orbit/internal/jsonvalue/canonical.go

// Package jsonvalue 比较 JSON 的实际内容，不将 JSON 数值转换为浮点数。
package jsonvalue

import (
	"bytes"
	"encoding/json"
	"errors"
	"io"
	"sort"
	"strconv"
	"strings"
	"unicode/utf8"
)

// Canonical 生成内部 orbit-json-v1 摘要输入：对象键排序、重复键拒绝、
// 精确十进制数规范化。该格式不是对外 JSON 编码承诺，也不宣称实现 RFC 8785。
func Canonical(raw []byte) ([]byte, error) {
	if !utf8.Valid(raw) {
		return nil, errors.New("JSON must be valid UTF-8")
	}
	if err := validEscapes(raw); err != nil {
		return nil, err
	}
	d := json.NewDecoder(bytes.NewReader(raw))
	d.UseNumber()
	b, err := value(d, 0)
	if err != nil {
		return nil, err
	}
	if _, err := d.Token(); err != io.EOF {
		return nil, errors.New("exactly one JSON value required")
	}
	return b, nil
}
func value(d *json.Decoder, depth int) ([]byte, error) {
	if depth > 32 {
		return nil, errors.New("JSON nesting exceeds 32 levels")
	}
	t, err := d.Token()
	if err != nil {
		return nil, err
	}
	switch v := t.(type) {
	case json.Delim:
		switch v {
		case '{':
			fields := map[string][]byte{}
			keys := []string{}
			for d.More() {
				token, err := d.Token()
				if err != nil {
					return nil, err
				}
				key, ok := token.(string)
				if !ok {
					return nil, errors.New("invalid object key")
				}
				if _, ok := fields[key]; ok {
					return nil, errors.New("duplicate JSON object key")
				}
				b, err := value(d, depth+1)
				if err != nil {
					return nil, err
				}
				fields[key] = b
				keys = append(keys, key)
			}
			if close, err := d.Token(); err != nil || close != json.Delim('}') {
				return nil, errors.New("unterminated JSON object")
			}
			sort.Strings(keys)
			var b bytes.Buffer
			b.WriteByte('{')
			for i, k := range keys {
				if i > 0 {
					b.WriteByte(',')
				}
				key, _ := json.Marshal(k)
				b.Write(key)
				b.WriteByte(':')
				b.Write(fields[k])
			}
			b.WriteByte('}')
			return b.Bytes(), nil
		case '[':
			var b bytes.Buffer
			b.WriteByte('[')
			n := 0
			for d.More() {
				v, err := value(d, depth+1)
				if err != nil {
					return nil, err
				}
				if n > 0 {
					b.WriteByte(',')
				}
				b.Write(v)
				n++
			}
			if close, err := d.Token(); err != nil || close != json.Delim(']') {
				return nil, errors.New("unterminated JSON array")
			}
			b.WriteByte(']')
			return b.Bytes(), nil
		default:
			return nil, errors.New("unexpected JSON delimiter")
		}
	case json.Number:
		return number(string(v))
	default:
		return json.Marshal(v)
	}
}
func number(s string) ([]byte, error) {
	if len(s) > 256 {
		return nil, errors.New("JSON number exceeds 256 characters")
	}
	// 仅用作有限范围检查；摘要始终由下面的十进制文本运算生成。
	if _, err := strconv.ParseFloat(s, 64); err != nil {
		return nil, errors.New("JSON number outside supported range")
	}
	negative := strings.HasPrefix(s, "-")
	s = strings.TrimPrefix(s, "-")
	exponent := 0
	if i := strings.IndexAny(s, "eE"); i >= 0 {
		var err error
		exponent, err = strconv.Atoi(s[i+1:])
		if err != nil || exponent < -308 || exponent > 308 {
			return nil, errors.New("JSON exponent outside supported range")
		}
		s = s[:i]
	}
	if i := strings.IndexByte(s, '.'); i >= 0 {
		exponent -= len(s) - i - 1
		s = s[:i] + s[i+1:]
	}
	s = strings.TrimLeft(s, "0")
	if s == "" {
		return []byte("0"), nil
	}
	for strings.HasSuffix(s, "0") {
		s = s[:len(s)-1]
		exponent++
	}
	if exponent < -308 || exponent > 308 {
		return nil, errors.New("normalized JSON exponent outside supported range")
	}
	if exponent+len(s)-1 > 308 {
		return nil, errors.New("JSON number outside supported range")
	}
	if negative {
		s = "-" + s
	}
	if exponent != 0 {
		s += "e" + strconv.Itoa(exponent)
	}
	return []byte(s), nil
}

// encoding/json 会替换孤立 UTF-16 代理项；拒绝它们，避免两个不同输入被悄悄合并。
func validEscapes(raw []byte) error {
	for i := 0; i < len(raw); i++ {
		if raw[i] != '\\' {
			continue
		}
		i++
		if i >= len(raw) || raw[i] != 'u' {
			continue
		}
		if i+4 >= len(raw) {
			return errors.New("invalid unicode escape")
		}
		v, err := strconv.ParseUint(string(raw[i+1:i+5]), 16, 16)
		if err != nil {
			return err
		}
		i += 4
		if v >= 0xdc00 && v <= 0xdfff {
			return errors.New("unpaired unicode surrogate")
		}
		if v < 0xd800 || v > 0xdbff {
			continue
		}
		if i+6 >= len(raw) || raw[i+1] != '\\' || raw[i+2] != 'u' {
			return errors.New("unpaired unicode surrogate")
		}
		v, err = strconv.ParseUint(string(raw[i+3:i+7]), 16, 16)
		if err != nil || v < 0xdc00 || v > 0xdfff {
			return errors.New("unpaired unicode surrogate")
		}
		i += 6
	}
	return nil
}