simd-json 0.19.0

High performance JSON parser based on a port of simdjson
Documentation 
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

use std::mem::MaybeUninit;
use std::ops::Range;

use crate::charutils::{codepoint_to_utf8, hex_to_u32_nocheck};
use crate::error::ErrorType;
use crate::safer_unchecked::GetSaferUnchecked;

/// begin copypasta
/// These chars yield themselves: " \ /
/// b -> backspace, f -> formfeed, n -> newline, r -> cr, t -> horizontal tab
/// u not handled in this table as it's Container
pub(crate) const ESCAPE_MAP: [u8; 256] = [
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x0.
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x22, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0x2f, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, // 0x4.
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x5c, 0, 0, 0, // 0x5.
    0, 0, 0x08, 0, 0, 0, 0x0c, 0, 0, 0, 0, 0, 0, 0, 0x0a, 0, // 0x6.
    0, 0, 0x0d, 0, 0x09, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x7.
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
];

const HIGH_SURROGATES: Range<u32> = 0xd800..0xdc00;
const LOW_SURROGATES: Range<u32> = 0xdc00..0xe000;

/// handle a unicode codepoint
/// write appropriate values into dest
#[cfg_attr(not(feature = "no-inline"), inline)]
#[allow(dead_code)]
pub(crate) fn handle_unicode_codepoint(
    src_ptr: &[u8],
    dst_ptr: &mut [MaybeUninit<u8>],
) -> Result<(usize, usize), ErrorType> {
    let (code_point, src_offset) = get_unicode_codepoint(src_ptr)?;
    let offset: usize = codepoint_to_utf8(code_point, dst_ptr);
    Ok((offset, src_offset))
}

/// Decodes a `\uXXXX` escape or surrogate pair, returning the code point and the bytes consumed.
#[cfg_attr(not(feature = "no-inline"), inline)]
pub(crate) fn get_unicode_codepoint(mut src_ptr: &[u8]) -> Result<(u32, usize), ErrorType> {
    let mut code_point: u32 = hex_to_u32_nocheck(unsafe { src_ptr.get_kinda_unchecked(2..) });
    // Invalid hex digits set the high 16 bits; valid ones give at most 0xFFFF.
    if code_point > 0xFFFF {
        return Err(ErrorType::InvalidUnicodeCodepoint);
    }
    src_ptr = unsafe { src_ptr.get_kinda_unchecked(6..) };
    let mut src_offset = 6;
    // check for low surrogate for characters outside the Basic
    // Multilingual Plane.
    if HIGH_SURROGATES.contains(&code_point) {
        if (unsafe { *src_ptr.get_kinda_unchecked(0) } != b'\\')
            || unsafe { *src_ptr.get_kinda_unchecked(1) } != b'u'
        {
            // A high surrogate must be followed by a `\u` low surrogate.
            return Err(ErrorType::InvalidUnicodeCodepoint);
        }

        let code_point_2: u32 = hex_to_u32_nocheck(unsafe { src_ptr.get_kinda_unchecked(2..) });
        // Invalid hex digits set the high bits, which also fall outside this range.
        if !LOW_SURROGATES.contains(&code_point_2) {
            return Err(ErrorType::InvalidUnicodeCodepoint);
        }
        code_point = (((code_point - 0xd800) << 10) | (code_point_2 - 0xdc00)) + 0x10000;
        src_offset += 6;
    } else if LOW_SURROGATES.contains(&code_point) {
        // This is a low surrogate on it's own, which is invalid.
        return Err(ErrorType::InvalidUtf8);
    }
    Ok((code_point, src_offset))
}