Bug 628949 - Update visible region / glass regions after we paint. r=roc a=2.0.
[mozilla-central.git] / netwerk / dns / punycode.c
blob56f6b021e02604d14c505b2413bb0c8cd5ede819
1 /*
2 punycode.c from RFC 3492
3 http://www.nicemice.net/idn/
4 Adam M. Costello
5 http://www.nicemice.net/amc/
7 This is ANSI C code (C89) implementing Punycode (RFC 3492).
10 C. Disclaimer and license
12 Regarding this entire document or any portion of it (including
13 the pseudocode and C code), the author makes no guarantees and
14 is not responsible for any damage resulting from its use. The
15 author grants irrevocable permission to anyone to use, modify,
16 and distribute it in any way that does not diminish the rights
17 of anyone else to use, modify, and distribute it, provided that
18 redistributed derivative works do not contain misleading author or
19 version information. Derivative works need not be licensed under
20 similar terms.
23 #include "punycode.h"
25 /**********************************************************/
26 /* Implementation (would normally go in its own .c file): */
28 #include <string.h>
30 /*** Bootstring parameters for Punycode ***/
32 enum { base = 36, tmin = 1, tmax = 26, skew = 38, damp = 700,
33 initial_bias = 72, initial_n = 0x80, delimiter = 0x2D };
35 /* basic(cp) tests whether cp is a basic code point: */
36 #define basic(cp) ((punycode_uint)(cp) < 0x80)
38 /* delim(cp) tests whether cp is a delimiter: */
39 #define delim(cp) ((cp) == delimiter)
41 /* decode_digit(cp) returns the numeric value of a basic code */
42 /* point (for use in representing integers) in the range 0 to */
43 /* base-1, or base if cp is does not represent a value. */
45 static punycode_uint decode_digit(punycode_uint cp)
47 return cp - 48 < 10 ? cp - 22 : cp - 65 < 26 ? cp - 65 :
48 cp - 97 < 26 ? cp - 97 : base;
51 /* encode_digit(d,flag) returns the basic code point whose value */
52 /* (when used for representing integers) is d, which needs to be in */
53 /* the range 0 to base-1. The lowercase form is used unless flag is */
54 /* nonzero, in which case the uppercase form is used. The behavior */
55 /* is undefined if flag is nonzero and digit d has no uppercase form. */
57 static char encode_digit(punycode_uint d, int flag)
59 return d + 22 + 75 * (d < 26) - ((flag != 0) << 5);
60 /* 0..25 map to ASCII a..z or A..Z */
61 /* 26..35 map to ASCII 0..9 */
64 /* flagged(bcp) tests whether a basic code point is flagged */
65 /* (uppercase). The behavior is undefined if bcp is not a */
66 /* basic code point. */
68 #define flagged(bcp) ((punycode_uint)(bcp) - 65 < 26)
70 /* encode_basic(bcp,flag) forces a basic code point to lowercase */
71 /* if flag is zero, uppercase if flag is nonzero, and returns */
72 /* the resulting code point. The code point is unchanged if it */
73 /* is caseless. The behavior is undefined if bcp is not a basic */
74 /* code point. */
76 static char encode_basic(punycode_uint bcp, int flag)
78 bcp -= (bcp - 97 < 26) << 5;
79 return bcp + ((!flag && (bcp - 65 < 26)) << 5);
82 /*** Platform-specific constants ***/
84 /* maxint is the maximum value of a punycode_uint variable: */
85 static const punycode_uint maxint = (punycode_uint) -1;
86 /* Because maxint is unsigned, -1 becomes the maximum value. */
88 /*** Bias adaptation function ***/
90 static punycode_uint adapt(
91 punycode_uint delta, punycode_uint numpoints, int firsttime )
93 punycode_uint k;
95 delta = firsttime ? delta / damp : delta >> 1;
96 /* delta >> 1 is a faster way of doing delta / 2 */
97 delta += delta / numpoints;
99 for (k = 0; delta > ((base - tmin) * tmax) / 2; k += base) {
100 delta /= base - tmin;
103 return k + (base - tmin + 1) * delta / (delta + skew);
106 /*** Main encode function ***/
108 enum punycode_status punycode_encode(
109 punycode_uint input_length,
110 const punycode_uint input[],
111 const unsigned char case_flags[],
112 punycode_uint *output_length,
113 char output[] )
115 punycode_uint n, delta, h, b, out, max_out, bias, j, m, q, k, t;
117 /* Initialize the state: */
119 n = initial_n;
120 delta = out = 0;
121 max_out = *output_length;
122 bias = initial_bias;
124 /* Handle the basic code points: */
126 for (j = 0; j < input_length; ++j) {
127 if (basic(input[j])) {
128 if (max_out - out < 2) return punycode_big_output;
129 output[out++] =
130 case_flags ? encode_basic(input[j], case_flags[j]) : (char)input[j];
132 /* else if (input[j] < n) return punycode_bad_input; */
