2010-07-27 Paolo Carlini <paolo.carlini@oracle.com>
[official-gcc/alias-decl.git] / gcc / ada / a-suenco.adb
blob42b7f719a5ba5669038bf0ea0f3cf8943fbba88f
1 ------------------------------------------------------------------------------
2 -- --
3 -- GNAT RUN-TIME COMPONENTS --
4 -- --
5 -- ADA.STRINGS.UTF_ENCODING.CONVERSIONS --
6 -- --
7 -- B o d y --
8 -- --
9 -- Copyright (C) 2010, Free Software Foundation, Inc. --
10 -- --
11 -- GNAT is free software; you can redistribute it and/or modify it under --
12 -- terms of the GNU General Public License as published by the Free Soft- --
13 -- ware Foundation; either version 2, or (at your option) any later ver- --
14 -- sion. GNAT is distributed in the hope that it will be useful, but WITH- --
15 -- OUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY --
16 -- or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License --
17 -- for more details. You should have received a copy of the GNU General --
18 -- Public License distributed with GNAT; see file COPYING. If not, write --
19 -- to the Free Software Foundation, 51 Franklin Street, Fifth Floor, --
20 -- Boston, MA 02110-1301, USA. --
21 -- --
22 -- As a special exception, if other files instantiate generics from this --
23 -- unit, or you link this unit with other files to produce an executable, --
24 -- this unit does not by itself cause the resulting executable to be --
25 -- covered by the GNU General Public License. This exception does not --
26 -- however invalidate any other reasons why the executable file might be --
27 -- covered by the GNU Public License. --
28 -- --
29 -- GNAT was originally developed by the GNAT team at New York University. --
30 -- Extensive contributions were provided by Ada Core Technologies Inc. --
31 -- --
32 ------------------------------------------------------------------------------
34 package body Ada.Strings.UTF_Encoding.Conversions is
35 use Interfaces;
37 -- Version convertion from UTF-8/UTF-16BE/LE to UTF-8/UTF-16BE/LE
39 function Convert
40 (Item : UTF_String;
41 Input_Scheme : Encoding_Scheme;
42 Output_Scheme : Encoding_Scheme;
43 Output_BOM : Boolean := False) return UTF_String
45 begin
46 -- Nothing to do if identical schemes
48 if Input_Scheme = Output_Scheme then
49 return Item;
51 -- For remaining cases, one or other of the operands is UTF-16BE/LE
52 -- encoded, so go through UTF-16 intermediate.
54 else
55 return Convert (UTF_16_Wide_String'(Convert (Item, Input_Scheme)),
56 Output_Scheme, Output_BOM);
57 end if;
58 end Convert;
60 -- Version converting UTF-8/UTF-16BE/LE to UTF-16
62 function Convert
63 (Item : UTF_String;
64 Input_Scheme : Encoding_Scheme;
65 Output_BOM : Boolean := False) return UTF_16_Wide_String
67 begin
68 if Input_Scheme = UTF_8 then
69 return Convert (Item, Output_BOM);
70 else
71 return To_UTF_16 (Item, Input_Scheme, Output_BOM);
72 end if;
73 end Convert;
75 -- Version converting UTF-8 to UTF-16
77 function Convert
78 (Item : UTF_8_String;
79 Output_BOM : Boolean := False) return UTF_16_Wide_String
81 Result : UTF_16_Wide_String (1 .. Item'Length + 1);
82 -- Maximum length of result, including possible BOM
84 Len : Natural := 0;
85 -- Number of characters stored so far in Result
87 Iptr : Natural;
88 -- Next character to process in Item
90 C : Unsigned_8;
91 -- Input UTF-8 code
93 R : Unsigned_16;
94 -- Output UTF-16 code
96 procedure Get_Continuation;
97 -- Reads a continuation byte of the form 10xxxxxx, shifts R left
98 -- by 6 bits, and or's in the xxxxxx to the low order 6 bits. On
99 -- return Ptr is incremented. Raises exceptioon if continuation
100 -- byte does not exist or is invalid.
