Skip to content

Commit a56ddc1

Browse files
ahuoguoahuoguo
committed
move wat parser from gensym
1 parent 1a56648 commit a56ddc1

11 files changed

+1149
-236
lines changed

‎.gitignore

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -4,3 +4,5 @@
44
.vscode/
55
project/
66
target/
7+
grammar/antlr-4.13.0-complete.jar
8+
.bsp

‎grammar/README.md

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,4 @@
1+
# Parser Documentation
2+
3+
The antlr java binary can be downloaded at https://github.com/antlr/website-antlr4/blob/gh-pages/download/antlr-4.13.0-complete.jar
4+

‎grammar/WatLexer.g4

Lines changed: 383 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,383 @@
1+
/*
2+
Copyright (c) 2019 Renata Hodovan.
3+
All rights reserved.
4+
5+
Modified by Guannan Wei (2023).
6+
7+
Modified by Mikail Khan (2023).
8+
9+
Redistribution and use in source and binary forms, with or without
10+
modification, are permitted provided that the following conditions are met:
11+
12+
1. Redistributions of source code must retain the above copyright notice, this
13+
list of conditions and the following disclaimer.
14+
15+
2. Redistributions in binary form must reproduce the above copyright notice,
16+
this list of conditions and the following disclaimer in the documentation
17+
and/or other materials provided with the distribution.
18+
19+
3. Neither the name of the copyright holder nor the names of its contributors
20+
may be used to endorse or promote products derived from this software
21+
without specific prior written permission.
22+
23+
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
24+
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
25+
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
26+
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
27+
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
28+
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
29+
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
30+
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
31+
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
32+
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
33+
*/
34+
35+
lexer grammar WatLexer;
36+
37+
LPAR : '(' ;
38+
RPAR : ')' ;
39+
40+
NAT : Nat ;
41+
INT : Int ;
42+
FLOAT : Float ;
43+
STRING_ : String_ ;
44+
VALUE_TYPE : NXX ;
45+
CONST : NXX '.const' ;
46+
SYMBOLIC : NXX '.symbolic' ;
47+
48+
FUNCREF: 'funcref' ;
49+
EXTERNREF: 'externref' ;
50+
MUT: 'mut' ;
51+
REF: 'ref' ;
52+
CONT: 'cont' ;
53+
NULL: 'null' ;
54+
55+
NOP: 'nop' ;
