///
/// * `integer`     - Slice containing the integer digits.
/// * `fraction`    - Slice containing the fraction digits.
#[inline]
fn parse_number_fast<'a, Iter1, Iter2>(
    integer: Iter1,
    fraction: Iter2,
    exponent: i32,
) -> Option<Number>
where
    Iter1: Iterator<Item = &'a u8>,
    Iter2: Iterator<Item = &'a u8>,
{
    let mut num = Number::default();
    let mut integer_count: usize = 0;
    let mut fraction_count: usize = 0;
    for &c in integer {
        integer_count += 1;
        let digit = c - b'0';
        num.mantissa = num.mantissa.wrapping_mul(10).wrapping_add(digit as u64);
    }
    for &c in fraction {
        fraction_count += 1;
        let digit = c - b'0';
        num.mantissa = num.mantissa.wrapping_mul(10).wrapping_add(digit as u64);
    }

    if integer_count + fraction_count <= 19 {
        // Can't overflow, since must be <= 19.
        num.exponent = exponent.saturating_sub(fraction_count as i32);
        Some(num)
    } else {
        None
    }
}

/// Parse the significant digits of the float and adjust the exponent.
///
/// * `integer`     - Slice containing the integer digits.
/// * `fraction`    - Slice containing the fraction digits.
#[inline]
fn parse_number<'a, Iter1, Iter2>(mut integer: Iter1, mut fraction: Iter2, exponent: i32) -> Number
where
    Iter1: Iterator<Item = &'a u8> + Clone,
    Iter2: Iterator<Item = &'a u8> + Clone,
{
    // NOTE: for performance, we do this in 2 passes:
    if let Some(num) = parse_number_fast(integer.clone(), fraction.clone(), exponent) {
        return num;
    }

    // Can only add 19 digits.
    let mut num = Number::default();
    let mut count = 0;
    while let Some(&c) = integer.next() {
        count += 1;
        if count == 20 {
            // Only the integer digits affect the exponent.
            num.many_digits = true;
            num.exponent = exponent.saturating_add(into_i32(1 + integer.count()));
            return num;
        } else {
            let digit = c - b'0';
            num.mantissa = num.mantissa * 10 + digit as u64;
        }
    }

    // Skip leading fraction zeros.
    // This is required otherwise we might have a 0 mantissa and many digits.
    let mut fraction_count: usize = 0;
    if count == 0 {
        for &c in &mut fraction {
            fraction_count += 1;
            if c != b'0' {
                count += 1;
                let digit = c - b'0';
                num.mantissa = num.mantissa * 10 + digit as u64;
                break;
            }
        }
    }
    for c in fraction {
        fraction_count += 1;
        count += 1;
        if count == 20 {
            num.many_digits = true;
            // This can't wrap, since we have at most 20 digits.
            // We've adjusted the exponent too high by `fraction_count - 1`.
            // Note: -1 is due to incrementing this loop iteration, which we
            // didn't use.
            num.exponent = exponent.saturating_sub(fraction_count as i32 - 1);
            return num;
        } else {
            let digit = c - b'0';
            num.mantissa = num.mantissa * 10 + digit as u64;
        }
    }

    // No truncated digits: easy.
    // Cannot overflow: <= 20 digits.
    num.exponent = exponent.saturating_sub(fraction_count as i32);
    num
}

/// Parse float from extracted float components.
///
/// * `integer`     - Cloneable, forward iterator over integer digits.
/// * `fraction`    - Cloneable, forward iterator over integer digits.
/// * `exponent`    - Parsed, 32-bit exponent.
///
/// # Preconditions
/// 1. The integer should not have leading zeros.
/// 2. The fraction should not have trailing zeros.
/// 3. All bytes in `integer` and `fraction` should be valid digits,
///     in the range [`b'0', b'9'].
///
/// # Panics
///
/// Although passing garbage input will not cause memory safety issues,
/// it is very likely to cause a panic with a large number of digits, or
/// in debug mode. The big-integer arithmetic without the `alloc` feature
/// assumes a maximum, fixed-width input, which assumes at maximum a
/// value of `10^(769 + 342)`, or ~4000 bits of storage. Passing in
/// nonsensical digits may require up to ~6000 bits of storage, which will
/// panic when attempting to add it to the big integer. It is therefore
/// up to the caller to validate this input.
///
/// We cannot efficiently remove trailing zeros while only accepting a
/// forward iterator.
pub fn parse_float<'a, F, Iter1, Iter2>(integer: Iter1, fraction: Iter2, exponent: i32) -> F
where
    F: Float,
    Iter1: Iterator<Item = &'a u8> + Clone,
    Iter2: Iterator<Item = &'a u8> + Clone,
{
    // Parse the mantissa and attempt the fast and moderate-path algorithms.
    let num = parse_number(integer.clone(), fraction.clone(), exponent);
    // Try the fast-path algorithm.
    if let Some(value) = num.try_fast_path() {
        return value;
    }

    // Now try the moderate path algorithm.
    let mut fp = moderate_path::<F>(&num);
    if fp.exp < 0 {
        // Undo the invalid extended float biasing.
        fp.exp -= F::INVALID_FP;
        fp = slow::<F, _, _>(num, fp, integer, fraction);
    }

    // Unable to correctly round the float using the fast or moderate algorithms.
    // Fallback to a slower, but always correct algorithm. If we have
    // lossy, we can't be here.
    extended_to_float::<F>(fp)
}

/// Wrapper for different moderate-path algorithms.
/// A return exponent of `-1` indicates an invalid value.
#[inline]
pub fn moderate_path<F: Float>(num: &Number) -> ExtendedFloat {
    #[cfg(not(feature = "compact"))]
    return lemire::<F>(num);

    #[cfg(feature = "compact")]
    return bellerophon::<F>(num);
}

/// Convert usize into i32 without overflow.
///
/// This is needed to ensure when adjusting the exponent relative to
/// the mantissa we do not overflow for comically-long exponents.
#[inline]
fn into_i32(value: usize) -> i32 {
    if value > i32::max_value() as usize {
        i32::max_value()
    } else {
        value as i32
    }
}

// Add digit to mantissa.
#[inline]
pub fn add_digit(value: u64, digit: u8) -> Option<u64> {
    value.checked_mul(10)?.checked_add(digit as u64)
}