102 ----------------------
103 -- Get_Continuation --
104 ----------------------
106 procedure Get_Continuation is
107 begin
108 if Iptr > Item'Last then
109 Raise_Encoding_Error (Iptr - 1);
111 else
112 C := To_Unsigned_8 (Item (Iptr));
113 Iptr := Iptr + 1;
115 if C < 2#10_000000# or else C > 2#10_111111# then
116 Raise_Encoding_Error (Iptr - 1);
118 else
119 R := Shift_Left (R, 6) or
120 Unsigned_16 (C and 2#00_111111#);
121 end if;
122 end if;
123 end Get_Continuation;
125 -- Start of processing for Convert
127 begin
128 -- Output BOM if required
130 if Output_BOM then
131 Len := Len + 1;
132 Result (Len) := BOM_16 (1);
133 end if;
135 -- Skip OK BOM
137 Iptr := Item'First;
139 if Item'Length >= 3 and then Item (Iptr .. Iptr + 2) = BOM_8 then
140 Iptr := Iptr + 3;
142 -- Error if bad BOM
144 elsif Item'Length >= 2
145 and then (Item (Iptr .. Iptr + 1) = BOM_16BE
146 or else
147 Item (Iptr .. Iptr + 1) = BOM_16LE)
148 then
149 Raise_Encoding_Error (Iptr);
151 -- No BOM present
153 else
154 Iptr := Item'First;
155 end if;
157 while Iptr <= Item'Last loop
158 C := To_Unsigned_8 (Item (Iptr));
159 Iptr := Iptr + 1;
161 -- Codes in the range 16#00# - 16#7F#
162 -- UTF-8: 0xxxxxxx
163 -- UTF-16: 00000000_0xxxxxxx
165 if C <= 16#7F# then
166 Len := Len + 1;
167 Result (Len) := Wide_Character'Val (C);
169 -- No initial code can be of the form 10xxxxxx. Such codes are used
170 -- only for continuations.
172 elsif C <= 2#10_111111# then
173 Raise_Encoding_Error (Iptr - 1);
175 -- Codes in the range 16#80# - 16#7FF#
176 -- UTF-8: 110yyyxx 10xxxxxx
177 -- UTF-16: 00000yyy_xxxxxxxx
179 elsif C <= 2#110_11111# then
180 R := Unsigned_16 (C and 2#000_11111#);
181 Get_Continuation;
182 Len := Len + 1;
183 Result (Len) := Wide_Character'Val (R);
185 -- Codes in the range 16#800# - 16#FFFF#
186 -- UTF-8: 1110yyyy 10yyyyxx 10xxxxxx
187 -- UTF-16: yyyyyyyy_xxxxxxxx
189 elsif C <= 2#1110_1111# then
190 R := Unsigned_16 (C and 2#0000_1111#);
191 Get_Continuation;
192 Get_Continuation;
193 Len := Len + 1;
194 Result (Len) := Wide_Character'Val (R);
196 -- Make sure that we don't have a result in the forbidden range
197 -- reserved for UTF-16 surrogate characters.
199 if R in 16#D800# .. 16#DF00# then
200 Raise_Encoding_Error (Iptr - 3);
201 end if;
203 -- Codes in the range 16#10000# - 16#10FFFF#
204 -- UTF-8: 11110zzz 10zzyyyy 10yyyyxx 10xxxxxx
205 -- UTF-16: 110110zz_zzyyyyyy 110111yy_xxxxxxxx
206 -- Note: zzzz in the output is input zzzzz - 1
208 elsif C <= 2#11110_111# then
209 R := Unsigned_16 (C and 2#00000_111#);
210 Get_Continuation;
212 -- R now has zzzzzyyyy
214 R := R - 2#0000_1_0000#;
216 -- R now has zzzzyyyy (zzzz minus one for the output)
218 Get_Continuation;
220 -- R now has zzzzyyyyyyyyxx
222 Len := Len + 1;
223 Result (Len) :=
224 Wide_Character'Val
225 (2#110110_00_0000_0000# or Shift_Right (R, 4));
227 R := R and 2#1111#;
228 Get_Continuation;
229 Len := Len + 1;
230 Result (Len) :=
231 Wide_Character'Val (2#110111_00_0000_0000# or R);
233 -- Any other code is an error
235 else
236 Raise_Encoding_Error (Iptr - 1);
237 end if;
238 end loop;
240 return Result (1 .. Len);
241 end Convert;
243 -- Convert from UTF-16 to UTF-8/UTF-16-BE/LE
245 function Convert
246 (Item : UTF_16_Wide_String;
247 Output_Scheme : Encoding_Scheme;
248 Output_BOM : Boolean := False) return UTF_String
250 begin
251 if Output_Scheme = UTF_8 then
252 return Convert (Item, Output_BOM);
253 else
