Commit | Line | Data |
---|---|---|
71aaa3f7 PP |
1 | # The MIT License (MIT) |
2 | # | |
3 | # Copyright (c) 2023 Philippe Proulx <eeppeliteloop@gmail.com> | |
4 | # | |
5 | # Permission is hereby granted, free of charge, to any person obtaining | |
6 | # a copy of this software and associated documentation files (the | |
7 | # "Software"), to deal in the Software without restriction, including | |
8 | # without limitation the rights to use, copy, modify, merge, publish, | |
9 | # distribute, sublicense, and/or sell copies of the Software, and to | |
10 | # permit persons to whom the Software is furnished to do so, subject to | |
11 | # the following conditions: | |
12 | # | |
13 | # The above copyright notice and this permission notice shall be | |
14 | # included in all copies or substantial portions of the Software. | |
15 | # | |
16 | # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, | |
17 | # EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF | |
18 | # MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. | |
19 | # IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY | |
20 | # CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, | |
21 | # TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE | |
22 | # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. | |
23 | ||
5a993698 PP |
24 | # This module is the portable Normand processor. It offers both the |
25 | # parse() function and the command-line tool (run the module itself) | |
26 | # without external dependencies except a `typing` module for Python 3.4. | |
27 | # | |
28 | # Feel free to copy this module file to your own project to use Normand. | |
29 | # | |
30 | # Upstream repository: <https://github.com/efficios/normand>. | |
31 | ||
71aaa3f7 | 32 | __author__ = "Philippe Proulx" |
ba11fb1d | 33 | __version__ = "0.20.0" |
71aaa3f7 | 34 | __all__ = [ |
320644e2 PP |
35 | "__author__", |
36 | "__version__", | |
71aaa3f7 | 37 | "ByteOrder", |
320644e2 | 38 | "LabelsT", |
71aaa3f7 PP |
39 | "parse", |
40 | "ParseError", | |
f5dcb24c | 41 | "ParseErrorMessage", |
71aaa3f7 | 42 | "ParseResult", |
e57a18e1 | 43 | "TextLocation", |
e57a18e1 | 44 | "VariablesT", |
71aaa3f7 PP |
45 | ] |
46 | ||
47 | import re | |
48 | import abc | |
49 | import ast | |
50 | import sys | |
320644e2 | 51 | import copy |
71aaa3f7 | 52 | import enum |
05f81895 | 53 | import math |
71aaa3f7 | 54 | import struct |
e57a18e1 PP |
55 | import typing |
56 | from typing import Any, Set, Dict, List, Union, Pattern, Callable, NoReturn, Optional | |
71aaa3f7 PP |
57 | |
58 | ||
59 | # Text location (line and column numbers). | |
e57a18e1 | 60 | class TextLocation: |
71aaa3f7 PP |
61 | @classmethod |
62 | def _create(cls, line_no: int, col_no: int): | |
63 | self = cls.__new__(cls) | |
64 | self._init(line_no, col_no) | |
65 | return self | |
66 | ||
67 | def __init__(*args, **kwargs): # type: ignore | |
68 | raise NotImplementedError | |
69 | ||
70 | def _init(self, line_no: int, col_no: int): | |
71 | self._line_no = line_no | |
72 | self._col_no = col_no | |
73 | ||
74 | # Line number. | |
75 | @property | |
76 | def line_no(self): | |
77 | return self._line_no | |
78 | ||
79 | # Column number. | |
80 | @property | |
81 | def col_no(self): | |
82 | return self._col_no | |
83 | ||
2adf4336 | 84 | def __repr__(self): |
e57a18e1 | 85 | return "TextLocation({}, {})".format(self._line_no, self._col_no) |
2adf4336 | 86 | |
71aaa3f7 PP |
87 | |
88 | # Any item. | |
89 | class _Item: | |
e57a18e1 | 90 | def __init__(self, text_loc: TextLocation): |
71aaa3f7 PP |
91 | self._text_loc = text_loc |
92 | ||
93 | # Source text location. | |
94 | @property | |
95 | def text_loc(self): | |
96 | return self._text_loc | |
97 | ||
2adf4336 PP |
98 | |
99 | # Scalar item. | |
100 | class _ScalarItem(_Item): | |
71aaa3f7 PP |
101 | # Returns the size, in bytes, of this item. |
102 | @property | |
103 | @abc.abstractmethod | |
104 | def size(self) -> int: | |
105 | ... | |
106 | ||
107 | ||
108 | # A repeatable item. | |
2adf4336 | 109 | class _RepableItem: |
71aaa3f7 PP |
110 | pass |
111 | ||
112 | ||
113 | # Single byte. | |
2adf4336 | 114 | class _Byte(_ScalarItem, _RepableItem): |
e57a18e1 | 115 | def __init__(self, val: int, text_loc: TextLocation): |
71aaa3f7 PP |
116 | super().__init__(text_loc) |
117 | self._val = val | |
118 | ||
119 | # Byte value. | |
120 | @property | |
121 | def val(self): | |
122 | return self._val | |
123 | ||
124 | @property | |
125 | def size(self): | |
126 | return 1 | |
127 | ||
128 | def __repr__(self): | |
676f6189 | 129 | return "_Byte({}, {})".format(hex(self._val), repr(self._text_loc)) |
71aaa3f7 PP |
130 | |
131 | ||
7a7b31e8 PP |
132 | # Literal string. |
133 | class _LitStr(_ScalarItem, _RepableItem): | |
e57a18e1 | 134 | def __init__(self, data: bytes, text_loc: TextLocation): |
71aaa3f7 PP |
135 | super().__init__(text_loc) |
136 | self._data = data | |
137 | ||
138 | # Encoded bytes. | |
139 | @property | |
140 | def data(self): | |
141 | return self._data | |
142 | ||
143 | @property | |
144 | def size(self): | |
145 | return len(self._data) | |
146 | ||
147 | def __repr__(self): | |
7a7b31e8 | 148 | return "_LitStr({}, {})".format(repr(self._data), repr(self._text_loc)) |
71aaa3f7 PP |
149 | |
150 | ||
151 | # Byte order. | |
152 | @enum.unique | |
153 | class ByteOrder(enum.Enum): | |
154 | # Big endian. | |
155 | BE = "be" | |
156 | ||
157 | # Little endian. | |
158 | LE = "le" | |
159 | ||
160 | ||
2adf4336 PP |
161 | # Byte order setting. |
162 | class _SetBo(_Item): | |
e57a18e1 | 163 | def __init__(self, bo: ByteOrder, text_loc: TextLocation): |
0e8e3169 | 164 | super().__init__(text_loc) |
71aaa3f7 PP |
165 | self._bo = bo |
166 | ||
167 | @property | |
168 | def bo(self): | |
169 | return self._bo | |
170 | ||
2adf4336 | 171 | def __repr__(self): |
676f6189 | 172 | return "_SetBo({}, {})".format(repr(self._bo), repr(self._text_loc)) |
71aaa3f7 PP |
173 | |
174 | ||
175 | # Label. | |
176 | class _Label(_Item): | |
e57a18e1 | 177 | def __init__(self, name: str, text_loc: TextLocation): |
71aaa3f7 PP |
178 | super().__init__(text_loc) |
179 | self._name = name | |
180 | ||
181 | # Label name. | |
182 | @property | |
183 | def name(self): | |
184 | return self._name | |
185 | ||
71aaa3f7 | 186 | def __repr__(self): |
676f6189 | 187 | return "_Label({}, {})".format(repr(self._name), repr(self._text_loc)) |
71aaa3f7 PP |
188 | |
189 | ||
2adf4336 PP |
190 | # Offset setting. |
191 | class _SetOffset(_Item): | |
e57a18e1 | 192 | def __init__(self, val: int, text_loc: TextLocation): |
71aaa3f7 PP |
193 | super().__init__(text_loc) |
194 | self._val = val | |
195 | ||
676f6189 | 196 | # Offset value (bytes). |
