• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 /* Copyright (c) 2012 The Chromium Authors. All rights reserved.
2  * Use of this source code is governed by a BSD-style license that can be
3  * found in the LICENSE file.
4  */
5 #ifndef PPAPI_C_DEV_PPB_CHAR_SET_DEV_H_
6 #define PPAPI_C_DEV_PPB_CHAR_SET_DEV_H_
7 
8 #include "ppapi/c/pp_instance.h"
9 #include "ppapi/c/pp_macros.h"
10 #include "ppapi/c/pp_stdint.h"
11 #include "ppapi/c/pp_var.h"
12 
13 #define PPB_CHAR_SET_DEV_INTERFACE_0_4 "PPB_CharSet(Dev);0.4"
14 #define PPB_CHAR_SET_DEV_INTERFACE PPB_CHAR_SET_DEV_INTERFACE_0_4
15 
16 // Specifies the error behavior for the character set conversion functions.
17 // This will affect two cases: where the input is not encoded correctly, and
18 // when the requested character can not be converted to the destination
19 // character set.
20 enum PP_CharSet_ConversionError {
21   // Causes the entire conversion to fail if an error is encountered. The
22   // conversion function will return NULL.
23   PP_CHARSET_CONVERSIONERROR_FAIL,
24 
25   // Silently skips over errors. Unrepresentable characters and input encoding
26   // errors will be removed from the output.
27   PP_CHARSET_CONVERSIONERROR_SKIP,
28 
29   // Replaces the error or unrepresentable character with a substitution
30   // character. When converting to a Unicode character set (UTF-8 or UTF-16)
31   // it will use the unicode "substitution character" U+FFFD. When converting
32   // to another character set, the character will be charset-specific. For
33   // many languages this will be the representation of the '?' character.
34   PP_CHARSET_CONVERSIONERROR_SUBSTITUTE
35 };
36 PP_COMPILE_ASSERT_ENUM_SIZE_IN_BYTES(PP_CharSet_ConversionError, 4);
37 
38 struct PPB_CharSet_Dev_0_4 {
39   // Converts the UTF-16 string pointed to in |*utf16| to an 8-bit string in the
40   // specified code page. |utf16_len| is measured in UTF-16 units, not bytes.
41   // This value may not be NULL.
42   //
43   // The return value is a NULL-terminated 8-bit string corresponding to the
44   // new character set, or NULL on failure. THIS STRING MUST BE FREED USING
45   // PPB_Core::MemFree(). The length of the returned string, not including the
46   // terminating NULL, will be placed into *output_length. When there is no
47   // error, the result will always be non-NULL, even if the output is 0-length.
48   // In this case, it will only contain the terminator. You must still call
49   // MemFree any time the return value is non-NULL.
50   //
51   // This function will return NULL if there was an error converting the string
52   // and you requested PP_CHARSET_CONVERSIONERROR_FAIL, or the output character
53   // set was unknown.
54   char* (*UTF16ToCharSet)(PP_Instance instance,
55                           const uint16_t* utf16, uint32_t utf16_len,
56                           const char* output_char_set,
57                           enum PP_CharSet_ConversionError on_error,
58                           uint32_t* output_length);
59 
60   // Same as UTF16ToCharSet except converts in the other direction. The input
61   // is in the given charset, and the |input_len| is the number of bytes in
62   // the |input| string. |*output_length| is the number of 16-bit values in
63   // the output not counting the terminating NULL.
64   //
65   // Since UTF16 can represent every Unicode character, the only time the
66   // replacement character will be used is if the encoding in the input string
67   // is incorrect.
68   uint16_t* (*CharSetToUTF16)(PP_Instance instance,
69                               const char* input, uint32_t input_len,
70                               const char* input_char_set,
71                               enum PP_CharSet_ConversionError on_error,
72                               uint32_t* output_length);
73 
74   // Returns a string var representing the current multi-byte character set of
75   // the current system.
76   //
77   // WARNING: You really shouldn't be using this function unless you're dealing
78   // with legacy data. You should be using UTF-8 or UTF-16 and you don't have
79   // to worry about the character sets.
80   struct PP_Var (*GetDefaultCharSet)(PP_Instance instance);
81 };
82 
83 typedef struct PPB_CharSet_Dev_0_4 PPB_CharSet_Dev;
84 
85 #endif  /* PPAPI_C_DEV_PPB_CHAR_SET_DEV_H_ */
86