c6d25d8d |
1 | /* |
2 | * cp949.c - CP949 / KS_C_5601_1987 multibyte encoding |
3 | */ |
4 | |
5 | #ifndef ENUM_CHARSETS |
6 | |
7 | #include "charset.h" |
8 | #include "internal.h" |
9 | |
10 | /* |
11 | * CP949 has no associated data, so `charset' may be ignored. |
12 | */ |
13 | |
14 | static void read_cp949(charset_spec const *charset, long int input_chr, |
15 | charset_state *state, |
16 | void (*emit)(void *ctx, long int output), void *emitctx) |
17 | { |
18 | UNUSEDARG(charset); |
19 | |
20 | /* |
21 | * For reading CP949, state->s0 simply contains the single |
22 | * stored lead byte when we are half way through a double-byte |
23 | * character, or 0 if we aren't. |
24 | */ |
25 | |
26 | if (state->s0 == 0) { |
27 | if (input_chr >= 0x81 && input_chr <= 0xFE) { |
28 | /* |
29 | * Lead byte. Just store it. |
30 | */ |
31 | state->s0 = input_chr; |
32 | } else { |
33 | /* |
34 | * Anything else we pass straight through unchanged. |
35 | */ |
36 | emit(emitctx, input_chr); |
37 | } |
38 | } else { |
39 | /* |
40 | * We have a stored lead byte. We expect a valid followup |
41 | * byte. |
42 | */ |
43 | if ((input_chr >= 0x40 && input_chr <= 0xFF)) { |
44 | emit(emitctx, cp949_to_unicode(state->s0 - 0x80, |
45 | input_chr - 0x40)); |
46 | } else { |
47 | emit(emitctx, ERROR); |
48 | } |
49 | state->s0 = 0; |
50 | } |
51 | } |
52 | |
53 | /* |
54 | * CP949 is a stateless multi-byte encoding (in the sense that just |
55 | * after any character has been completed, the state is always the |
56 | * same); hence when writing it, there is no need to use the |
57 | * charset_state. |
58 | */ |
59 | |
60 | static int write_cp949(charset_spec const *charset, long int input_chr, |
61 | charset_state *state, |
62 | void (*emit)(void *ctx, long int output), |
63 | void *emitctx) |
64 | { |
65 | UNUSEDARG(charset); |
66 | UNUSEDARG(state); |
67 | |
68 | if (input_chr == -1) |
69 | return TRUE; /* stateless; no cleanup required */ |
70 | |
71 | if (input_chr < 0x80) { |
72 | emit(emitctx, input_chr); |
73 | return TRUE; |
74 | } else { |
75 | int r, c; |
76 | if (unicode_to_cp949(input_chr, &r, &c)) { |
77 | emit(emitctx, r + 0x80); |
78 | emit(emitctx, c + 0x40); |
79 | return TRUE; |
80 | } else { |
81 | return FALSE; |
82 | } |
83 | } |
84 | } |
85 | |
86 | const charset_spec charset_CS_CP949 = { |
87 | CS_CP949, read_cp949, write_cp949, NULL |
88 | }; |
89 | |
90 | #else /* ENUM_CHARSETS */ |
91 | |
92 | ENUM_CHARSET(CS_CP949) |
93 | |
94 | #endif /* ENUM_CHARSETS */ |