71aaa3f7 PP |
197 | @property |
198 | def val(self): | |
199 | return self._val | |
200 | ||
71aaa3f7 | 201 | def __repr__(self): |
676f6189 PP |
202 | return "_SetOffset({}, {})".format(repr(self._val), repr(self._text_loc)) |
203 | ||
204 | ||
205 | # Offset alignment. | |
206 | class _AlignOffset(_Item): | |
e57a18e1 | 207 | def __init__(self, val: int, pad_val: int, text_loc: TextLocation): |
676f6189 PP |
208 | super().__init__(text_loc) |
209 | self._val = val | |
210 | self._pad_val = pad_val | |
211 | ||
212 | # Alignment value (bits). | |
213 | @property | |
214 | def val(self): | |
215 | return self._val | |
216 | ||
217 | # Padding byte value. | |
218 | @property | |
219 | def pad_val(self): | |
220 | return self._pad_val | |
221 | ||
222 | def __repr__(self): | |
223 | return "_AlignOffset({}, {}, {})".format( | |
224 | repr(self._val), repr(self._pad_val), repr(self._text_loc) | |
225 | ) | |
71aaa3f7 PP |
226 | |
227 | ||
228 | # Mixin of containing an AST expression and its string. | |
229 | class _ExprMixin: | |
230 | def __init__(self, expr_str: str, expr: ast.Expression): | |
231 | self._expr_str = expr_str | |
232 | self._expr = expr | |
233 | ||
234 | # Expression string. | |
235 | @property | |
236 | def expr_str(self): | |
237 | return self._expr_str | |
238 | ||
239 | # Expression node to evaluate. | |
240 | @property | |
241 | def expr(self): | |
242 | return self._expr | |
243 | ||
244 | ||
25ca454b PP |
245 | # Fill until some offset. |
246 | class _FillUntil(_Item, _ExprMixin): | |
247 | def __init__( | |
248 | self, expr_str: str, expr: ast.Expression, pad_val: int, text_loc: TextLocation | |
249 | ): | |
250 | super().__init__(text_loc) | |
251 | _ExprMixin.__init__(self, expr_str, expr) | |
252 | self._pad_val = pad_val | |
253 | ||
254 | # Padding byte value. | |
255 | @property | |
256 | def pad_val(self): | |
257 | return self._pad_val | |
258 | ||
259 | def __repr__(self): | |
260 | return "_FillUntil({}, {}, {}, {})".format( | |
261 | repr(self._expr_str), | |
262 | repr(self._expr), | |
263 | repr(self._pad_val), | |
264 | repr(self._text_loc), | |
265 | ) | |
266 | ||
267 | ||
2adf4336 PP |
268 | # Variable assignment. |
269 | class _VarAssign(_Item, _ExprMixin): | |
71aaa3f7 | 270 | def __init__( |
e57a18e1 | 271 | self, name: str, expr_str: str, expr: ast.Expression, text_loc: TextLocation |
71aaa3f7 PP |
272 | ): |
273 | super().__init__(text_loc) | |
274 | _ExprMixin.__init__(self, expr_str, expr) | |
275 | self._name = name | |
276 | ||
277 | # Name. | |
278 | @property | |
279 | def name(self): | |
280 | return self._name | |
281 | ||
71aaa3f7 | 282 | def __repr__(self): |
2adf4336 | 283 | return "_VarAssign({}, {}, {}, {})".format( |
676f6189 PP |
284 | repr(self._name), |
285 | repr(self._expr_str), | |
286 | repr(self._expr), | |
287 | repr(self._text_loc), | |
71aaa3f7 PP |
288 | ) |
289 | ||
290 | ||
269f6eb3 PP |
291 | # Fixed-length number, possibly needing more than one byte. |
292 | class _FlNum(_ScalarItem, _RepableItem, _ExprMixin): | |
71aaa3f7 | 293 | def __init__( |
e57a18e1 | 294 | self, expr_str: str, expr: ast.Expression, len: int, text_loc: TextLocation |
71aaa3f7 PP |
295 | ): |
296 | super().__init__(text_loc) | |
297 | _ExprMixin.__init__(self, expr_str, expr) | |
298 | self._len = len | |
299 | ||
300 | # Length (bits). | |
301 | @property | |
302 | def len(self): | |
303 | return self._len | |
304 | ||
305 | @property | |
306 | def size(self): | |
307 | return self._len // 8 | |
308 | ||
309 | def __repr__(self): | |
269f6eb3 | 310 | return "_FlNum({}, {}, {}, {})".format( |
676f6189 PP |
311 | repr(self._expr_str), |
312 | repr(self._expr), | |
313 | repr(self._len), | |
314 | repr(self._text_loc), | |
71aaa3f7 PP |
315 | ) |
316 | ||
317 | ||
05f81895 PP |
318 | # LEB128 integer. |
319 | class _Leb128Int(_Item, _RepableItem, _ExprMixin): | |
e57a18e1 | 320 | def __init__(self, expr_str: str, expr: ast.Expression, text_loc: TextLocation): |
05f81895 PP |
321 | super().__init__(text_loc) |
322 | _ExprMixin.__init__(self, expr_str, expr) | |
323 | ||
324 | def __repr__(self): | |
325 | return "{}({}, {}, {})".format( | |
326 | self.__class__.__name__, | |
327 | repr(self._expr_str), | |
328 | repr(self._expr), | |
676f6189 | 329 | repr(self._text_loc), |
05f81895 PP |
330 | ) |
331 | ||
332 | ||
333 | # Unsigned LEB128 integer. | |
334 | class _ULeb128Int(_Leb128Int, _RepableItem, _ExprMixin): | |
335 | pass | |
336 | ||
337 | ||
338 | # Signed LEB128 integer. | |
339 | class _SLeb128Int(_Leb128Int, _RepableItem, _ExprMixin): | |
340 | pass | |
341 | ||
342 | ||
7a7b31e8 PP |
343 | # String. |
344 | class _Str(_Item, _RepableItem, _ExprMixin): | |
345 | def __init__( | |
346 | self, expr_str: str, expr: ast.Expression, codec: str, text_loc: TextLocation | |
347 | ): | |
348 | super().__init__(text_loc) | |
349 | _ExprMixin.__init__(self, expr_str, expr) | |
350 | self._codec = codec | |
351 | ||
352 | # Codec name. | |
353 | @property | |
354 | def codec(self): | |
355 | return self._codec | |
356 | ||
357 | def __repr__(self): | |
358 | return "_Str({}, {}, {}, {})".format( | |
359 | self.__class__.__name__, | |
360 | repr(self._expr_str), | |
361 | repr(self._expr), | |
362 | repr(self._codec), | |
363 | repr(self._text_loc), | |
364 | ) | |
365 | ||
366 | ||
71aaa3f7 | 367 | # Group of items. |
2adf4336 | 368 | class _Group(_Item, _RepableItem): |
e57a18e1 | 369 | def __init__(self, items: List[_Item], text_loc: TextLocation): |
71aaa3f7 PP |
370 | super().__init__(text_loc) |
371 | self._items = items | |
71aaa3f7 PP |
372 | |
373 | # Contained items. | |
374 | @property | |
375 | def items(self): | |
376 | return self._items | |
377 | ||
71aaa3f7 | 378 | def __repr__(self): |
676f6189 | 379 | return "_Group({}, {})".format(repr(self._items), repr(self._text_loc)) |
71aaa3f7 PP |
380 | |
381 | ||
382 | # Repetition item. | |
2adf4336 PP |
383 | class _Rep(_Item, _ExprMixin): |
384 | def __init__( | |
e57a18e1 | 385 | self, item: _Item, expr_str: str, expr: ast.Expression, text_loc: TextLocation |
2adf4336 | 386 | ): |
71aaa3f7 | 387 | super().__init__(text_loc) |
2adf4336 | 388 | _ExprMixin.__init__(self, expr_str, expr) |
71aaa3f7 | 389 | self._item = item |
71aaa3f7 PP |
390 | |
391 | # Item to repeat. | |
392 | @property | |
393 | def item(self): | |
394 | return self._item | |
395 | ||
71aaa3f7 | 396 | def __repr__(self): |
2adf4336 | 397 | return "_Rep({}, {}, {}, {})".format( |
676f6189 PP |
398 | repr(self._item), |
399 | repr(self._expr_str), | |
400 | repr(self._expr), | |
401 | repr(self._text_loc), | |
71aaa3f7 PP |
402 | ) |
403 | ||
404 | ||
27d52a19 PP |
405 | # Conditional item. |
