2008-05-30 Vladimir Makarov <vmakarov@redhat.com>
[official-gcc.git] / gcc / ada / scn.adb
blob805caab574e698d012c038f2de9a8be534e9d8a8
1 ------------------------------------------------------------------------------
2 -- --
3 -- GNAT COMPILER COMPONENTS --
4 -- --
5 -- S C N --
6 -- --
7 -- B o d y --
8 -- --
9 -- Copyright (C) 1992-2008, 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 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. 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 COPYING3. If not, go to --
19 -- http://www.gnu.org/licenses for a complete copy of the license. --
20 -- --
21 -- GNAT was originally developed by the GNAT team at New York University. --
22 -- Extensive contributions were provided by Ada Core Technologies Inc. --
23 -- --
24 ------------------------------------------------------------------------------
26 with Atree; use Atree;
27 with Csets; use Csets;
28 with Hostparm; use Hostparm;
29 with Namet; use Namet;
30 with Opt; use Opt;
31 with Output; use Output;
32 with Restrict; use Restrict;
33 with Rident; use Rident;
34 with Scans; use Scans;
35 with Sinfo; use Sinfo;
36 with Sinput; use Sinput;
37 with Uintp; use Uintp;
39 with GNAT.Byte_Order_Mark; use GNAT.Byte_Order_Mark;
41 with System.WCh_Con; use System.WCh_Con;
43 package body Scn is
45 use ASCII;
47 Used_As_Identifier : array (Token_Type) of Boolean;
48 -- Flags set True if a given keyword is used as an identifier (used to
49 -- make sure that we only post an error message for incorrect use of a
50 -- keyword as an identifier once for a given keyword).
52 procedure Check_End_Of_Line;
53 -- Called when end of line encountered. Checks that line is not too long,
54 -- and that other style checks for the end of line are met.
56 function Determine_License return License_Type;
57 -- Scan header of file and check that it has an appropriate GNAT-style
58 -- header with a proper license statement. Returns GPL, Unrestricted,
59 -- or Modified_GPL depending on header. If none of these, returns Unknown.
61 procedure Error_Long_Line;
62 -- Signal error of excessively long line
64 -----------------------
65 -- Check_End_Of_Line --
66 -----------------------
68 procedure Check_End_Of_Line is
69 Len : constant Int := Int (Scan_Ptr) - Int (Current_Line_Start);
70 begin
71 if Style_Check then
72 Style.Check_Line_Terminator (Len);
73 elsif Len > Max_Line_Length then
74 Error_Long_Line;
75 end if;
76 end Check_End_Of_Line;
78 -----------------------
79 -- Determine_License --
80 -----------------------
82 function Determine_License return License_Type is
83 GPL_Found : Boolean := False;
84 Result : License_Type;
86 function Contains (S : String) return Boolean;
87 -- See if current comment contains successive non-blank characters
88 -- matching the contents of S. If so leave Scan_Ptr unchanged and
89 -- return True, otherwise leave Scan_Ptr unchanged and return False.
91 procedure Skip_EOL;
92 -- Skip to line terminator character
94 --------------
95 -- Contains --
96 --------------
98 function Contains (S : String) return Boolean is
99 CP : Natural;
100 SP : Source_Ptr;
101 SS : Source_Ptr;
103 begin
104 -- Loop to check characters. This loop is terminated by end of
105 -- line, and also we need to check for the EOF case, to take
106 -- care of files containing only comments.
