1 ------------------------------------------------------------------------------
3 -- GNAT RUN-TIME COMPONENTS --
5 -- ADA.STRINGS.UTF_ENCODING.CONVERSIONS --
9 -- Copyright (C) 2010-2013, Free Software Foundation, Inc. --
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 3, 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. --
18 -- As a special exception under Section 7 of GPL version 3, you are granted --
19 -- additional permissions described in the GCC Runtime Library Exception, --
20 -- version 3.1, as published by the Free Software Foundation. --
22 -- You should have received a copy of the GNU General Public License and --
23 -- a copy of the GCC Runtime Library Exception along with this program; --
24 -- see the files COPYING3 and COPYING.RUNTIME respectively. If not, see --
25 -- <http://www.gnu.org/licenses/>. --
27 -- GNAT was originally developed by the GNAT team at New York University. --
28 -- Extensive contributions were provided by Ada Core Technologies Inc. --
30 ------------------------------------------------------------------------------
32 package body Ada
.Strings
.UTF_Encoding
.Conversions
is
35 -- Convert from UTF-8/UTF-16BE/LE to UTF-8/UTF-16BE/LE
39 Input_Scheme
: Encoding_Scheme
;
40 Output_Scheme
: Encoding_Scheme
;
41 Output_BOM
: Boolean := False) return UTF_String
44 -- Nothing to do if identical schemes, but for UTF_8 we need to
45 -- exclude overlong encodings, so need to do the full conversion.
47 if Input_Scheme
= Output_Scheme
48 and then Input_Scheme
/= UTF_8
52 -- For remaining cases, one or other of the operands is UTF-16BE/LE
53 -- encoded, so go through UTF-16 intermediate.
56 return Convert
(UTF_16_Wide_String
'(Convert (Item, Input_Scheme)),
57 Output_Scheme, Output_BOM);
61 -- Convert from UTF-8/UTF-16BE/LE to UTF-16
65 Input_Scheme : Encoding_Scheme;
66 Output_BOM : Boolean := False) return UTF_16_Wide_String
69 if Input_Scheme = UTF_8 then
70 return Convert (Item, Output_BOM);
72 return To_UTF_16 (Item, Input_Scheme, Output_BOM);
76 -- Convert from UTF-8 to UTF-16
80 Output_BOM : Boolean := False) return UTF_16_Wide_String
82 Result : UTF_16_Wide_String (1 .. Item'Length + 1);
83 -- Maximum length of result, including possible BOM
86 -- Number of characters stored so far in Result
89 -- Next character to process in Item
97 procedure Get_Continuation;
98 -- Reads a continuation byte of the form 10xxxxxx, shifts R left by 6
99 -- bits, and or's in the xxxxxx to the low order 6 bits. On return Ptr
100 -- is incremented. Raises exception if continuation byte does not exist
103 ----------------------
104 -- Get_Continuation --
105 ----------------------
107 procedure Get_Continuation is
109 if Iptr > Item'Last then
110 Raise_Encoding_Error (Iptr - 1);
113 C := To_Unsigned_8 (Item (Iptr));
116 if C < 2#10_000000# or else C > 2#10_111111# then
117 Raise_Encoding_Error (Iptr - 1);
121 Shift_Left (R, 6) or Unsigned_16 (C and 2#00_111111#);
124 end Get_Continuation;
126 -- Start of processing for Convert
129 -- Output BOM if required
133 Result (Len) := BOM_16 (1);
140 if Item'Length >= 3 and then Item (Iptr .. Iptr + 2) = BOM_8 then
145 elsif Item'Length >= 2
146 and then (Item (Iptr .. Iptr + 1) = BOM_16BE
148 Item (Iptr .. Iptr + 1) = BOM_16LE)
150 Raise_Encoding_Error (Iptr);
158 while Iptr <= Item'Last loop
159 C := To_Unsigned_8 (Item (Iptr));
162 -- Codes in the range 16#00# - 16#7F#
164 -- UTF-16: 00000000_0xxxxxxx
168 Result (Len) := Wide_Character'Val (C);
170 -- No initial code can be of the form 10xxxxxx. Such codes are used
171 -- only for continuations.
173 elsif C <= 2#10_111111# then
174 Raise_Encoding_Error (Iptr - 1);
176 -- Codes in the range 16#80# - 16#7FF#
177 -- UTF-8: 110yyyxx 10xxxxxx
178 -- UTF-16: 00000yyy_xxxxxxxx
180 elsif C <= 2#110_11111# then
181 R := Unsigned_16 (C and 2#000_11111#);
184 Result (Len) := Wide_Character'Val (R);
186 -- Codes in the range 16#800# - 16#FFFF#
187 -- UTF-8: 1110yyyy 10yyyyxx 10xxxxxx
188 -- UTF-16: yyyyyyyy_xxxxxxxx
190 elsif C <= 2#1110_1111# then
191 R := Unsigned_16 (C and 2#0000_1111#);
195 Result (Len) := Wide_Character'Val (R);
197 -- Make sure that we don't have a result in the forbidden range
198 -- reserved for UTF-16 surrogate characters.