406 | class _Cond(_Item, _ExprMixin): | |
407 | def __init__( | |
12b5dbc0 PP |
408 | self, |
409 | true_item: _Item, | |
410 | false_item: _Item, | |
411 | expr_str: str, | |
412 | expr: ast.Expression, | |
413 | text_loc: TextLocation, | |
27d52a19 PP |
414 | ): |
415 | super().__init__(text_loc) | |
416 | _ExprMixin.__init__(self, expr_str, expr) | |
12b5dbc0 PP |
417 | self._true_item = true_item |
418 | self._false_item = false_item | |
27d52a19 | 419 | |
12b5dbc0 | 420 | # Item when condition is true. |
27d52a19 | 421 | @property |
12b5dbc0 PP |
422 | def true_item(self): |
423 | return self._true_item | |
424 | ||
425 | # Item when condition is false. | |
426 | @property | |
427 | def false_item(self): | |
428 | return self._false_item | |
27d52a19 PP |
429 | |
430 | def __repr__(self): | |
12b5dbc0 PP |
431 | return "_Cond({}, {}, {}, {}, {})".format( |
432 | repr(self._true_item), | |
433 | repr(self._false_item), | |
27d52a19 PP |
434 | repr(self._expr_str), |
435 | repr(self._expr), | |
436 | repr(self._text_loc), | |
437 | ) | |
438 | ||
439 | ||
320644e2 PP |
440 | # Macro definition item. |
441 | class _MacroDef(_Item): | |
442 | def __init__( | |
443 | self, name: str, param_names: List[str], group: _Group, text_loc: TextLocation | |
444 | ): | |
445 | super().__init__(text_loc) | |
446 | self._name = name | |
447 | self._param_names = param_names | |
448 | self._group = group | |
449 | ||
450 | # Name. | |
451 | @property | |
452 | def name(self): | |
453 | return self._name | |
454 | ||
455 | # Parameters. | |
456 | @property | |
457 | def param_names(self): | |
458 | return self._param_names | |
459 | ||
460 | # Contained items. | |
461 | @property | |
462 | def group(self): | |
463 | return self._group | |
464 | ||
465 | def __repr__(self): | |
466 | return "_MacroDef({}, {}, {}, {})".format( | |
467 | repr(self._name), | |
468 | repr(self._param_names), | |
469 | repr(self._group), | |
470 | repr(self._text_loc), | |
471 | ) | |
472 | ||
473 | ||
474 | # Macro expansion parameter. | |
475 | class _MacroExpParam: | |
476 | def __init__(self, expr_str: str, expr: ast.Expression, text_loc: TextLocation): | |
477 | self._expr_str = expr_str | |
478 | self._expr = expr | |
479 | self._text_loc = text_loc | |
480 | ||
481 | # Expression string. | |
482 | @property | |
483 | def expr_str(self): | |
484 | return self._expr_str | |
485 | ||
486 | # Expression. | |
487 | @property | |
488 | def expr(self): | |
489 | return self._expr | |
490 | ||
491 | # Source text location. | |
492 | @property | |
493 | def text_loc(self): | |
494 | return self._text_loc | |
495 | ||
496 | def __repr__(self): | |
497 | return "_MacroExpParam({}, {}, {})".format( | |
498 | repr(self._expr_str), repr(self._expr), repr(self._text_loc) | |
499 | ) | |
500 | ||
501 | ||
502 | # Macro expansion item. | |
503 | class _MacroExp(_Item, _RepableItem): | |
504 | def __init__( | |
505 | self, | |
506 | name: str, | |
507 | params: List[_MacroExpParam], | |
508 | text_loc: TextLocation, | |
509 | ): | |
510 | super().__init__(text_loc) | |
511 | self._name = name | |
512 | self._params = params | |
513 | ||
514 | # Name. | |
515 | @property | |
516 | def name(self): | |
517 | return self._name | |
518 | ||
519 | # Parameters. | |
520 | @property | |
521 | def params(self): | |
522 | return self._params | |
523 | ||
524 | def __repr__(self): | |
525 | return "_MacroExp({}, {}, {})".format( | |
526 | repr(self._name), | |
527 | repr(self._params), | |
528 | repr(self._text_loc), | |
529 | ) | |
2adf4336 PP |
530 | |
531 | ||
f5dcb24c PP |
532 | # A parsing error message: a string and a text location. |
533 | class ParseErrorMessage: | |
534 | @classmethod | |
535 | def _create(cls, text: str, text_loc: TextLocation): | |
536 | self = cls.__new__(cls) | |
537 | self._init(text, text_loc) | |
538 | return self | |
539 | ||
540 | def __init__(self, *args, **kwargs): # type: ignore | |
541 | raise NotImplementedError | |
542 | ||
543 | def _init(self, text: str, text_loc: TextLocation): | |
544 | self._text = text | |
545 | self._text_loc = text_loc | |
546 | ||
547 | # Message text. | |
548 | @property | |
549 | def text(self): | |
550 | return self._text | |
551 | ||
552 | # Source text location. | |
553 | @property | |
554 | def text_location(self): | |
555 | return self._text_loc | |
556 | ||
557 | ||
558 | # A parsing error containing one or more messages (`ParseErrorMessage`). | |
71aaa3f7 PP |
559 | class ParseError(RuntimeError): |
560 | @classmethod | |
e57a18e1 | 561 | def _create(cls, msg: str, text_loc: TextLocation): |
71aaa3f7 PP |
562 | self = cls.__new__(cls) |
563 | self._init(msg, text_loc) | |
564 | return self | |
565 | ||
566 | def __init__(self, *args, **kwargs): # type: ignore | |
567 | raise NotImplementedError | |
568 | ||
e57a18e1 | 569 | def _init(self, msg: str, text_loc: TextLocation): |
71aaa3f7 | 570 | super().__init__(msg) |
f5dcb24c PP |
571 | self._msgs = [] # type: List[ParseErrorMessage] |
572 | self._add_msg(msg, text_loc) | |
71aaa3f7 | 573 | |
f5dcb24c PP |
574 | def _add_msg(self, msg: str, text_loc: TextLocation): |
575 | self._msgs.append( | |
576 | ParseErrorMessage._create( # pyright: ignore[reportPrivateUsage] | |
577 | msg, text_loc | |
578 | ) | |
579 | ) | |
580 | ||
581 | # Parsing error messages. | |
582 | # | |
583 | # The first message is the most specific one. | |
71aaa3f7 | 584 | @property |
f5dcb24c PP |
585 | def messages(self): |
586 | return self._msgs | |
71aaa3f7 PP |
587 | |
588 | ||
589 | # Raises a parsing error, forwarding the parameters to the constructor. | |
e57a18e1 | 590 | def _raise_error(msg: str, text_loc: TextLocation) -> NoReturn: |
71aaa3f7 PP |
591 | raise ParseError._create(msg, text_loc) # pyright: ignore[reportPrivateUsage] |
592 | ||
593 | ||
f5dcb24c PP |
594 | # Adds a message to the parsing error `exc`. |
595 | def _add_error_msg(exc: ParseError, msg: str, text_loc: TextLocation): | |
596 | exc._add_msg(msg, text_loc) # pyright: ignore[reportPrivateUsage] | |
597 | ||
598 | ||
599 | # Appends a message to the parsing error `exc` and reraises it. | |
600 | def _augment_error(exc: ParseError, msg: str, text_loc: TextLocation) -> NoReturn: | |
601 | _add_error_msg(exc, msg, text_loc) | |
602 | raise exc | |
603 | ||
604 | ||
b2410769 PP |
605 | # Returns a normalized version (so as to be parseable by int()) of |
606 | # the constant integer string `s`, possibly negative, dealing with | |
607 | # any radix suffix. | |
608 | def _norm_const_int(s: str): | |
609 | neg = "" | |
610 | pos = s | |
611 | ||
612 | if s.startswith("-"): | |
613 | neg = "-" | |
614 | pos = s[1:] | |
615 | ||
616 | for r in "xXoObB": | |
617 | if pos.startswith("0" + r): | |
