-
Notifications
You must be signed in to change notification settings - Fork 11
/
openqasm3.ungram
435 lines (348 loc) · 10.1 KB
/
openqasm3.ungram
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
// This grammar specifies the structure of the OpenQASM 3 concrete syntax tree.
// It does not specify parsing rules (ambiguities, precedence, etc are out of scope).
// Tokens are processed -- contextual keywords are recognised, compound operators glued.
//
// Legend:
//
// // -- comment
// Name = -- non-terminal definition
// 'ident' -- token (terminal)
// A B -- sequence
// A | B -- alternation
// A* -- zero or more repetition
// A? -- zero or one repetition
// (A) -- same as A
// label:A -- suggested name for field of AST node
// Notes for OQ3
//
// Introducing a literal keyword that is also a legal rust identifer here is not difficult, but easy, just do it.
// 'measure' below requires that the literal "measure" is present.
//
// However '++' takes more work.
// If you introduce a literal that cannot start a rust identifier, say `++`:
// 1. Map it to an identifer in `fn method_name` in sourcegen_ast.rs. For example:
// `"++" => "double_plus"`
// 2. Add ++ to KINDS_SRC in ast_src.rs
// 3. Add T![++] to `fn eat` in parser/parser.rs
// 4. Add T![++] to `fn nth_at` in parser/parser.rs
// 5. Add ++ to current_op in parser/src/grammar/expressions.rs.
// Specify binding power and associtivity.
// 6. Add ++ to BinExpr below if it's a binary op.
//
// Instead of `"++" => "double_plus"`, something like `myop:(++)` also works as long as `myop` is entered into the
// list of `manually_implemented` labels in sourcegen_ast.rs. But, I think it is preferable to
// use `"++" => "double_plus"` as discussed above.
// The operator must then of course be actually manually_implemented elsewhere.
// For example "then_branch" is marked as manually_implemented and it is handled (I think)
// in expr_ext.rs.
// Compare with qasm3Parser.g4
// qasm3Parser.g4 has expressions such as rangeExpression that are:
// > Special-case expressions that are only valid in certain contexts. These are
// > not in the expression tree, but can contain elements that are within it.
// Here, we *can* include these in the the expression tree (in the enum) because
// part of the syntax rules is enforced by the hand-written parser.
//********//
// Names //
//********//
// Some apparent tokens, such as 'ident' are intercepted in sourcegen_ast.rs and are
// treated specially.
Name =
'ident'
//*************************//
// [Stmt]s //
//*************************//
SourceFile =
statements:(Stmt*)
// FIXME: Stmt vs Statement ?
Stmt =
';'
| AliasDeclarationStatement
| AnnotationStatement
| AssignmentStmt
| Barrier
| BreakStmt
| Cal
| ClassicalDeclarationStatement
| ContinueStmt
| Def
| DefCal
| DefCalGrammar
| DelayStmt
| EndStmt
| ExprStmt
| ForStmt
| Gate
| IfStmt
| Include
| IODeclarationStatement
| LetStmt
| Measure
| PragmaStatement
| QuantumDeclarationStatement
| Reset
| SwitchCaseStmt
| VersionString
| WhileStmt
// "thelabel" is in the list of manually implemented elements.
AnnotationStatement =
thelabel:'fakekw'
// Like 'ident' the remainder is manually coded
PragmaStatement =
'pragma'
SwitchCaseStmt =
'switch' '(' control:Expr ')' '{' (CaseExpr*) ('default' default_block:BlockExpr)? '}'
// Not in `Expr` tree
CaseExpr =
'case' ExpressionList BlockExpr
BreakStmt =
'break' ';'
ContinueStmt =
'continue' ';'
DelayStmt =
'delay' Designator QubitList ';'
EndStmt =
'end' ';'
VersionString =
'OPENQASM' version:Version ';'
// Maybe 'int_number' '. 'int_number' would be better.
// But the lexer lexes this as a float number.