108 SP := Scan_Ptr;
109 while Source (SP) /= CR and then
110 Source (SP) /= LF and then
111 Source (SP) /= EOF
112 loop
113 if Source (SP) = S (S'First) then
114 SS := SP;
115 CP := S'First;
117 loop
118 SS := SS + 1;
119 CP := CP + 1;
121 if CP > S'Last then
122 return True;
123 end if;
125 while Source (SS) = ' ' loop
126 SS := SS + 1;
127 end loop;
129 exit when Source (SS) /= S (CP);
130 end loop;
131 end if;
133 SP := SP + 1;
134 end loop;
136 return False;
137 end Contains;
139 --------------
140 -- Skip_EOL --
141 --------------
143 procedure Skip_EOL is
144 begin
145 while Source (Scan_Ptr) /= CR
146 and then Source (Scan_Ptr) /= LF
147 and then Source (Scan_Ptr) /= EOF
148 loop
149 Scan_Ptr := Scan_Ptr + 1;
150 end loop;
151 end Skip_EOL;
153 -- Start of processing for Determine_License
155 begin
156 loop
157 if Source (Scan_Ptr) /= '-'
158 or else Source (Scan_Ptr + 1) /= '-'
159 then
160 if GPL_Found then
161 Result := GPL;
162 exit;
163 else
164 Result := Unknown;
165 exit;
166 end if;
168 elsif Contains ("Asaspecialexception") then
169 if GPL_Found then
170 Result := Modified_GPL;
171 exit;
172 end if;
174 elsif Contains ("GNUGeneralPublicLicense") then
175 GPL_Found := True;
177 elsif
178 Contains
179 ("ThisspecificationisadaptedfromtheAdaSemanticInterface")
180 or else
181 Contains
182 ("ThisspecificationisderivedfromtheAdaReferenceManual")
183 then
184 Result := Unrestricted;
185 exit;
186 end if;
188 Skip_EOL;
190 Check_End_Of_Line;
192 if Source (Scan_Ptr) /= EOF then
194 -- We have to take into account a degenerate case when the source
195 -- file contains only comments and no Ada code.
197 declare
198 Physical : Boolean;
200 begin
201 Skip_Line_Terminators (Scan_Ptr, Physical);
203 -- If we are at start of physical line, update scan pointers
204 -- to reflect the start of the new line.
206 if Physical then
207 Current_Line_Start := Scan_Ptr;
208 Start_Column := Scanner.Set_Start_Column;
209 First_Non_Blank_Location := Scan_Ptr;
210 end if;
211 end;
212 end if;
213 end loop;
215 return Result;
216 end Determine_License;
218 ----------------------------
219 -- Determine_Token_Casing --
220 ----------------------------
222 function Determine_Token_Casing return Casing_Type is
223 begin
224 return Scanner.Determine_Token_Casing;
225 end Determine_Token_Casing;
227 ---------------------
228 -- Error_Long_Line --
229 ---------------------
231 procedure Error_Long_Line is
232 begin
233 Error_Msg
234 ("this line is too long",
235 Current_Line_Start + Source_Ptr (Max_Line_Length));
236 end Error_Long_Line;
238 ------------------------
239 -- Initialize_Scanner --
240 ------------------------
242 procedure Initialize_Scanner
243 (Unit : Unit_Number_Type;
244 Index : Source_File_Index)
246 GNAT_Hedr : constant Text_Buffer (1 .. 78) := (others => '-');
248 begin
249 Scanner.Initialize_Scanner (Index);
251 if Index /= Internal_Source_File then
252 Set_Unit (Index, Unit);
253 end if;
255 Current_Source_Unit := Unit;
257 -- Set default for Comes_From_Source (except if we are going to process
258 -- an artificial string internally created within the compiler and
259 -- placed into internal source duffer). All nodes built now until we
260 -- reenter the analyzer will have Comes_From_Source set to True
262 if Index /= Internal_Source_File then
263 Set_Comes_From_Source_Default (True);
264 end if;
266 -- Check license if GNAT type header possibly present
268 if Source_Last (Index) - Scan_Ptr > 80
269 and then Source (Scan_Ptr .. Scan_Ptr + 77) = GNAT_Hedr
270 then
271 Set_License (Current_Source_File, Determine_License);
272 end if;
274 -- Check for BOM
276 declare
277 BOM : BOM_Kind;
278 Len : Natural;
279 Tst : String (1 .. 5);
281 begin
282 for J in 1 .. 5 loop
283 Tst (J) := Source (Scan_Ptr + Source_Ptr (J) - 1);
284 end loop;
286 Read_BOM (Tst, Len, BOM, False);
288 case BOM is
289 when UTF8_All =>
290 Scan_Ptr := Scan_Ptr + Source_Ptr (Len);
291 Wide_Character_Encoding_Method := WCEM_UTF8;
292 Upper_Half_Encoding := True;
294 when UTF16_LE | UTF16_BE =>
295 Set_Standard_Error;
296 Write_Line ("UTF-16 encoding format not recognized");
297 Set_Standard_Output;
298 raise Unrecoverable_Error;
300 when UTF32_LE | UTF32_BE =>
301 Set_Standard_Error;
302 Write_Line ("UTF-32 encoding format not recognized");
303 Set_Standard_Output;
304 raise Unrecoverable_Error;
306 when Unknown =>
307 null;
309 when others =>
310 raise Program_Error;
311 end case;
312 end;
314 -- Because of the License stuff above, Scng.Initialize_Scanner cannot
315 -- call Scan. Scan initial token (note this initializes Prev_Token,
316 -- Prev_Token_Ptr).