618 | # Already correct | |
619 | return s | |
620 | ||
621 | # Try suffix | |
622 | asm_suf_base = { | |
623 | "h": "x", | |
624 | "H": "x", | |
625 | "q": "o", | |
626 | "Q": "o", | |
627 | "o": "o", | |
628 | "O": "o", | |
629 | "b": "b", | |
630 | "B": "B", | |
631 | } | |
632 | ||
633 | for suf in asm_suf_base: | |
634 | if pos[-1] == suf: | |
635 | s = "{}0{}{}".format(neg, asm_suf_base[suf], pos.rstrip(suf)) | |
636 | ||
637 | return s | |
638 | ||
639 | ||
7a7b31e8 PP |
640 | # Encodes the string `s` using the codec `codec`, raising `ParseError` |
641 | # with `text_loc` on encoding error. | |
642 | def _encode_str(s: str, codec: str, text_loc: TextLocation): | |
643 | try: | |
644 | return s.encode(codec) | |
645 | except UnicodeEncodeError: | |
646 | _raise_error( | |
647 | "Cannot encode `{}` with the `{}` encoding".format(s, codec), text_loc | |
648 | ) | |
649 | ||
650 | ||
e57a18e1 | 651 | # Variables dictionary type (for type hints). |
7a7b31e8 | 652 | VariablesT = Dict[str, Union[int, float, str]] |
e57a18e1 PP |
653 | |
654 | ||
655 | # Labels dictionary type (for type hints). | |
656 | LabelsT = Dict[str, int] | |
71aaa3f7 PP |
657 | |
658 | ||
b2410769 | 659 | # Common patterns. |
71aaa3f7 | 660 | _py_name_pat = re.compile(r"[a-zA-Z_][a-zA-Z0-9_]*") |
b2410769 PP |
661 | _pos_const_int_pat = re.compile( |
662 | r"(?:0[Xx][A-Fa-f0-9]+|0[Oo][0-7]+|0[Bb][01]+|[A-Fa-f0-9]+[hH]|[0-7]+[qQoO]|[01]+[bB]|\d+)\b" | |
663 | ) | |
664 | _const_int_pat = re.compile(r"(?P<neg>-)?(?:{})".format(_pos_const_int_pat.pattern)) | |
665 | _const_float_pat = re.compile( | |
666 | r"[-+]?(?:(?:\d*\.\d+)|(?:\d+\.))(?:[Ee][+-]?\d+)?(?=\W|)" | |
667 | ) | |
71aaa3f7 PP |
668 | |
669 | ||
320644e2 PP |
670 | # Macro definition dictionary. |
671 | _MacroDefsT = Dict[str, _MacroDef] | |
672 | ||
673 | ||
71aaa3f7 PP |
674 | # Normand parser. |
675 | # | |
676 | # The constructor accepts a Normand input. After building, use the `res` | |
677 | # property to get the resulting main group. | |
678 | class _Parser: | |
679 | # Builds a parser to parse the Normand input `normand`, parsing | |
680 | # immediately. | |
e57a18e1 | 681 | def __init__(self, normand: str, variables: VariablesT, labels: LabelsT): |
71aaa3f7 PP |
682 | self._normand = normand |
683 | self._at = 0 | |
684 | self._line_no = 1 | |
685 | self._col_no = 1 | |
686 | self._label_names = set(labels.keys()) | |
687 | self._var_names = set(variables.keys()) | |
320644e2 | 688 | self._macro_defs = {} # type: _MacroDefsT |
71aaa3f7 PP |
689 | self._parse() |
690 | ||
691 | # Result (main group). | |
692 | @property | |
693 | def res(self): | |
694 | return self._res | |
695 | ||
320644e2 PP |
696 | # Macro definitions. |
697 | @property | |
698 | def macro_defs(self): | |
699 | return self._macro_defs | |
700 | ||
71aaa3f7 PP |
701 | # Current text location. |
702 | @property | |
703 | def _text_loc(self): | |
e57a18e1 | 704 | return TextLocation._create( # pyright: ignore[reportPrivateUsage] |
71aaa3f7 PP |
705 | self._line_no, self._col_no |
706 | ) | |
707 | ||
708 | # Returns `True` if this parser is done parsing. | |
709 | def _is_done(self): | |
710 | return self._at == len(self._normand) | |
711 | ||
712 | # Returns `True` if this parser isn't done parsing. | |
713 | def _isnt_done(self): | |
714 | return not self._is_done() | |
715 | ||
716 | # Raises a parse error, creating it using the message `msg` and the | |
717 | # current text location. | |
718 | def _raise_error(self, msg: str) -> NoReturn: | |
719 | _raise_error(msg, self._text_loc) | |
720 | ||
721 | # Tries to make the pattern `pat` match the current substring, | |
722 | # returning the match object and updating `self._at`, | |
723 | # `self._line_no`, and `self._col_no` on success. | |
724 | def _try_parse_pat(self, pat: Pattern[str]): | |
725 | m = pat.match(self._normand, self._at) | |
726 | ||
727 | if m is None: | |
728 | return | |
729 | ||
730 | # Skip matched string | |
731 | self._at += len(m.group(0)) | |
732 | ||
733 | # Update line number | |
734 | self._line_no += m.group(0).count("\n") | |
735 | ||
736 | # Update column number | |
737 | for i in reversed(range(self._at)): | |
738 | if self._normand[i] == "\n" or i == 0: | |
739 | if i == 0: | |
740 | self._col_no = self._at + 1 | |
741 | else: | |
742 | self._col_no = self._at - i | |
743 | ||
744 | break | |
745 | ||
746 | # Return match object | |
747 | return m | |
748 | ||
749 | # Expects the pattern `pat` to match the current substring, | |
750 | # returning the match object and updating `self._at`, | |
751 | # `self._line_no`, and `self._col_no` on success, or raising a parse | |
752 | # error with the message `error_msg` on error. | |
753 | def _expect_pat(self, pat: Pattern[str], error_msg: str): | |
754 | # Match | |
755 | m = self._try_parse_pat(pat) | |
756 | ||
757 | if m is None: | |
758 | # No match: error | |
759 | self._raise_error(error_msg) | |
760 | ||
761 | # Return match object | |
762 | return m | |
763 | ||
ba11fb1d PP |
764 | # Patterns for _skip_*() |
765 | _comment_pat = re.compile(r"#[^#]*?(?:$|#)", re.M) | |
766 | _ws_or_comments_pat = re.compile(r"(?:\s|{})*".format(_comment_pat.pattern), re.M) | |
71aaa3f7 | 767 | _ws_or_syms_or_comments_pat = re.compile( |
ba11fb1d | 768 | r"(?:[\s/\\?&:;.,[\]_=|-]|{})*".format(_comment_pat.pattern), re.M |
71aaa3f7 PP |
769 | ) |
770 | ||
ba11fb1d PP |
771 | # Skips as many whitespaces and comments as possible, but not |
772 | # insignificant symbol characters. | |
773 | def _skip_ws_and_comments(self): | |
774 | self._try_parse_pat(self._ws_or_comments_pat) | |
775 | ||
71aaa3f7 PP |
776 | # Skips as many whitespaces, insignificant symbol characters, and |
777 | # comments as possible. | |
ba11fb1d | 778 | def _skip_ws_and_comments_and_syms(self): |
71aaa3f7 PP |
779 | self._try_parse_pat(self._ws_or_syms_or_comments_pat) |
780 | ||
781 | # Pattern for _try_parse_hex_byte() | |
782 | _nibble_pat = re.compile(r"[A-Fa-f0-9]") | |
783 | ||
784 | # Tries to parse a hexadecimal byte, returning a byte item on | |
785 | # success. | |
786 | def _try_parse_hex_byte(self): | |
0e8e3169 PP |
787 | begin_text_loc = self._text_loc |
788 | ||
71aaa3f7 PP |
789 | # Match initial nibble |
790 | m_high = self._try_parse_pat(self._nibble_pat) | |
791 | ||
792 | if m_high is None: | |
793 | # No match | |
794 | return | |
795 | ||
796 | # Expect another nibble | |
ba11fb1d | 797 | self._skip_ws_and_comments_and_syms() |
71aaa3f7 PP |
798 | m_low = self._expect_pat( |
799 | self._nibble_pat, "Expecting another hexadecimal nibble" | |
800 | ) | |
801 | ||
802 | # Return item | |
0e8e3169 | 803 | return _Byte(int(m_high.group(0) + m_low.group(0), 16), begin_text_loc) |