// Also 'float_number' causes error in nodes.rs for some reason.
Version =
'int_number'
// TODO: make this more precise.
// I was getting errors when trying
Reset =
'reset' GateOperand
Measure =
'measure' qubit:Expr ';'
// FIXME, args to barrier are more general that this.
Barrier =
'barrier' QubitList? ';'
Cal =
'cal' (body:BlockExpr | ';')
DefCalGrammar =
'defcalgrammar' file:FilePath ';'
Include =
'include' file:FilePath ';'
// code for extracting string from FilePath is in expr_ext.rs
FilePath =
'string'
// Subroutine definition
Def =
'def' Name TypedParamList ReturnSignature?
(body:BlockExpr | ';')
// Defcal definition
DefCal =
'defcal' Name ParamList QubitList ReturnSignature?
(body:BlockExpr | ';')
// Gate definition
// sourceget_ast.rs is not smart enough to handle two ParamList's here.
// It would return the same thing for angle_params and qubit_args.
// So in sourcegen_ast.rs we enter angle_params and qubit_args in `manually_implemented`.
// We have to choose sort of unique names. If, say qubit_args appears elsewhere,
// in a method intended to be automatically generated, then autogen will skip it as well.
// Finally, we code the methods by hand in expr_ext.rs. We are not required to
// give the methods in expr_ext.rs the same names as the labels here, but we do so,
// to make this slightly less complex.
Gate =
'gate' Name angle_params:ParamList qubit_args:ParamList
(body:BlockExpr | ';')
// Paren delimited list
ParamList =
'(' (Param (',' Param)* ','?)? ')'
TypedParamList =
'(' (TypedParam (',' TypedParam)* ','?)? ')'
TypedParam =
ScalarType Name
// List with no delimeters
QubitList =
(GateOperand (',' GateOperand)* ','?)?
Param =
Name
//****************************//
// Statements and Expressions //
//****************************//
ExprStmt =
Expr ';'?
// We do not want to include all expressions here.
Expr =
ArrayExpr
| ArrayLiteral
| BinExpr
| BlockExpr
| BoxExpr
| CallExpr
| CastExpression
| GateCallExpr
| GPhaseCallExpr
| HardwareQubit
| Identifier
| IndexExpr
| IndexedIdentifier
| Literal
| TimingLiteral
| MeasureExpression
| ModifiedGateCallExpr
| ParenExpr
| PrefixExpr
| RangeExpr
| ReturnExpr
Identifier =
'ident'
Literal =
value:(
'int_number' | 'float_number'
| 'string'
| 'bit_string'
| 'true' | 'false'
)
TimingLiteral =
Literal Identifier
LetStmt =
'let' Name '=' Expr ';'
BlockExpr =
'{'
statements:Stmt*
'}'
PrefixExpr =
op:('~' | '!' | '-') Expr
// Adding a binary op here requires changes in several places. See notes at the top of this file.
BinExpr =
lhs:Expr
op:(
'||' | '&&'
| '==' | '!=' | '<=' | '>=' | '<' | '>'
| '+' | '*' | '-' | '/' | '%' | '<<' | '>>' | '^' | '|' | '&'
| '=' | '+=' | '/=' | '*=' | '%=' | '>>=' | '<<=' | '-=' | '|=' | '&=' | '^='
| '++'
)
rhs:Expr
// This is somehow used in parsing IfElse. Don't see how though.
ParenExpr =
'(' Expr ')'
// FIXME: replace part inside with ExpressionList
ArrayExpr =
'[' (
(Expr (',' Expr)* ','?)?
| Expr ';' Expr
) ']'
IndexExpr =
Expr IndexOperator
// Note: I don't know if the modifier '?' is used at all. Not in methods on `struct ArgList`.
// FIXME: use ExpressionList
ArgList =
'(' ExpressionList? ')'
// Gate arg list has different reqs from function call arg list
// But too hard to backtrack, so we don't make the distinction
// GateArgList =
// Expr (',' Expr)* ','?