254 return From_UTF_16 (Item, Output_Scheme, Output_BOM);
255 end if;
256 end Convert;
258 -- Convert from UTF-16 to UTF-8
260 function Convert
261 (Item : UTF_16_Wide_String;
262 Output_BOM : Boolean := False) return UTF_8_String
264 Result : UTF_8_String (1 .. 3 * Item'Length + 3);
265 -- Worst case is 3 output codes for each input code + BOM space
267 Len : Natural;
268 -- Number of result codes stored
270 Iptr : Natural;
271 -- Pointer to next input character
273 C1, C2 : Unsigned_16;
275 zzzzz : Unsigned_16;
276 yyyyyyyy : Unsigned_16;
277 xxxxxxxx : Unsigned_16;
278 -- Components of double length case
280 begin
281 Iptr := Item'First;
283 -- Skip BOM at start of input
285 if Item'Length > 0 and then Item (Iptr) = BOM_16 (1) then
286 Iptr := Iptr + 1;
287 end if;
289 -- Generate output BOM if required
291 if Output_BOM then
292 Result (1 .. 3) := BOM_8;
293 Len := 3;
294 else
295 Len := 0;
296 end if;
298 -- Loop through input
300 while Iptr <= Item'Last loop
301 C1 := To_Unsigned_16 (Item (Iptr));
302 Iptr := Iptr + 1;
304 -- Codes in the range 16#0000# - 16#007F#
305 -- UTF-16: 000000000xxxxxxx
306 -- UTF-8: 0xxxxxxx
308 if C1 <= 16#007F# then
309 Result (Len + 1) := Character'Val (C1);
310 Len := Len + 1;
312 -- Codes in the range 16#80# - 16#7FF#
313 -- UTF-16: 00000yyyxxxxxxxx
314 -- UTF-8: 110yyyxx 10xxxxxx
316 elsif C1 <= 16#07FF# then
317 Result (Len + 1) :=
318 Character'Val
319 (2#110_000000# or Shift_Right (C1, 6));
320 Result (Len + 2) :=
321 Character'Val
322 (2#10_000000# or (C1 and 2#00_111111#));
323 Len := Len + 2;
325 -- Codes in the range 16#800# - 16#D7FF# or 16#E000# - 16#FFFF#
326 -- UTF-16: yyyyyyyyxxxxxxxx
327 -- UTF-8: 1110yyyy 10yyyyxx 10xxxxxx
329 elsif C1 <= 16#D7FF# or else C1 >= 16#E000# then
330 Result (Len + 1) :=
331 Character'Val
332 (2#1110_0000# or Shift_Right (C1, 12));
333 Result (Len + 2) :=
334 Character'Val
335 (2#10_000000# or (Shift_Right (C1, 6) and 2#00_111111#));
336 Result (Len + 3) :=
337 Character'Val
338 (2#10_000000# or (C1 and 2#00_111111#));
339 Len := Len + 3;
341 -- Codes in the range 16#10000# - 16#10FFFF#
342 -- UTF-16: 110110zzzzyyyyyy 110111yyxxxxxxxx
343 -- UTF-8: 11110zzz 10zzyyyy 10yyyyxx 10xxxxxx
344 -- Note: zzzzz in the output is input zzzz + 1
346 elsif C1 <= 2#110110_11_11111111# then
347 if Iptr > Item'Last then
348 Raise_Encoding_Error (Iptr - 1);
349 else
350 C2 := To_Unsigned_16 (Item (Iptr));
351 Iptr := Iptr + 1;
352 end if;
354 if (C2 and 2#111111_00_00000000#) /= 2#110111_00_00000000# then
355 Raise_Encoding_Error (Iptr - 1);
356 end if;
358 zzzzz := (Shift_Right (C1, 6) and 2#1111#) + 1;
359 yyyyyyyy := ((Shift_Left (C1, 2) and 2#111111_00#)
361 (Shift_Right (C2, 8) and 2#000000_11#));
362 xxxxxxxx := C2 and 2#11111111#;
364 Result (Len + 1) :=
365 Character'Val
366 (2#11110_000# or (Shift_Right (zzzzz, 2)));
367 Result (Len + 2) :=
368 Character'Val
369 (2#10_000000# or Shift_Left (zzzzz and 2#11#, 4)
370 or Shift_Right (yyyyyyyy, 4));
371 Result (Len + 3) :=
372 Character'Val
373 (2#10_000000# or Shift_Left (yyyyyyyy and 2#1111#, 4)
374 or Shift_Right (xxxxxxxx, 6));
375 Result (Len + 4) :=
376 Character'Val
377 (2#10_000000# or (xxxxxxxx and 2#00_111111#));
378 Len := Len + 4;
380 -- Error if input in 16#DC00# - 16#DFFF# (2nd surrogate with no 1st)
382 else
383 Raise_Encoding_Error (Iptr - 2);
384 end if;
385 end loop;
387 return Result (1 .. Len);
388 end Convert;
390 end Ada.Strings.UTF_Encoding.Conversions;