71aaa3f7 PP |
804 | |
805 | # Patterns for _try_parse_bin_byte() | |
806 | _bin_byte_bit_pat = re.compile(r"[01]") | |
6dd69a2a | 807 | _bin_byte_prefix_pat = re.compile(r"%+") |
71aaa3f7 PP |
808 | |
809 | # Tries to parse a binary byte, returning a byte item on success. | |
810 | def _try_parse_bin_byte(self): | |
0e8e3169 PP |
811 | begin_text_loc = self._text_loc |
812 | ||
71aaa3f7 | 813 | # Match prefix |
6dd69a2a PP |
814 | m = self._try_parse_pat(self._bin_byte_prefix_pat) |
815 | ||
816 | if m is None: | |
71aaa3f7 PP |
817 | # No match |
818 | return | |
819 | ||
6dd69a2a PP |
820 | # Expect as many bytes as there are `%` prefixes |
821 | items = [] # type: List[_Item] | |
71aaa3f7 | 822 | |
6dd69a2a | 823 | for _ in range(len(m.group(0))): |
ba11fb1d | 824 | self._skip_ws_and_comments_and_syms() |
6dd69a2a PP |
825 | byte_text_loc = self._text_loc |
826 | bits = [] # type: List[str] | |
827 | ||
828 | # Expect eight bits | |
829 | for _ in range(8): | |
ba11fb1d | 830 | self._skip_ws_and_comments_and_syms() |
6dd69a2a PP |
831 | m = self._expect_pat( |
832 | self._bin_byte_bit_pat, "Expecting a bit (`0` or `1`)" | |
833 | ) | |
834 | bits.append(m.group(0)) | |
835 | ||
836 | items.append(_Byte(int("".join(bits), 2), byte_text_loc)) | |
71aaa3f7 PP |
837 | |
838 | # Return item | |
6dd69a2a PP |
839 | if len(items) == 1: |
840 | return items[0] | |
841 | ||
842 | # As group | |
843 | return _Group(items, begin_text_loc) | |
71aaa3f7 PP |
844 | |
845 | # Patterns for _try_parse_dec_byte() | |
320644e2 | 846 | _dec_byte_prefix_pat = re.compile(r"\$") |
71aaa3f7 PP |
847 | _dec_byte_val_pat = re.compile(r"(?P<neg>-?)(?P<val>\d+)") |
848 | ||
849 | # Tries to parse a decimal byte, returning a byte item on success. | |
850 | def _try_parse_dec_byte(self): | |
0e8e3169 PP |
851 | begin_text_loc = self._text_loc |
852 | ||
71aaa3f7 PP |
853 | # Match prefix |
854 | if self._try_parse_pat(self._dec_byte_prefix_pat) is None: | |
855 | # No match | |
856 | return | |
857 | ||
858 | # Expect the value | |
ba11fb1d | 859 | self._skip_ws_and_comments() |
71aaa3f7 PP |
860 | m = self._expect_pat(self._dec_byte_val_pat, "Expecting a decimal constant") |
861 | ||
862 | # Compute value | |
863 | val = int(m.group("val")) * (-1 if m.group("neg") == "-" else 1) | |
864 | ||
865 | # Validate | |
866 | if val < -128 or val > 255: | |
0e8e3169 | 867 | _raise_error("Invalid decimal byte value {}".format(val), begin_text_loc) |
71aaa3f7 PP |
868 | |
869 | # Two's complement | |
05f81895 | 870 | val %= 256 |
71aaa3f7 PP |
871 | |
872 | # Return item | |
0e8e3169 | 873 | return _Byte(val, begin_text_loc) |
71aaa3f7 PP |
874 | |
875 | # Tries to parse a byte, returning a byte item on success. | |
876 | def _try_parse_byte(self): | |
877 | # Hexadecimal | |
878 | item = self._try_parse_hex_byte() | |
879 | ||
880 | if item is not None: | |
881 | return item | |
882 | ||
883 | # Binary | |
884 | item = self._try_parse_bin_byte() | |
885 | ||
886 | if item is not None: | |
887 | return item | |
888 | ||
889 | # Decimal | |
890 | item = self._try_parse_dec_byte() | |
891 | ||
892 | if item is not None: | |
893 | return item | |
894 | ||
71aaa3f7 | 895 | # Strings corresponding to escape sequence characters |
7a7b31e8 | 896 | _lit_str_escape_seq_strs = { |
71aaa3f7 PP |
897 | "0": "\0", |
898 | "a": "\a", | |
899 | "b": "\b", | |
900 | "e": "\x1b", | |
901 | "f": "\f", | |
902 | "n": "\n", | |
903 | "r": "\r", | |
904 | "t": "\t", | |
905 | "v": "\v", | |
906 | "\\": "\\", | |
907 | '"': '"', | |
908 | } | |
909 | ||
7a7b31e8 PP |
910 | # Patterns for _try_parse_lit_str() |
911 | _lit_str_prefix_suffix_pat = re.compile(r'"') | |
912 | _lit_str_contents_pat = re.compile(r'(?:(?:\\.)|[^"])*') | |
0e8e3169 | 913 | |
7a7b31e8 PP |
914 | # Parses a literal string between double quotes (without an encoding |
915 | # prefix) and returns the resulting string. | |
916 | def _try_parse_lit_str(self, with_prefix: bool): | |
917 | # Match prefix if needed | |
918 | if with_prefix: | |
919 | if self._try_parse_pat(self._lit_str_prefix_suffix_pat) is None: | |
920 | # No match | |
921 | return | |
71aaa3f7 | 922 | |
7a7b31e8 PP |
923 | # Expect literal string |
924 | m = self._expect_pat(self._lit_str_contents_pat, "Expecting a literal string") | |
925 | ||
926 | # Expect end of string | |
927 | self._expect_pat( | |
928 | self._lit_str_prefix_suffix_pat, 'Expecting `"` (end of literal string)' | |
929 | ) | |
930 | ||
931 | # Replace escape sequences | |
932 | val = m.group(0) | |
933 | ||
934 | for ec in '0abefnrtv"\\': | |
935 | val = val.replace(r"\{}".format(ec), self._lit_str_escape_seq_strs[ec]) | |
936 | ||
937 | # Return string | |
938 | return val | |
939 | ||
940 | # Patterns for _try_parse_utf_str_encoding() | |
941 | _str_encoding_utf_prefix_pat = re.compile(r"u") | |
942 | _str_encoding_utf_pat = re.compile(r"(?:8|(?:(?:16|32)(?:[bl]e)))\b") | |
943 | ||
944 | # Tries to parse a UTF encoding specification, returning the Python | |
945 | # codec name on success. | |
946 | def _try_parse_utf_str_encoding(self): | |
947 | # Match prefix | |
948 | if self._try_parse_pat(self._str_encoding_utf_prefix_pat) is None: | |
71aaa3f7 PP |
949 | # No match |
950 | return | |
951 | ||
7a7b31e8 PP |
952 | # Expect UTF specification |
953 | m = self._expect_pat( | |
954 | self._str_encoding_utf_pat, | |
955 | "Expecting `8`, `16be`, `16le`, `32be` or `32le`", | |
956 | ) | |
71aaa3f7 | 957 | |
7a7b31e8 PP |
958 | # Convert to codec name |
959 | return { | |
960 | "8": "utf_8", | |
961 | "16be": "utf_16_be", | |
962 | "16le": "utf_16_le", | |
963 | "32be": "utf_32_be", | |
964 | "32le": "utf_32_le", | |
965 | }[m.group(0)] | |
966 | ||
967 | # Patterns for _try_parse_str_encoding() | |
968 | _str_encoding_gen_prefix_pat = re.compile(r"s") | |
969 | _str_encoding_colon_pat = re.compile(r":") | |
970 | _str_encoding_non_utf_pat = re.compile(r"latin(?:[1-9]|10)\b") | |
971 | ||
972 | # Tries to parse a string encoding specification, returning the | |
973 | # Python codec name on success. | |
974 | # | |
975 | # Requires the general prefix (`s:`) if `req_gen_prefix` is `True`. | |
976 | def _try_parse_str_encoding(self, req_gen_prefix: bool = False): | |
977 | # General prefix? | |
978 | if self._try_parse_pat(self._str_encoding_gen_prefix_pat) is not None: | |
979 | # Expect `:` | |
ba11fb1d | 980 | self._skip_ws_and_comments() |
7a7b31e8 | 981 | self._expect_pat(self._str_encoding_colon_pat, "Expecting `:`") |
71aaa3f7 | 982 | |
7a7b31e8 | 983 | # Expect encoding specification |
ba11fb1d | 984 | self._skip_ws_and_comments() |
71aaa3f7 | 985 | |
7a7b31e8 PP |
986 | # UTF? |
