1 ------------------------------------------------------------------------------
3 -- GNAT RUN-TIME COMPONENTS --
5 -- ADA.STRINGS.UTF_ENCODING.CONVERSIONS --
9 -- Copyright (C) 2010-2011, 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
46 if Input_Scheme
= Output_Scheme
then
49 -- For remaining cases, one or other of the operands is UTF-16BE/LE
50 -- encoded, so go through UTF-16 intermediate.
53 return Convert
(UTF_16_Wide_String
'(Convert (Item, Input_Scheme)),
54 Output_Scheme, Output_BOM);
58 -- Convert from UTF-8/UTF-16BE/LE to UTF-16
62 Input_Scheme : Encoding_Scheme;
63 Output_BOM : Boolean := False) return UTF_16_Wide_String
66 if Input_Scheme = UTF_8 then
67 return Convert (Item, Output_BOM);
69 return To_UTF_16 (Item, Input_Scheme, Output_BOM);
73 -- Convert from UTF-8 to UTF-16
77 Output_BOM : Boolean := False) return UTF_16_Wide_String
79 Result : UTF_16_Wide_String (1 .. Item'Length + 1);
80 -- Maximum length of result, including possible BOM
83 -- Number of characters stored so far in Result
86 -- Next character to process in Item
94 procedure Get_Continuation;
95 -- Reads a continuation byte of the form 10xxxxxx, shifts R left by 6
96 -- bits, and or's in the xxxxxx to the low order 6 bits. On return Ptr
97 -- is incremented. Raises exception if continuation byte does not exist
100 ----------------------
101 -- Get_Continuation --
102 ----------------------
104 procedure Get_Continuation is
106 if Iptr > Item'Last then
107 Raise_Encoding_Error (Iptr - 1);
110 C := To_Unsigned_8 (Item (Iptr));
113 if C < 2#10_000000# or else C > 2#10_111111# then
114 Raise_Encoding_Error (Iptr - 1);
118 Shift_Left (R, 6) or Unsigned_16 (C and 2#00_111111#);
121 end Get_Continuation;
123 -- Start of processing for Convert
126 -- Output BOM if required
130 Result (Len) := BOM_16 (1);
137 if Item'Length >= 3 and then Item (Iptr .. Iptr + 2) = BOM_8 then
142 elsif Item'Length >= 2
143 and then (Item (Iptr .. Iptr + 1) = BOM_16BE
145 Item (Iptr .. Iptr + 1) = BOM_16LE)
147 Raise_Encoding_Error (Iptr);
155 while Iptr <= Item'Last loop
156 C := To_Unsigned_8 (Item (Iptr));
159 -- Codes in the range 16#00# - 16#7F#
161 -- UTF-16: 00000000_0xxxxxxx
165 Result (Len) := Wide_Character'Val (C);
167 -- No initial code can be of the form 10xxxxxx. Such codes are used
168 -- only for continuations.
170 elsif C <= 2#10_111111# then
171 Raise_Encoding_Error (Iptr - 1);
173 -- Codes in the range 16#80# - 16#7FF#
174 -- UTF-8: 110yyyxx 10xxxxxx
175 -- UTF-16: 00000yyy_xxxxxxxx
177 elsif C <= 2#110_11111# then
178 R := Unsigned_16 (C and 2#000_11111#);
181 Result (Len) := Wide_Character'Val (R);
183 -- Codes in the range 16#800# - 16#FFFF#
184 -- UTF-8: 1110yyyy 10yyyyxx 10xxxxxx
185 -- UTF-16: yyyyyyyy_xxxxxxxx
187 elsif C <= 2#1110_1111# then
188 R := Unsigned_16 (C and 2#0000_1111#);
192 Result (Len) := Wide_Character'Val (R);
194 -- Make sure that we don't have a result in the forbidden range
195 -- reserved for UTF-16 surrogate characters.