56+
SYM_ASSERT: 'sym_assert' ;
57+
ALLOC: 'alloc' ;
58+
FREE: 'free' ;
59+
UNREACHABLE: 'unreachable' ;
60+
DROP: 'drop' ;
61+
BLOCK: 'block' ;
62+
LOOP: 'loop' ;
63+
FOR : 'for';
64+
VBAR: '|';
65+
END: 'end' ;
66+
BR: 'br' ;
67+
BR_IF: 'br_if' ;
68+
BR_TABLE: 'br_table' ;
69+
RETURN: 'return' ;
70+
IF: 'if' ;
71+
THEN: 'then' ;
72+
ELSE: 'else' ;
73+
SELECT: '.select' ;
74+
CALL: 'call' ;
75+
CALL_INDIRECT: 'call_indirect' ;
76+
RETURN_CALL: 'return_call' ;
77+
RETURN_CALL_INDIRECT: 'return_call_indirect' ;
78+
REFFUNC: 'ref.func' ;
79+
CALLREF: 'call_ref' ;
80+
RESUME: 'resume' ;
81+
ON: 'on' ;
82+
CONTNEW: 'cont.new' ;
83+
CONTBIND: 'cont.bind' ;
84+
SUSPEND: 'suspend' ;
85+
REFNULL: 'ref.null' ;
86+
REFISNULL: 'ref.is_null' ;
87+
88+
// resumable try-catch extension:
89+
TRY: 'try' ;
90+
CATCH: 'catch' ;
91+
THROW: 'throw' ;
92+
RESUME0: 'resume0' ;
93+
94+
LOCAL_GET: 'local.get' ;
95+
LOCAL_SET: 'local.set' ;
96+
LOCAL_TEE: 'local.tee' ;
97+
GLOBAL_GET: 'global.get' ;
98+
GLOBAL_SET: 'global.set' ;
99+
100+
LOAD : '.load' (MEM_SIZE UNDERSCORE SIGN_POSTFIX)?;
101+
102+
STORE : '.store' MEM_SIZE? ;
103+
UNDERSCORE : '_';
104+
105+
OFFSET_EQ : 'offset=' ;
106+
ALIGN_EQ : 'align=' ;
107+
108+
SIGN_POSTFIX : 's' | 'u';
109+
110+
MEM_SIZE : '8' | '16' | '32' | '64';
111+
112+
I32 : 'i32';
113+
I64 : 'i64';
114+
115+
F32 : 'f32';
116+
F64 : 'f64';
117+
118+
IXX : I32 | I64;
119+
FXX : F32 | F64;
120+
121+
OP_EQZ : '.eqz';
122+
123+
OP_EQ : '.eq';
124+
OP_NE : '.ne';
125+
OP_LT : '.lt';
126+
OP_LTS : '.lt_s';
127+
OP_LTU : '.lt_u';
128+
OP_LE : '.le';
129+
OP_LES : '.le_s';
130+
OP_LEU : '.le_u';
131+
OP_GT : '.gt';
132+
OP_GTS : '.gt_s';
133+
OP_GTU : '.gt_u';
134+
OP_GE : '.ge';
135+
OP_GES : '.ge_s';
136+
OP_GEU : '.ge_u';
137+
138+
OP_CLZ : '.clz';
139+
OP_CTZ : '.ctz';
140+
OP_POPCNT : '.popcnt';
141+
OP_NEG : '.neg';
142+
OP_ABS : '.abs';
143+
OP_SQRT : '.sqrt';
144+
OP_CEIL : '.ceil';
145+
OP_FLOOR : '.floor';
146+
OP_TRUNC : '.trunc';
147+
OP_NEAREST : '.nearest';
148+
149+
OP_ADD : '.add';
150+
OP_SUB : '.sub';
151+
OP_MUL : '.mul';
152+
OP_DIV : '.div';
153+
OP_DIV_S : '.div_s';
154+
OP_DIV_U : '.div_u';
155+
OP_REM_S : '.rem_s';
156+
OP_REM_U : '.rem_u';
157+
OP_AND : '.and';
158+
OP_OR : '.or';
159+
OP_XOR : '.xor';
160+
OP_SHL : '.shl';
161+
OP_SHR_S : '.shr_s';
162+
OP_SHR_U : '.shr_u';
163+
OP_ROTL : '.rotl';
164+
OP_ROTR : '.rotr';
165+
OP_MIN : '.min';
166+
OP_MAX : '.max';
167+
OP_COPYSIGN : '.copysign';
168+
169+
OP_WRAP : '.wrap_';
170+
OP_TRUNC_ : '.trunc_';
