unquote.go raw

   1  /*
   2   * Copyright 2021 ByteDance Inc.
   3   *
   4   * Licensed under the Apache License, Version 2.0 (the "License");
   5   * you may not use this file except in compliance with the License.
   6   * You may obtain a copy of the License at
   7   *
   8   *     http://www.apache.org/licenses/LICENSE-2.0
   9   *
  10   * Unless required by applicable law or agreed to in writing, software
  11   * distributed under the License is distributed on an "AS IS" BASIS,
  12   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  13   * See the License for the specific language governing permissions and
  14   * limitations under the License.
  15   */
  16  
  17  package unquote
  18  
  19  import (
  20      `unsafe`
  21      `runtime`
  22  
  23      `github.com/bytedance/sonic/internal/native`
  24      `github.com/bytedance/sonic/internal/native/types`
  25      `github.com/bytedance/sonic/internal/rt`
  26  )
  27  
  28  // String unescapes an escaped string (not including `"` at beginning and end)
  29  // It validates invalid UTF8 and replace with `\ufffd`
  30  func String(s string) (ret string, err types.ParsingError) {
  31      mm := make([]byte, 0, len(s))
  32      err = intoBytesUnsafe(s, &mm, true)
  33      ret = rt.Mem2Str(mm)
  34      return
  35  }
  36  
  37  // IntoBytes is same with String besides it output result into a buffer m
  38  func IntoBytes(s string, m *[]byte) types.ParsingError {
  39      if cap(*m) < len(s) {
  40          return types.ERR_EOF
  41      } else {
  42          return intoBytesUnsafe(s, m, true)
  43      }
  44  }
  45  
  46  // String unescapes an escaped string (not including `"` at beginning and end)
  47  //   - replace enables replacing invalid utf8 escaped char with `\uffd`
  48  func _String(s string, replace bool) (ret string, err error) {
  49      mm := make([]byte, 0, len(s))
  50      err = intoBytesUnsafe(s, &mm, replace)
  51      ret = rt.Mem2Str(mm)
  52      return
  53  }
  54  
  55  func intoBytesUnsafe(s string, m *[]byte, replace bool) types.ParsingError {
  56      pos := -1
  57      slv := (*rt.GoSlice)(unsafe.Pointer(m))
  58      str := (*rt.GoString)(unsafe.Pointer(&s))
  59  
  60      flags := uint64(0)
  61      if replace {
  62          /* unquote as the default configuration, replace invalid unicode with \ufffd */
  63          flags |= types.F_UNICODE_REPLACE
  64      }
  65  
  66      ret := native.Unquote(str.Ptr, str.Len, slv.Ptr, &pos, flags)
  67  
  68      /* check for errors */
  69      if ret < 0 {
  70          return types.ParsingError(-ret)
  71      }
  72  
  73      /* update the length */
  74      slv.Len = ret
  75      runtime.KeepAlive(s)
  76      return 0
  77  }
  78  
  79  
  80  
  81