197 if R in 16#D800# .. 16#DF00# then
198 Raise_Encoding_Error (Iptr - 3);
201 -- Codes in the range 16#10000# - 16#10FFFF#
202 -- UTF-8: 11110zzz 10zzyyyy 10yyyyxx 10xxxxxx
203 -- UTF-16: 110110zz_zzyyyyyy 110111yy_xxxxxxxx
204 -- Note: zzzz in the output is input zzzzz - 1
206 elsif C <= 2#11110_111# then
207 R := Unsigned_16 (C and 2#00000_111#);
210 -- R now has zzzzzyyyy
212 R := R - 2#0000_1_0000#;
214 -- R now has zzzzyyyy (zzzz minus one for the output)
218 -- R now has zzzzyyyyyyyyxx
223 (2#110110_00_0000_0000# or Shift_Right (R, 4));
229 Wide_Character'Val (2#110111_00_0000_0000# or R);
231 -- Any other code is an error
234 Raise_Encoding_Error (Iptr - 1);
238 return Result (1 .. Len);
241 -- Convert from UTF-16 to UTF-8/UTF-16-BE/LE
244 (Item : UTF_16_Wide_String;
245 Output_Scheme : Encoding_Scheme;
246 Output_BOM : Boolean := False) return UTF_String
249 if Output_Scheme = UTF_8 then
250 return Convert (Item, Output_BOM);
252 return From_UTF_16 (Item, Output_Scheme, Output_BOM);
256 -- Convert from UTF-16 to UTF-8
259 (Item : UTF_16_Wide_String;
260 Output_BOM : Boolean := False) return UTF_8_String
262 Result : UTF_8_String (1 .. 3 * Item'Length + 3);
263 -- Worst case is 3 output codes for each input code + BOM space
266 -- Number of result codes stored
269 -- Pointer to next input character
271 C1, C2 : Unsigned_16;
274 yyyyyyyy : Unsigned_16;
275 xxxxxxxx : Unsigned_16;
276 -- Components of double length case
281 -- Skip BOM at start of input
283 if Item'Length > 0 and then Item (Iptr) = BOM_16 (1) then
287 -- Generate output BOM if required
290 Result (1 .. 3) := BOM_8;
296 -- Loop through input
298 while Iptr <= Item'Last loop
299 C1 := To_Unsigned_16 (Item (Iptr));
302 -- Codes in the range 16#0000# - 16#007F#
303 -- UTF-16: 000000000xxxxxxx
306 if C1 <= 16#007F# then
307 Result (Len + 1) := Character'Val (C1);
310 -- Codes in the range 16#80# - 16#7FF#
311 -- UTF-16: 00000yyyxxxxxxxx
312 -- UTF-8: 110yyyxx 10xxxxxx
314 elsif C1 <= 16#07FF# then
317 (2#110_00000# or Shift_Right (C1, 6));
320 (2#10_000000# or (C1 and 2#00_111111#));
323 -- Codes in the range 16#800# - 16#D7FF# or 16#E000# - 16#FFFF#
324 -- UTF-16: yyyyyyyyxxxxxxxx
325 -- UTF-8: 1110yyyy 10yyyyxx 10xxxxxx
327 elsif C1 <= 16#D7FF# or else C1 >= 16#E000# then
330 (2#1110_0000# or Shift_Right (C1, 12));
333 (2#10_000000# or (Shift_Right (C1, 6) and 2#00_111111#));
336 (2#10_000000# or (C1 and 2#00_111111#));
339 -- Codes in the range 16#10000# - 16#10FFFF#
340 -- UTF-16: 110110zzzzyyyyyy 110111yyxxxxxxxx
341 -- UTF-8: 11110zzz 10zzyyyy 10yyyyxx 10xxxxxx
342 -- Note: zzzzz in the output is input zzzz + 1
344 elsif C1 <= 2#110110_11_11111111# then
345 if Iptr > Item'Last then
346 Raise_Encoding_Error (Iptr - 1);
348 C2 := To_Unsigned_16 (Item (Iptr));
352 if (C2 and 2#111111_00_00000000#) /= 2#110111_00_00000000# then
353 Raise_Encoding_Error (Iptr - 1);
356 zzzzz := (Shift_Right (C1, 6) and 2#1111#) + 1;
357 yyyyyyyy := ((Shift_Left (C1, 2) and 2#111111_00#)
359 (Shift_Right (C2, 8) and 2#000000_11#));
360 xxxxxxxx := C2 and 2#11111111#;
364 (2#11110_000# or (Shift_Right (zzzzz, 2)));
367 (2#10_000000# or Shift_Left (zzzzz and 2#11#, 4)
368 or Shift_Right (yyyyyyyy, 4));
371 (2#10_000000# or Shift_Left (yyyyyyyy and 2#1111#, 4)
372 or Shift_Right (xxxxxxxx, 6));
375 (2#10_000000# or (xxxxxxxx and 2#00_111111#));
378 -- Error if input in 16#DC00# - 16#DFFF# (2nd surrogate with no 1st)
381 Raise_Encoding_Error (Iptr - 2);
385 return Result (1 .. Len);
388 end Ada.Strings.UTF_Encoding.Conversions;