171+
OP_TRUNC_SAT : '.trunc_sat_';
172+
OP_CONVERT : '.convert_';
173+
OP_EXTEND : '.extend_';
174+
OP_DEMOTE : '.demote_';
175+
OP_PROMOTE : '.promote_';
176+
OP_REINTER : '.reinterpret_';
177+
178+
MEMORY_SIZE : 'memory.size' ;
179+
MEMORY_GROW : 'memory.grow' ;
180+
MEMORY_FILL : 'memory.fill' ;
181+
MEMORY_COPY : 'memory.copy' ;
182+
MEMORY_INIT : 'memory.init' ;
183+
184+
TEST : IXX OP_EQZ;
185+
186+
COMPARE :
187+
IXX '.eq'
188+
| IXX '.ne'
189+
| IXX '.lt_s'
190+
| IXX '.lt_u'
191+
| IXX '.le_s'
192+
| IXX '.le_u'
193+
| IXX '.gt_s'
194+
| IXX '.gt_u'
195+
| IXX '.ge_s'
196+
| IXX '.ge_u'
197+
| FXX '.eq'
198+
| FXX '.ne'
199+
| FXX '.lt'
200+
| FXX '.le'
201+
| FXX '.gt'
202+
| FXX '.ge'
203+
;
204+
205+
UNARY
206+
: IXX '.clz'
207+
| IXX '.ctz'
208+
| IXX '.popcnt'
209+
| FXX '.neg'
210+
| FXX '.abs'
211+
| FXX '.sqrt'
212+
| FXX '.ceil'
213+
| FXX '.floor'
214+
| FXX '.trunc'
215+
| FXX '.nearest'
216+
;
217+
218+
BINARY
219+
: IXX '.add'
220+
| IXX '.sub'
221+
| IXX '.mul'
222+
| IXX '.div_s'
223+
| IXX '.div_u'
224+
| IXX '.rem_s'
225+
| IXX '.rem_u'
226+
| IXX '.and'
227+
| IXX '.or'
228+
| IXX '.xor'
229+
| IXX '.shl'
230+
| IXX '.shr_s'
231+
| IXX '.shr_u'
232+
| IXX '.rotl'
233+
| IXX '.rotr'
234+
| FXX '.add'
235+
| FXX '.sub'
236+
| FXX '.mul'
237+
| FXX '.div'
238+
| FXX '.min'
239+
| FXX '.max'
240+
| FXX '.copysign'
241+
;
242+
243+
CONVERT
244+
: I32 '.wrap_' I64
245+
| IXX '.trunc_' FXX UNDERSCORE SIGN_POSTFIX
246+
| IXX '.trunc_sat_' FXX UNDERSCORE SIGN_POSTFIX
247+
| I64 '.extend_' I32 UNDERSCORE SIGN_POSTFIX
248+
| FXX '.convert_' IXX UNDERSCORE SIGN_POSTFIX
249+
| F32 '.demote_' F64
250+
| F64 '.promote_' F32
251+
| F32 '.reinterpret_' I32
252+
| F64 '.reinterpret_' I64
253+
| I32 '.reinterpret_' F32
254+
| I64 '.reinterpret_' F64
255+
;
256+
257+
TYPE: 'type' ;
258+
FUNC: 'func' ;
259+
EXTERN: 'extern' ;
260+
START_: 'start' ;
261+
PARAM: 'param' ;
262+
RESULT: 'result' ;
263+
LOCAL: 'local' ;
264+
GLOBAL: 'global' ;
265+
TABLE: 'table' ;
266+
MEMORY: 'memory' ;
267+
ELEM: 'elem' ;
268+
DATA: 'data' ;
269+
OFFSET: 'offset' ;
270+
IMPORT: 'import' ;
271+
EXPORT: 'export' ;
272+
TAG: 'tag' ;
273+
274+
DECLARE: 'declare' ;
275+
276+
MODULE : 'module' ;
277+
BIN : 'binary' ;
278+
QUOTE : 'quote' ;
279+
DEFINITION : 'definition' ;
280+
INSTANCE : 'instance' ;
281+
282+
SCRIPT: 'script' ;
283+
REGISTER: 'register' ;
284+
INVOKE: 'invoke' ;
285+
GET: 'get' ;
286+
ASSERT_MALFORMED: 'assert_malformed' ;
287+
ASSERT_INVALID: 'assert_invalid' ;
288+