133 /* (not needed for Punycode with unsigned code points) */
136 h = b = out;
138 /* h is the number of code points that have been handled, b is the */
139 /* number of basic code points, and out is the number of characters */
140 /* that have been output. */
142 if (b > 0) output[out++] = delimiter;
144 /* Main encoding loop: */
146 while (h < input_length) {
147 /* All non-basic code points < n have been */
148 /* handled already. Find the next larger one: */
150 for (m = maxint, j = 0; j < input_length; ++j) {
151 /* if (basic(input[j])) continue; */
152 /* (not needed for Punycode) */
153 if (input[j] >= n && input[j] < m) m = input[j];
156 /* Increase delta enough to advance the decoder's */
157 /* <n,i> state to <m,0>, but guard against overflow: */
159 if (m - n > (maxint - delta) / (h + 1)) return punycode_overflow;
160 delta += (m - n) * (h + 1);
161 n = m;
163 for (j = 0; j < input_length; ++j) {
164 /* Punycode does not need to check whether input[j] is basic: */
165 if (input[j] < n /* || basic(input[j]) */ ) {
166 if (++delta == 0) return punycode_overflow;
169 if (input[j] == n) {
170 /* Represent delta as a generalized variable-length integer: */
172 for (q = delta, k = base; ; k += base) {
173 if (out >= max_out) return punycode_big_output;
174 t = k <= bias /* + tmin */ ? tmin : /* +tmin not needed */
175 k >= bias + tmax ? tmax : k - bias;
176 if (q < t) break;
177 output[out++] = encode_digit(t + (q - t) % (base - t), 0);
178 q = (q - t) / (base - t);
181 output[out++] = encode_digit(q, case_flags && case_flags[j]);
182 bias = adapt(delta, h + 1, h == b);
183 delta = 0;
184 ++h;
188 ++delta, ++n;
191 *output_length = out;
192 return punycode_success;
195 /*** Main decode function ***/
197 enum punycode_status punycode_decode(
198 punycode_uint input_length,
199 const char input[],
200 punycode_uint *output_length,
201 punycode_uint output[],
202 unsigned char case_flags[] )
204 punycode_uint n, out, i, max_out, bias,
205 b, j, in, oldi, w, k, digit, t;
207 /* Initialize the state: */
209 n = initial_n;
210 out = i = 0;
211 max_out = *output_length;
212 bias = initial_bias;
214 /* Handle the basic code points: Let b be the number of input code */
215 /* points before the last delimiter, or 0 if there is none, then */
216 /* copy the first b code points to the output. */
218 for (b = j = 0; j < input_length; ++j) if (delim(input[j])) b = j;
219 if (b > max_out) return punycode_big_output;
221 for (j = 0; j < b; ++j) {
222 if (case_flags) case_flags[out] = flagged(input[j]);
223 if (!basic(input[j])) return punycode_bad_input;
224 output[out++] = input[j];
227 /* Main decoding loop: Start just after the last delimiter if any */
228 /* basic code points were copied; start at the beginning otherwise. */
230 for (in = b > 0 ? b + 1 : 0; in < input_length; ++out) {
232 /* in is the index of the next character to be consumed, and */
233 /* out is the number of code points in the output array. */
235 /* Decode a generalized variable-length integer into delta, */
236 /* which gets added to i. The overflow checking is easier */
237 /* if we increase i as we go, then subtract off its starting */
238 /* value at the end to obtain delta. */
240 for (oldi = i, w = 1, k = base; ; k += base) {
241 if (in >= input_length) return punycode_bad_input;
242 digit = decode_digit(input[in++]);
243 if (digit >= base) return punycode_bad_input;
244 if (digit > (maxint - i) / w) return punycode_overflow;
245 i += digit * w;
246 t = k <= bias /* + tmin */ ? tmin : /* +tmin not needed */
247 k >= bias + tmax ? tmax : k - bias;
248 if (digit < t) break;
249 if (w > maxint / (base - t)) return punycode_overflow;
250 w *= (base - t);
253 bias = adapt(i - oldi, out + 1, oldi == 0);
255 /* i was supposed to wrap around from out+1 to 0, */
256 /* incrementing n each time, so we'll fix that now: */
258 if (i / (out + 1) > maxint - n) return punycode_overflow;
259 n += i / (out + 1);
260 i %= (out + 1);
262 /* Insert n at position i of the output: */
264 /* not needed for Punycode: */
265 /* if (decode_digit(n) <= base) return punycode_invalid_input; */
266 if (out >= max_out) return punycode_big_output;
268 if (case_flags) {
269 memmove(case_flags + i + 1, case_flags + i, out - i);
270 /* Case of last character determines uppercase flag: */
271 case_flags[i] = flagged(input[in - 1]);
274 memmove(output + i + 1, output + i, (out - i) * sizeof *output);
275 output[i++] = n;
278 *output_length = out;
279 return punycode_success;