Files
neon/libs/proxy/json/src/str.rs
Conrad Ludgate 55aef2993d introduce a JSON serialization lib (#12417)
See #11992 and #11961 for some examples of usecases.

This introduces a JSON serialization lib, designed for more flexibility
than serde_json offers.

## Dynamic construction

Sometimes you have dynamic values you want to serialize, that are not
already in a serde-aware model like a struct or a Vec etc. To achieve
this with serde, you need to implement a lot of different traits on a
lot of different new-types. Because of this, it's often easier to
give-in and pull all the data into a serde-aware model
(serde_json::Value or some intermediate struct), but that is often not
very efficient.

This crate allows full control over the JSON encoding without needing to
implement any extra traits. Just call the relevant functions, and it
will guarantee a correctly encoded JSON value.

## Async construction

Similar to the above, sometimes the values arrive asynchronously. Often
collecting those values in memory is more expensive than writing them as
JSON, since the overheads of `Vec` and `String` is much higher, however
there are exceptions.

Serializing to JSON all in one go is also more CPU intensive and can
cause lag spikes, whereas serializing values incrementally spreads out
the CPU load and reduces lag.
2025-07-07 15:12:02 +00:00

167 lines
5.6 KiB
Rust

//! Helpers for serializing escaped strings.
//!
//! ## License
//!
//! <https://github.com/serde-rs/json/blob/c1826ebcccb1a520389c6b78ad3da15db279220d/src/ser.rs#L1514-L1552>
//! <https://github.com/serde-rs/json/blob/c1826ebcccb1a520389c6b78ad3da15db279220d/src/ser.rs#L2081-L2157>
//! Licensed by David Tolnay under MIT or Apache-2.0.
//!
//! With modifications by Conrad Ludgate on behalf of Databricks.
use std::fmt::{self, Write};
/// Represents a character escape code in a type-safe manner.
pub enum CharEscape {
/// An escaped quote `"`
Quote,
/// An escaped reverse solidus `\`
ReverseSolidus,
// /// An escaped solidus `/`
// Solidus,
/// An escaped backspace character (usually escaped as `\b`)
Backspace,
/// An escaped form feed character (usually escaped as `\f`)
FormFeed,
/// An escaped line feed character (usually escaped as `\n`)
LineFeed,
/// An escaped carriage return character (usually escaped as `\r`)
CarriageReturn,
/// An escaped tab character (usually escaped as `\t`)
Tab,
/// An escaped ASCII plane control character (usually escaped as
/// `\u00XX` where `XX` are two hex characters)
AsciiControl(u8),
}
impl CharEscape {
#[inline]
fn from_escape_table(escape: u8, byte: u8) -> CharEscape {
match escape {
self::BB => CharEscape::Backspace,
self::TT => CharEscape::Tab,
self::NN => CharEscape::LineFeed,
self::FF => CharEscape::FormFeed,
self::RR => CharEscape::CarriageReturn,
self::QU => CharEscape::Quote,
self::BS => CharEscape::ReverseSolidus,
self::UU => CharEscape::AsciiControl(byte),
_ => unreachable!(),
}
}
}
pub(crate) fn format_escaped_str(writer: &mut Vec<u8>, value: &str) {
writer.reserve(2 + value.len());
writer.push(b'"');
let rest = format_escaped_str_contents(writer, value);
writer.extend_from_slice(rest);
writer.push(b'"');
}
pub(crate) fn format_escaped_fmt(writer: &mut Vec<u8>, args: fmt::Arguments) {
writer.push(b'"');
Collect { buf: writer }
.write_fmt(args)
.expect("formatting should not error");
writer.push(b'"');
}
struct Collect<'buf> {
buf: &'buf mut Vec<u8>,
}
impl fmt::Write for Collect<'_> {
fn write_str(&mut self, s: &str) -> fmt::Result {
let last = format_escaped_str_contents(self.buf, s);
self.buf.extend(last);
Ok(())
}
}
// writes any escape sequences, and returns the suffix still needed to be written.
fn format_escaped_str_contents<'a>(writer: &mut Vec<u8>, value: &'a str) -> &'a [u8] {
let bytes = value.as_bytes();
let mut start = 0;
for (i, &byte) in bytes.iter().enumerate() {
let escape = ESCAPE[byte as usize];
if escape == 0 {
continue;
}
writer.extend_from_slice(&bytes[start..i]);
let char_escape = CharEscape::from_escape_table(escape, byte);
write_char_escape(writer, char_escape);
start = i + 1;
}
&bytes[start..]
}
const BB: u8 = b'b'; // \x08
const TT: u8 = b't'; // \x09
const NN: u8 = b'n'; // \x0A
const FF: u8 = b'f'; // \x0C
const RR: u8 = b'r'; // \x0D
const QU: u8 = b'"'; // \x22
const BS: u8 = b'\\'; // \x5C
const UU: u8 = b'u'; // \x00...\x1F except the ones above
const __: u8 = 0;
// Lookup table of escape sequences. A value of b'x' at index i means that byte
// i is escaped as "\x" in JSON. A value of 0 means that byte i is not escaped.
static ESCAPE: [u8; 256] = [
// 1 2 3 4 5 6 7 8 9 A B C D E F
UU, UU, UU, UU, UU, UU, UU, UU, BB, TT, NN, UU, FF, RR, UU, UU, // 0
UU, UU, UU, UU, UU, UU, UU, UU, UU, UU, UU, UU, UU, UU, UU, UU, // 1
__, __, QU, __, __, __, __, __, __, __, __, __, __, __, __, __, // 2
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // 3
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // 4
__, __, __, __, __, __, __, __, __, __, __, __, BS, __, __, __, // 5
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // 6
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // 7
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // 8
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // 9
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // A
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // B
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // C
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // D
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // E
__, __, __, __, __, __, __, __, __, __, __, __, __, __, __, __, // F
];
fn write_char_escape(writer: &mut Vec<u8>, char_escape: CharEscape) {
let s = match char_escape {
CharEscape::Quote => b"\\\"",
CharEscape::ReverseSolidus => b"\\\\",
// CharEscape::Solidus => b"\\/",
CharEscape::Backspace => b"\\b",
CharEscape::FormFeed => b"\\f",
CharEscape::LineFeed => b"\\n",
CharEscape::CarriageReturn => b"\\r",
CharEscape::Tab => b"\\t",
CharEscape::AsciiControl(byte) => {
static HEX_DIGITS: [u8; 16] = *b"0123456789abcdef";
let bytes = &[
b'\\',
b'u',
b'0',
b'0',
HEX_DIGITS[(byte >> 4) as usize],
HEX_DIGITS[(byte & 0xF) as usize],
];
return writer.extend_from_slice(bytes);
}
};
writer.extend_from_slice(s);
}