987 | codec = self._try_parse_utf_str_encoding() | |
71aaa3f7 | 988 | |
7a7b31e8 PP |
989 | if codec is not None: |
990 | return codec | |
71aaa3f7 | 991 | |
7a7b31e8 PP |
992 | # Expect Latin |
993 | m = self._expect_pat( | |
994 | self._str_encoding_non_utf_pat, | |
995 | "Expecting `u8`, `u16be`, `u16le`, `u32be`, `u32le`, or `latin1` to `latin10`", | |
996 | ) | |
997 | return m.group(0) | |
71aaa3f7 | 998 | |
7a7b31e8 PP |
999 | # UTF? |
1000 | if not req_gen_prefix: | |
1001 | return self._try_parse_utf_str_encoding() | |
71aaa3f7 | 1002 | |
7a7b31e8 PP |
1003 | # Patterns for _try_parse_str() |
1004 | _lit_str_prefix_pat = re.compile(r'"') | |
1005 | _str_prefix_pat = re.compile(r'"|\{') | |
1006 | _str_expr_pat = re.compile(r"[^}]+") | |
1007 | _str_expr_suffix_pat = re.compile(r"\}") | |
1008 | ||
1009 | # Tries to parse a string, returning a literal string or string item | |
1010 | # on success. | |
1011 | def _try_parse_str(self): | |
1012 | begin_text_loc = self._text_loc | |
1013 | ||
1014 | # Encoding | |
1015 | codec = self._try_parse_str_encoding() | |
1016 | ||
1017 | # Match prefix (expect if there's an encoding specification) | |
ba11fb1d | 1018 | self._skip_ws_and_comments() |
7a7b31e8 PP |
1019 | |
1020 | if codec is None: | |
1021 | # No encoding: only a literal string (UTF-8) is legal | |
1022 | m_prefix = self._try_parse_pat(self._lit_str_prefix_pat) | |
1023 | ||
1024 | if m_prefix is None: | |
1025 | return | |
1026 | else: | |
1027 | # Encoding present: expect a string prefix | |
1028 | m_prefix = self._expect_pat(self._str_prefix_pat, 'Expecting `"` or `{`') | |
1029 | ||
1030 | # Literal string or expression? | |
1031 | prefix = m_prefix.group(0) | |
1032 | ||
1033 | if prefix == '"': | |
1034 | # Expect literal string | |
1035 | str_text_loc = self._text_loc | |
1036 | val = self._try_parse_lit_str(False) | |
1037 | ||
1038 | if val is None: | |
1039 | self._raise_error("Expecting a literal string") | |
1040 | ||
1041 | # Encode string | |
1042 | data = _encode_str(val, "utf_8" if codec is None else codec, str_text_loc) | |
1043 | ||
1044 | # Return item | |
1045 | return _LitStr(data, begin_text_loc) | |
1046 | else: | |
1047 | # Expect expression | |
ba11fb1d | 1048 | self._skip_ws_and_comments() |
7a7b31e8 PP |
1049 | expr_text_loc = self._text_loc |
1050 | m = self._expect_pat(self._str_expr_pat, "Expecting an expression") | |
1051 | ||
1052 | # Expect `}` | |
1053 | self._expect_pat(self._str_expr_suffix_pat, "Expecting `}`") | |
1054 | ||
1055 | # Create an expression node from the expression string | |
1056 | expr_str, expr = self._ast_expr_from_str(m.group(0), expr_text_loc) | |
1057 | ||
1058 | # Return item | |
1059 | assert codec is not None | |
1060 | return _Str(expr_str, expr, codec, begin_text_loc) | |
71aaa3f7 | 1061 | |
320644e2 PP |
1062 | # Common right parenthesis pattern |
1063 | _right_paren_pat = re.compile(r"\)") | |
1064 | ||
71aaa3f7 | 1065 | # Patterns for _try_parse_group() |
320644e2 | 1066 | _group_prefix_pat = re.compile(r"\(|!g(?:roup)?\b") |
71aaa3f7 PP |
1067 | |
1068 | # Tries to parse a group, returning a group item on success. | |
1069 | def _try_parse_group(self): | |
0e8e3169 PP |
1070 | begin_text_loc = self._text_loc |
1071 | ||
71aaa3f7 | 1072 | # Match prefix |
261c5ecf PP |
1073 | m_open = self._try_parse_pat(self._group_prefix_pat) |
1074 | ||
1075 | if m_open is None: | |
71aaa3f7 PP |
1076 | # No match |
1077 | return | |
1078 | ||
1079 | # Parse items | |
1080 | items = self._parse_items() | |
1081 | ||
1082 | # Expect end of group | |
ba11fb1d | 1083 | self._skip_ws_and_comments_and_syms() |
261c5ecf PP |
1084 | |
1085 | if m_open.group(0) == "(": | |
320644e2 | 1086 | pat = self._right_paren_pat |
261c5ecf PP |
1087 | exp = ")" |
1088 | else: | |
1089 | pat = self._block_end_pat | |
1090 | exp = "!end" | |
1091 | ||
1092 | self._expect_pat(pat, "Expecting an item or `{}` (end of group)".format(exp)) | |
71aaa3f7 PP |
1093 | |
1094 | # Return item | |
0e8e3169 | 1095 | return _Group(items, begin_text_loc) |
71aaa3f7 PP |
1096 | |
1097 | # Returns a stripped expression string and an AST expression node | |
1098 | # from the expression string `expr_str` at text location `text_loc`. | |
e57a18e1 | 1099 | def _ast_expr_from_str(self, expr_str: str, text_loc: TextLocation): |
71aaa3f7 PP |
1100 | # Create an expression node from the expression string |
1101 | expr_str = expr_str.strip().replace("\n", " ") | |
1102 | ||
1103 | try: | |
1104 | expr = ast.parse(expr_str, mode="eval") | |
1105 | except SyntaxError: | |
1106 | _raise_error( | |
1107 | "Invalid expression `{}`: invalid syntax".format(expr_str), | |
1108 | text_loc, | |
1109 | ) | |
1110 | ||
1111 | return expr_str, expr | |
1112 | ||
7a7b31e8 | 1113 | # Patterns for _try_parse_val() |
05f81895 | 1114 | _val_expr_pat = re.compile(r"([^}:]+):\s*") |
7a7b31e8 PP |
1115 | _fl_num_len_fmt_pat = re.compile(r"8|16|24|32|40|48|56|64") |
1116 | _leb128_int_fmt_pat = re.compile(r"(u|s)leb128") | |
71aaa3f7 | 1117 | |
7a7b31e8 PP |
1118 | # Tries to parse a value (number or string) and format (fixed length |
1119 | # in bits, `uleb128`, `sleb128`, or `s:` followed with an encoding | |
1120 | # name), returning an item on success. | |
1121 | def _try_parse_val(self): | |
71aaa3f7 PP |
1122 | begin_text_loc = self._text_loc |
1123 | ||
1124 | # Match | |
1125 | m_expr = self._try_parse_pat(self._val_expr_pat) | |
1126 | ||
1127 | if m_expr is None: | |
1128 | # No match | |
1129 | return | |
1130 | ||
71aaa3f7 PP |
1131 | # Create an expression node from the expression string |
1132 | expr_str, expr = self._ast_expr_from_str(m_expr.group(1), begin_text_loc) | |
1133 | ||
7a7b31e8 | 1134 | # Fixed length? |
ba11fb1d | 1135 | self._skip_ws_and_comments() |
7a7b31e8 | 1136 | m_fmt = self._try_parse_pat(self._fl_num_len_fmt_pat) |
05f81895 | 1137 | |
7a7b31e8 | 1138 | if m_fmt is None: |
05f81895 | 1139 | # LEB128? |
7a7b31e8 | 1140 | m_fmt = self._try_parse_pat(self._leb128_int_fmt_pat) |
05f81895 | 1141 | |
7a7b31e8 PP |
1142 | if m_fmt is None: |
1143 | # String encoding? | |
1144 | codec = self._try_parse_str_encoding(True) | |
1145 | ||
1146 | if codec is None: | |
1147 | # At this point it's invalid | |
1148 | self._raise_error( | |
1149 | "Expecting a fixed length (multiple of eight bits), `uleb128`, `sleb128`, or `s:` followed with a valid encoding (`u8`, `u16be`, `u16le`, `u32be`, `u32le`, or `latin1` to `latin10`)" | |
1150 | ) | |
1151 | else: | |
1152 | # Return string item | |
1153 | return _Str(expr_str, expr, codec, begin_text_loc) | |
05f81895 PP |
1154 | |
1155 | # Return LEB128 integer item | |
7a7b31e8 | 1156 | cls = _ULeb128Int if m_fmt.group(1) == "u" else _SLeb128Int |