318 -- There are two reasons not to do the Scan step in case if we
319 -- initialize the scanner for the internal source buffer:
321 -- - The artificial string may not be created by the compiler in this
322 -- buffer when we call Initialize_Scanner
324 -- - For these artificial strings a special way of scanning is used, so
325 -- the standard step of the scanner may just break the algorithm of
326 -- processing these strings.
328 if Index /= Internal_Source_File then
329 Scan;
330 end if;
332 -- Clear flags for reserved words used as identifiers
334 for J in Token_Type loop
335 Used_As_Identifier (J) := False;
336 end loop;
337 end Initialize_Scanner;
339 -----------------------
340 -- Obsolescent_Check --
341 -----------------------
343 procedure Obsolescent_Check (S : Source_Ptr) is
344 begin
345 -- This is a pain in the neck case, since we normally need a node to
346 -- call Check_Restrictions, and all we have is a source pointer. The
347 -- easiest thing is to construct a dummy node. A bit kludgy, but this
348 -- is a marginal case. It's not worth trying to do things more cleanly.
350 Check_Restriction (No_Obsolescent_Features, New_Node (N_Empty, S));
351 end Obsolescent_Check;
353 ---------------
354 -- Post_Scan --
355 ---------------
357 procedure Post_Scan is
358 begin
359 case Token is
360 when Tok_Char_Literal =>
361 Token_Node := New_Node (N_Character_Literal, Token_Ptr);
362 Set_Char_Literal_Value (Token_Node, UI_From_CC (Character_Code));
363 Set_Chars (Token_Node, Token_Name);
365 when Tok_Identifier =>
366 Token_Node := New_Node (N_Identifier, Token_Ptr);
367 Set_Chars (Token_Node, Token_Name);
369 when Tok_Real_Literal =>
370 Token_Node := New_Node (N_Real_Literal, Token_Ptr);
371 Set_Realval (Token_Node, Real_Literal_Value);
373 when Tok_Integer_Literal =>
374 Token_Node := New_Node (N_Integer_Literal, Token_Ptr);
375 Set_Intval (Token_Node, Int_Literal_Value);
377 when Tok_String_Literal =>
378 Token_Node := New_Node (N_String_Literal, Token_Ptr);
379 Set_Has_Wide_Character (Token_Node, Wide_Character_Found);
380 Set_Strval (Token_Node, String_Literal_Id);
382 when Tok_Operator_Symbol =>
383 Token_Node := New_Node (N_Operator_Symbol, Token_Ptr);
384 Set_Chars (Token_Node, Token_Name);
385 Set_Strval (Token_Node, String_Literal_Id);
387 when others =>
388 null;
389 end case;
390 end Post_Scan;
392 ------------------------------
393 -- Scan_Reserved_Identifier --
394 ------------------------------
396 procedure Scan_Reserved_Identifier (Force_Msg : Boolean) is
397 Token_Chars : constant String := Token_Type'Image (Token);
399 begin
400 -- We have in Token_Chars the image of the Token name, i.e. Tok_xxx.
401 -- This code extracts the xxx and makes an identifier out of it.
403 Name_Len := 0;
405 for J in 5 .. Token_Chars'Length loop
406 Name_Len := Name_Len + 1;
407 Name_Buffer (Name_Len) := Fold_Lower (Token_Chars (J));
408 end loop;
410 Token_Name := Name_Find;
412 if not Used_As_Identifier (Token) or else Force_Msg then
413 Error_Msg_Name_1 := Token_Name;
414 Error_Msg_SC ("reserved word* cannot be used as identifier!");
415 Used_As_Identifier (Token) := True;
416 end if;
418 Token := Tok_Identifier;
419 Token_Node := New_Node (N_Identifier, Token_Ptr);
420 Set_Chars (Token_Node, Token_Name);
421 end Scan_Reserved_Identifier;
423 end Scn;