CastExpression =
(ScalarType | ArrayType) '(' Expr ')'
GPhaseCallExpr =
'gphase' arg:Expr
// FIXME: Explain why head of call expression is Expr and not Name
CallExpr =
Expr ArgList
InvModifier =
'inv' '@'
PowModifier =
'pow' ParenExpr '@'
CtrlModifier =
'ctrl' ParenExpr? '@'
NegCtrlModifier =
'negctrl' ParenExpr? '@'
GateCallExpr =
Name ArgList? QubitList
Modifier =
InvModifier | PowModifier | CtrlModifier | NegCtrlModifier
ModifiedGateCallExpr =
(Modifier Modifier*) (GateCallExpr | GPhaseCallExpr)
// label 'then_branch' is handled manually in expr_ext.rs
IfStmt =
'if' '(' condition:Expr ')' then_branch:Expr
('else' else_branch:Expr)?
// label 'iterable' is handled manually in node_ext.rs
// ForStmt =
// 'for' loop_var:Name 'in' iterable:Expr
// loop_body:Expr
ForStmt =
'for' ScalarType loop_var:Name 'in' ForIterable
body:BlockExpr | stmt:Stmt
// Recall, the label prevents code generation from implementing this alternation
// as an `enum`. Leaving out the label would cause an error during codegen.
ForIterable =
SetExpression | RangeExpr | for_iterable_expr:Expr
WhileStmt =
'while' '(' condition:Expr ')'
loop_body:Expr
// For OQ3
// FIXME: Decide on how to organize range with and without square brackets. We
// need both. Now following OQ3 ANTLR grammar, includes brackets elsewhere.
// FIXME: 'start' is intercepted and destroyed somewhere. So I used thestart.
RangeExpr =
thestart:Expr ':' step:Expr? ':' stop:Expr
// '[' thestart:Expr ':' step:Expr? ':' stop:Expr ']'
// For OQ3
ReturnExpr =
'return' Expr?
BoxExpr =
'box' Expr
//*************************//
// Types //
//*************************//
// FIXME: move optional `const` here. This is currently
// in ClassicalDeclarationStatement.
ScalarType =
'bit' Designator?
| 'int' Designator?
| 'uint' Designator?
| 'float' Designator?
| 'angle' Designator?
| 'bool'
| 'duration'
| 'stretch'
| 'complex' ('[' ScalarType ']')?
Designator =
'[' Expr ']'
QubitType =
'qubit' Designator?
ArrayType =
'array' '[' ScalarType ',' ExpressionList ']'
ExpressionList =
Expr (',' Expr)* ','?
ReturnSignature =
'->' ScalarType
// Primitive declaration statements.
AliasDeclarationStatement =
'let' Name '=' Expr ';'
MeasureExpression =
'measure' GateOperand
// We keep Identifier and IndexedIdentifier separate
GateOperand =
Identifier | IndexedIdentifier | HardwareQubit
SetExpression =
'{' ExpressionList '}'
// But, probably better to make ExprRange an expression and do this:
IndexOperator =
'[' IndexKind ']'
// We do this just to get an enum from codegen.
// codgen is not smart enough (and lacks info) to handle '[' SetExpression | ExpressionList ']'
IndexKind =
SetExpression | ExpressionList
IndexedIdentifier =
Name IndexOperator*
ArrayLiteral =
'{' ExpressionList '}'
// // FIXME
HardwareQubit =
Name
ClassicalDeclarationStatement =
'const'? (ScalarType | ArrayType) Name ('=' Expr)? ';'
IODeclarationStatement =
('input' | 'output') (ScalarType | ArrayType) Name ';'
OldStyleDeclarationStatement =
('creg' | 'qreg') Name Designator? ';'
QuantumDeclarationStatement =
QubitType (Name | HardwareQubit) ';'
AssignmentStmt =
(Name | IndexedIdentifier) '=' rhs:Expr ';'