05f81895 PP |
1157 | return cls(expr_str, expr, begin_text_loc) |
1158 | else: | |
269f6eb3 PP |
1159 | # Return fixed-length number item |
1160 | return _FlNum( | |
05f81895 PP |
1161 | expr_str, |
1162 | expr, | |
7a7b31e8 | 1163 | int(m_fmt.group(0)), |
05f81895 PP |
1164 | begin_text_loc, |
1165 | ) | |
71aaa3f7 | 1166 | |
320644e2 | 1167 | # Patterns for _try_parse_var_assign() |
7a7b31e8 PP |
1168 | _var_assign_name_equal_pat = re.compile( |
1169 | r"({})\s*=(?!=)".format(_py_name_pat.pattern) | |
1170 | ) | |
320644e2 | 1171 | _var_assign_expr_pat = re.compile(r"[^}]+") |
71aaa3f7 | 1172 | |
2adf4336 PP |
1173 | # Tries to parse a variable assignment, returning a variable |
1174 | # assignment item on success. | |
1175 | def _try_parse_var_assign(self): | |
71aaa3f7 PP |
1176 | begin_text_loc = self._text_loc |
1177 | ||
1178 | # Match | |
320644e2 | 1179 | m = self._try_parse_pat(self._var_assign_name_equal_pat) |
71aaa3f7 PP |
1180 | |
1181 | if m is None: | |
1182 | # No match | |
1183 | return | |
1184 | ||
1185 | # Validate name | |
320644e2 | 1186 | name = m.group(1) |
71aaa3f7 PP |
1187 | |
1188 | if name == _icitte_name: | |
0e8e3169 PP |
1189 | _raise_error( |
1190 | "`{}` is a reserved variable name".format(_icitte_name), begin_text_loc | |
1191 | ) | |
71aaa3f7 PP |
1192 | |
1193 | if name in self._label_names: | |
0e8e3169 | 1194 | _raise_error("Existing label named `{}`".format(name), begin_text_loc) |
71aaa3f7 | 1195 | |
320644e2 | 1196 | # Expect an expression |
ba11fb1d | 1197 | self._skip_ws_and_comments() |
320644e2 | 1198 | m = self._expect_pat(self._var_assign_expr_pat, "Expecting an expression") |
71aaa3f7 PP |
1199 | |
1200 | # Create an expression node from the expression string | |
320644e2 PP |
1201 | expr_str, expr = self._ast_expr_from_str(m.group(0), begin_text_loc) |
1202 | ||
1203 | # Add to known variable names | |
1204 | self._var_names.add(name) | |
71aaa3f7 PP |
1205 | |
1206 | # Return item | |
2adf4336 | 1207 | return _VarAssign( |
71aaa3f7 PP |
1208 | name, |
1209 | expr_str, | |
1210 | expr, | |
0e8e3169 | 1211 | begin_text_loc, |
71aaa3f7 PP |
1212 | ) |
1213 | ||
2adf4336 | 1214 | # Pattern for _try_parse_set_bo() |
71aaa3f7 PP |
1215 | _bo_pat = re.compile(r"[bl]e") |
1216 | ||
2adf4336 PP |
1217 | # Tries to parse a byte order name, returning a byte order setting |
1218 | # item on success. | |
1219 | def _try_parse_set_bo(self): | |
0e8e3169 PP |
1220 | begin_text_loc = self._text_loc |
1221 | ||
71aaa3f7 PP |
1222 | # Match |
1223 | m = self._try_parse_pat(self._bo_pat) | |
1224 | ||
1225 | if m is None: | |
1226 | # No match | |
1227 | return | |
1228 | ||
1229 | # Return corresponding item | |
1230 | if m.group(0) == "be": | |
2adf4336 | 1231 | return _SetBo(ByteOrder.BE, begin_text_loc) |
71aaa3f7 PP |
1232 | else: |
1233 | assert m.group(0) == "le" | |
2adf4336 | 1234 | return _SetBo(ByteOrder.LE, begin_text_loc) |
71aaa3f7 PP |
1235 | |
1236 | # Patterns for _try_parse_val_or_bo() | |
320644e2 PP |
1237 | _val_var_assign_set_bo_prefix_pat = re.compile(r"\{") |
1238 | _val_var_assign_set_bo_suffix_pat = re.compile(r"\}") | |
71aaa3f7 | 1239 | |
2adf4336 PP |
1240 | # Tries to parse a value, a variable assignment, or a byte order |
1241 | # setting, returning an item on success. | |
1242 | def _try_parse_val_or_var_assign_or_set_bo(self): | |
71aaa3f7 | 1243 | # Match prefix |
2adf4336 | 1244 | if self._try_parse_pat(self._val_var_assign_set_bo_prefix_pat) is None: |
71aaa3f7 PP |
1245 | # No match |
1246 | return | |
1247 | ||
ba11fb1d | 1248 | self._skip_ws_and_comments() |
320644e2 | 1249 | |
2adf4336 PP |
1250 | # Variable assignment item? |
1251 | item = self._try_parse_var_assign() | |
71aaa3f7 PP |
1252 | |
1253 | if item is None: | |
7a7b31e8 PP |
1254 | # Value item? |
1255 | item = self._try_parse_val() | |
71aaa3f7 PP |
1256 | |
1257 | if item is None: | |
2adf4336 PP |
1258 | # Byte order setting item? |
1259 | item = self._try_parse_set_bo() | |
71aaa3f7 PP |
1260 | |
1261 | if item is None: | |
1262 | # At this point it's invalid | |
2adf4336 | 1263 | self._raise_error( |
7a7b31e8 | 1264 | "Expecting a fixed-length number, a string, a variable assignment, or a byte order setting" |
2adf4336 | 1265 | ) |
71aaa3f7 PP |
1266 | |
1267 | # Expect suffix | |
ba11fb1d | 1268 | self._skip_ws_and_comments() |
2adf4336 | 1269 | self._expect_pat(self._val_var_assign_set_bo_suffix_pat, "Expecting `}`") |
71aaa3f7 PP |
1270 | return item |
1271 | ||
2adf4336 PP |
1272 | # Tries to parse an offset setting value (after the initial `<`), |
1273 | # returning an offset item on success. | |
1274 | def _try_parse_set_offset_val(self): | |
0e8e3169 PP |
1275 | begin_text_loc = self._text_loc |
1276 | ||
71aaa3f7 | 1277 | # Match |
b2410769 | 1278 | m = self._try_parse_pat(_pos_const_int_pat) |
71aaa3f7 PP |
1279 | |
1280 | if m is None: | |
1281 | # No match | |
1282 | return | |
1283 | ||
1284 | # Return item | |
b2410769 | 1285 | return _SetOffset(int(_norm_const_int(m.group(0)), 0), begin_text_loc) |
71aaa3f7 PP |
1286 | |
1287 | # Tries to parse a label name (after the initial `<`), returning a | |
1288 | # label item on success. | |
1289 | def _try_parse_label_name(self): | |
0e8e3169 PP |
1290 | begin_text_loc = self._text_loc |
1291 | ||
71aaa3f7 PP |
1292 | # Match |
1293 | m = self._try_parse_pat(_py_name_pat) | |
1294 | ||
1295 | if m is None: | |
1296 | # No match | |
1297 | return | |
1298 | ||
1299 | # Validate | |
1300 | name = m.group(0) | |
1301 | ||
1302 | if name == _icitte_name: | |
0e8e3169 PP |
1303 | _raise_error( |
1304 | "`{}` is a reserved label name".format(_icitte_name), begin_text_loc | |
1305 | ) | |
71aaa3f7 PP |
1306 | |
1307 | if name in self._label_names: | |
0e8e3169 | 1308 | _raise_error("Duplicate label name `{}`".format(name), begin_text_loc) |
71aaa3f7 PP |
1309 | |
1310 | if name in self._var_names: | |
0e8e3169 | 1311 | _raise_error("Existing variable named `{}`".format(name), begin_text_loc) |
71aaa3f7 PP |
1312 | |
1313 | # Add to known label names | |
1314 | self._label_names.add(name) | |
1315 | ||
1316 | # Return item | |
0e8e3169 | 1317 | return _Label(name, begin_text_loc) |
71aaa3f7 | 1318 | |
2adf4336 | 1319 | # Patterns for _try_parse_label_or_set_offset() |
320644e2 PP |
1320 | _label_set_offset_prefix_pat = re.compile(r"<") |
1321 | _label_set_offset_suffix_pat = re.compile(r">") | |
71aaa3f7 | 1322 | |
2adf4336 PP |
1323 | # Tries to parse a label or an offset setting, returning an item on |
1324 | # success. | |
1325 | def _try_parse_label_or_set_offset(self): | |
71aaa3f7 | 1326 | # Match prefix |
2adf4336 | 1327 | if self._try_parse_pat(self._label_set_offset_prefix_pat) is None: |
