处理AI胡乱生成的乱摊子

This commit is contained in:
2025-09-07 20:36:02 +08:00
parent ba513e0827
commit c4522b974b
403 changed files with 22915 additions and 44424 deletions

View File

@@ -1,3 +1,7 @@
//go:build (amd64 && go1.17 && !go1.26) || (arm64 && go1.20 && !go1.26)
// +build amd64,go1.17,!go1.26 arm64,go1.20,!go1.26
/*
* Copyright 2021 ByteDance Inc.
*
@@ -25,27 +29,45 @@ import (
`github.com/bytedance/sonic/internal/rt`
)
// String unescapes an escaped string (not including `"` at beginning and end)
// It validates invalid UTF8 and replace with `\ufffd`
func String(s string) (ret string, err types.ParsingError) {
mm := make([]byte, 0, len(s))
err = intoBytesUnsafe(s, &mm)
err = intoBytesUnsafe(s, &mm, true)
ret = rt.Mem2Str(mm)
return
}
// IntoBytes is same with String besides it output result into a buffer m
func IntoBytes(s string, m *[]byte) types.ParsingError {
if cap(*m) < len(s) {
return types.ERR_EOF
} else {
return intoBytesUnsafe(s, m)
return intoBytesUnsafe(s, m, true)
}
}
func intoBytesUnsafe(s string, m *[]byte) types.ParsingError {
// String unescapes an escaped string (not including `"` at beginning and end)
// - replace enables replacing invalid utf8 escaped char with `\uffd`
func _String(s string, replace bool) (ret string, err error) {
mm := make([]byte, 0, len(s))
err = intoBytesUnsafe(s, &mm, replace)
ret = rt.Mem2Str(mm)
return
}
func intoBytesUnsafe(s string, m *[]byte, replace bool) types.ParsingError {
pos := -1
slv := (*rt.GoSlice)(unsafe.Pointer(m))
str := (*rt.GoString)(unsafe.Pointer(&s))
/* unquote as the default configuration, replace invalid unicode with \ufffd */
ret := native.Unquote(str.Ptr, str.Len, slv.Ptr, &pos, types.F_UNICODE_REPLACE)
flags := uint64(0)
if replace {
/* unquote as the default configuration, replace invalid unicode with \ufffd */
flags |= types.F_UNICODE_REPLACE
}
ret := native.Unquote(str.Ptr, str.Len, slv.Ptr, &pos, flags)
/* check for errors */
if ret < 0 {
@@ -57,3 +79,6 @@ func intoBytesUnsafe(s string, m *[]byte) types.ParsingError {
runtime.KeepAlive(s)
return 0
}