200 if R in 16#D800# .. 16#DF00# then
201 Raise_Encoding_Error (Iptr - 3);
204 -- Codes in the range 16#10000# - 16#10FFFF#
205 -- UTF-8: 11110zzz 10zzyyyy 10yyyyxx 10xxxxxx
206 -- UTF-16: 110110zz_zzyyyyyy 110111yy_xxxxxxxx
207 -- Note: zzzz in the output is input zzzzz - 1
209 elsif C <= 2#11110_111# then
210 R := Unsigned_16 (C and 2#00000_111#);
213 -- R now has zzzzzyyyy
215 R := R - 2#0000_1_0000#;
217 -- R now has zzzzyyyy (zzzz minus one for the output)
221 -- R now has zzzzyyyyyyyyxx
226 (2#110110_00_0000_0000# or Shift_Right (R, 4));
232 Wide_Character'Val (2#110111_00_0000_0000# or R);
234 -- Any other code is an error
237 Raise_Encoding_Error (Iptr - 1);
241 return Result (1 .. Len);
244 -- Convert from UTF-16 to UTF-8/UTF-16-BE/LE
247 (Item : UTF_16_Wide_String;
248 Output_Scheme : Encoding_Scheme;
249 Output_BOM : Boolean := False) return UTF_String
252 if Output_Scheme = UTF_8 then
253 return Convert (Item, Output_BOM);
255 return From_UTF_16 (Item, Output_Scheme, Output_BOM);
259 -- Convert from UTF-16 to UTF-8
262 (Item : UTF_16_Wide_String;
263 Output_BOM : Boolean := False) return UTF_8_String
265 Result : UTF_8_String (1 .. 3 * Item'Length + 3);
266 -- Worst case is 3 output codes for each input code + BOM space
269 -- Number of result codes stored
272 -- Pointer to next input character
274 C1, C2 : Unsigned_16;
277 yyyyyyyy : Unsigned_16;
278 xxxxxxxx : Unsigned_16;
279 -- Components of double length case
284 -- Skip BOM at start of input
286 if Item'Length > 0 and then Item (Iptr) = BOM_16 (1) then
290 -- Generate output BOM if required
293 Result (1 .. 3) := BOM_8;
299 -- Loop through input
301 while Iptr <= Item'Last loop
302 C1 := To_Unsigned_16 (Item (Iptr));
305 -- Codes in the range 16#0000# - 16#007F#
306 -- UTF-16: 000000000xxxxxxx
309 if C1 <= 16#007F# then
310 Result (Len + 1) := Character'Val (C1);
313 -- Codes in the range 16#80# - 16#7FF#
314 -- UTF-16: 00000yyyxxxxxxxx
315 -- UTF-8: 110yyyxx 10xxxxxx
317 elsif C1 <= 16#07FF# then
320 (2#110_00000# or Shift_Right (C1, 6));
323 (2#10_000000# or (C1 and 2#00_111111#));
326 -- Codes in the range 16#800# - 16#D7FF# or 16#E000# - 16#FFFF#
327 -- UTF-16: yyyyyyyyxxxxxxxx
328 -- UTF-8: 1110yyyy 10yyyyxx 10xxxxxx
330 elsif C1 <= 16#D7FF# or else C1 >= 16#E000# then
333 (2#1110_0000# or Shift_Right (C1, 12));
336 (2#10_000000# or (Shift_Right (C1, 6) and 2#00_111111#));
339 (2#10_000000# or (C1 and 2#00_111111#));
342 -- Codes in the range 16#10000# - 16#10FFFF#
343 -- UTF-16: 110110zzzzyyyyyy 110111yyxxxxxxxx
344 -- UTF-8: 11110zzz 10zzyyyy 10yyyyxx 10xxxxxx
345 -- Note: zzzzz in the output is input zzzz + 1
347 elsif C1 <= 2#110110_11_11111111# then
348 if Iptr > Item'Last then
349 Raise_Encoding_Error (Iptr - 1);
351 C2 := To_Unsigned_16 (Item (Iptr));
355 if (C2 and 2#111111_00_00000000#) /= 2#110111_00_00000000# then
356 Raise_Encoding_Error (Iptr - 1);
359 zzzzz := (Shift_Right (C1, 6) and 2#1111#) + 1;
360 yyyyyyyy := ((Shift_Left (C1, 2) and 2#111111_00#)
362 (Shift_Right (C2, 8) and 2#000000_11#));
363 xxxxxxxx := C2 and 2#11111111#;
367 (2#11110_000# or (Shift_Right (zzzzz, 2)));
370 (2#10_000000# or Shift_Left (zzzzz and 2#11#, 4)
371 or Shift_Right (yyyyyyyy, 4));
374 (2#10_000000# or Shift_Left (yyyyyyyy and 2#1111#, 4)
375 or Shift_Right (xxxxxxxx, 6));
378 (2#10_000000# or (xxxxxxxx and 2#00_111111#));
381 -- Error if input in 16#DC00# - 16#DFFF# (2nd surrogate with no 1st)
384 Raise_Encoding_Error (Iptr - 2);
388 return Result (1 .. Len);
391 end Ada.Strings.UTF_Encoding.Conversions;