ASSERT_UNLINKABLE: 'assert_unlinkable' ;
289+
ASSERT_RETURN: 'assert_return' ;
290+
ASSERT_RETURN_CANONICAL_NAN: 'assert_return_canonical_nan' ;
291+
ASSERT_RETURN_ARITHMETIC_NAN: 'assert_return_arithmetic_nan' ;
292+
ASSERT_TRAP: 'assert_trap' ;
293+
ASSERT_EXHAUSTION: 'assert_exhaustion' ;
294+
INPUT: 'input' ;
295+
OUTPUT: 'output' ;
296+
297+
VAR : Name ;
298+
299+
V128 : 'v128' ;
300+
301+
SPACE
302+
: [ \t\r\n]+ -> skip
303+
;
304+
305+
COMMENT
306+
: ( '(;' .*? ';)'
307+
| ';;' .*? '\n')-> skip
308+
;
309+
310+
fragment Symbol
311+
: '.' | '+' | '-' | '*' | '/' | '\\' | '^' | '~' | '=' | '<' | '>' | '!' | '?' | '@' | '#' | '$' | '%' | '&' | '|' | ':' | '\'' | '`'
312+
;
313+
314+
fragment Num
315+
: Digit ('_'? Digit)*
316+
;
317+
318+
fragment HexNum
319+
: HexDigit ('_'? HexDigit)*
320+
;
321+
322+
fragment Sign
323+
: '+' | '-'
324+
;
325+
326+
fragment Digit
327+
: [0-9]
328+
;
329+
330+
fragment HexDigit
331+
: [0-9a-fA-F]
332+
;
333+
334+
fragment Letter
335+
: [a-zA-Z]
336+
;
337+
338+
fragment Nat : Num | ('0x' HexNum) ;
339+
fragment Int : Sign Nat ;
340+
fragment Frac : Num ;
341+
fragment HexFrac : HexNum ;
342+
343+
fragment Float
344+
: Sign? Num '.' Frac?
345+
| Sign? Num ('.' Frac?)? ('e' | 'E') Sign? Num
346+
| Sign? '0x' HexNum '.' HexFrac?
347+
| Sign? '0x' HexNum ('.' HexFrac?)? ('p' | 'P') Sign? Num
348+
| Sign? 'inf'
349+
| Sign? 'nan'
350+
| Sign? 'nan:' '0x' HexNum
351+
;
352+
353+
fragment String_
354+
: '"' ( Char | '\n' | '\t' | '\\' | '\'' | '\\' HexDigit HexDigit | '\\u{' HexDigit+ '}' )* '"'
355+
;
356+
357+
fragment Name
358+
: '$' (Letter | Digit | '_' | Symbol)+
359+
;
360+
361+
fragment Escape : [nrt'"\\] ;
362+
363+
fragment NXX : IXX | FXX ;
364+
/* fragment MIXX : 'i' ('8' | '16' | '32' | '64') ; */
365+
/* fragment MFXX : 'f' ('32' | '64') ; */
366+
/* fragment SIGN : 's' | 'u' ; */
367+
368+
fragment Char : ~["'\\\u0000-\u001f\u007f-\u00ff] ;
369+
fragment Ascii : [\u0000-\u007f] ;
370+
fragment Ascii_no_nl : [\u0000-\u0009\u000b-\u007f] ;
371+
fragment Utf8Cont : [\u0080-\u00bf] ;
372+
fragment Utf8 : Ascii | Utf8Enc ;
373+
fragment Utf8_no_nl : Ascii_no_nl | Utf8Enc ;
374+
375+
fragment Utf8Enc
376+
: [\u00c2-\u00df] Utf8Cont
377+
| [\u00e0] [\u00a0-\u00bf] Utf8Cont
378+
| [\u00ed] [\u0080-\u009f] Utf8Cont
379+
| [\u00e1-\u00ec\u00ee-\u00ef] Utf8Cont Utf8Cont
380+
| [\u00f0] [\u0090-\u00bf] Utf8Cont Utf8Cont
381+
| [\u00f4] [\u0080-\u008f] Utf8Cont Utf8Cont
382+
| [\u00f1-\u00f3] Utf8Cont Utf8Cont Utf8Cont
383+
;

0 commit comments

Comments
 (0)