71aaa3f7 PP |
1328 | # No match |
1329 | return | |
1330 | ||
2adf4336 | 1331 | # Offset setting item? |
ba11fb1d | 1332 | self._skip_ws_and_comments() |
2adf4336 | 1333 | item = self._try_parse_set_offset_val() |
71aaa3f7 PP |
1334 | |
1335 | if item is None: | |
1336 | # Label item? | |
1337 | item = self._try_parse_label_name() | |
1338 | ||
1339 | if item is None: | |
1340 | # At this point it's invalid | |
2adf4336 | 1341 | self._raise_error("Expecting a label name or an offset setting value") |
71aaa3f7 PP |
1342 | |
1343 | # Expect suffix | |
ba11fb1d | 1344 | self._skip_ws_and_comments() |
2adf4336 | 1345 | self._expect_pat(self._label_set_offset_suffix_pat, "Expecting `>`") |
71aaa3f7 PP |
1346 | return item |
1347 | ||
25ca454b PP |
1348 | # Pattern for _parse_pad_val() |
1349 | _pad_val_prefix_pat = re.compile(r"~") | |
1350 | ||
1351 | # Tries to parse a padding value, returning the padding value, or 0 | |
1352 | # if none. | |
1353 | def _parse_pad_val(self): | |
1354 | # Padding value? | |
ba11fb1d | 1355 | self._skip_ws_and_comments() |
25ca454b PP |
1356 | pad_val = 0 |
1357 | ||
1358 | if self._try_parse_pat(self._pad_val_prefix_pat) is not None: | |
ba11fb1d | 1359 | self._skip_ws_and_comments() |
25ca454b PP |
1360 | pad_val_text_loc = self._text_loc |
1361 | m = self._expect_pat( | |
b2410769 | 1362 | _pos_const_int_pat, |
25ca454b PP |
1363 | "Expecting a positive constant integer (byte value)", |
1364 | ) | |
1365 | ||
1366 | # Validate | |
b2410769 | 1367 | pad_val = int(_norm_const_int(m.group(0)), 0) |
25ca454b PP |
1368 | |
1369 | if pad_val > 255: | |
1370 | _raise_error( | |
1371 | "Invalid padding byte value {}".format(pad_val), | |
1372 | pad_val_text_loc, | |
1373 | ) | |
1374 | ||
1375 | return pad_val | |
1376 | ||
676f6189 | 1377 | # Patterns for _try_parse_align_offset() |
320644e2 PP |
1378 | _align_offset_prefix_pat = re.compile(r"@") |
1379 | _align_offset_val_pat = re.compile(r"\d+") | |
676f6189 PP |
1380 | |
1381 | # Tries to parse an offset alignment, returning an offset alignment | |
1382 | # item on success. | |
1383 | def _try_parse_align_offset(self): | |
1384 | begin_text_loc = self._text_loc | |
1385 | ||
1386 | # Match prefix | |
1387 | if self._try_parse_pat(self._align_offset_prefix_pat) is None: | |
1388 | # No match | |
1389 | return | |
1390 | ||
320644e2 | 1391 | # Expect an alignment |
ba11fb1d | 1392 | self._skip_ws_and_comments() |
676f6189 PP |
1393 | align_text_loc = self._text_loc |
1394 | m = self._expect_pat( | |
1395 | self._align_offset_val_pat, | |
1396 | "Expecting an alignment (positive multiple of eight bits)", | |
1397 | ) | |
1398 | ||
1399 | # Validate alignment | |
320644e2 | 1400 | val = int(m.group(0)) |
676f6189 PP |
1401 | |
1402 | if val <= 0 or (val % 8) != 0: | |
1403 | _raise_error( | |
1404 | "Invalid alignment value {} (not a positive multiple of eight)".format( | |
1405 | val | |
1406 | ), | |
1407 | align_text_loc, | |
1408 | ) | |
1409 | ||
25ca454b PP |
1410 | # Padding value |
1411 | pad_val = self._parse_pad_val() | |
676f6189 | 1412 | |
25ca454b PP |
1413 | # Return item |
1414 | return _AlignOffset(val, pad_val, begin_text_loc) | |
676f6189 | 1415 | |
dbd84e74 PP |
1416 | # Patterns for _expect_expr() |
1417 | _inner_expr_prefix_pat = re.compile(r"\{") | |
1418 | _inner_expr_pat = re.compile(r"[^}]+") | |
1419 | _inner_expr_suffix_pat = re.compile(r"\}") | |
dbd84e74 PP |
1420 | |
1421 | # Parses an expression outside a `{`/`}` context. | |
1422 | # | |
1423 | # This function accepts: | |
1424 | # | |
1425 | # • A Python expression within `{` and `}`. | |
1426 | # | |
1427 | # • A Python name. | |
1428 | # | |
1429 | # • If `accept_const_int` is `True`: a constant integer, which may | |
1430 | # be negative if `allow_neg_int` is `True`. | |
1431 | # | |
1432 | # • If `accept_float` is `True`: a constant floating point number. | |
1433 | # | |
1434 | # Returns the stripped expression string and AST expression. | |
1435 | def _expect_expr( | |
1436 | self, | |
1437 | accept_const_int: bool = False, | |
1438 | allow_neg_int: bool = False, | |
1439 | accept_const_float: bool = False, | |
7a7b31e8 | 1440 | accept_lit_str: bool = False, |
dbd84e74 PP |
1441 | ): |
1442 | begin_text_loc = self._text_loc | |
1443 | ||
1444 | # Constant floating point number? | |
dbd84e74 | 1445 | if accept_const_float: |
b2410769 | 1446 | m = self._try_parse_pat(_const_float_pat) |
dbd84e74 PP |
1447 | |
1448 | if m is not None: | |
1449 | return self._ast_expr_from_str(m.group(0), begin_text_loc) | |
1450 | ||
1451 | # Constant integer? | |
dbd84e74 | 1452 | if accept_const_int: |
b2410769 | 1453 | m = self._try_parse_pat(_const_int_pat) |
dbd84e74 PP |
1454 | |
1455 | if m is not None: | |
1456 | # Negative and allowed? | |
1457 | if m.group("neg") == "-" and not allow_neg_int: | |
1458 | _raise_error( | |
1459 | "Expecting a positive constant integer", begin_text_loc | |
1460 | ) | |
1461 | ||
b2410769 | 1462 | expr_str = _norm_const_int(m.group(0)) |
dbd84e74 PP |
1463 | return self._ast_expr_from_str(expr_str, begin_text_loc) |
1464 | ||
1465 | # Name? | |
1466 | m = self._try_parse_pat(_py_name_pat) | |
1467 | ||
1468 | if m is not None: | |
1469 | return self._ast_expr_from_str(m.group(0), begin_text_loc) | |
1470 | ||
7a7b31e8 PP |
1471 | # Literal string |
1472 | if accept_lit_str: | |
1473 | val = self._try_parse_lit_str(True) | |
1474 | ||
1475 | if val is not None: | |
1476 | return self._ast_expr_from_str(repr(val), begin_text_loc) | |
1477 | ||
dbd84e74 PP |
1478 | # Expect `{` |
1479 | msg_accepted_parts = ["a name", "or `{`"] | |
1480 | ||
7a7b31e8 PP |
1481 | if accept_lit_str: |
1482 | msg_accepted_parts.insert(0, "a literal string") | |
1483 | ||
dbd84e74 PP |
1484 | if accept_const_float: |
1485 | msg_accepted_parts.insert(0, "a constant floating point number") | |
1486 | ||
1487 | if accept_const_int: | |
1488 | msg_pos = "" if allow_neg_int else "positive " | |
1489 | msg_accepted_parts.insert(0, "a {}constant integer".format(msg_pos)) | |
1490 | ||
1491 | if len(msg_accepted_parts) == 2: | |
1492 | msg_accepted = " ".join(msg_accepted_parts) | |
1493 | else: | |
1494 | msg_accepted = ", ".join(msg_accepted_parts) | |
1495 | ||
1496 | self._expect_pat( | |
1497 | self._inner_expr_prefix_pat, | |
1498 | "Expecting {}".format(msg_accepted), | |
1499 | ) | |
1500 | ||
1501 | # Expect an expression | |
ba11fb1d | 1502 | self._skip_ws_and_comments() |
dbd84e74 PP |
1503 | expr_text_loc = self._text_loc |
1504 | m = self._expect_pat(self._inner_expr_pat, "Expecting an expression") | |
1505 | expr_str = m.group(0) | |
1506 | ||
1507 | # Expect `}` | |
ba11fb1d | 1508 | self._skip_ws_and_comments() |
dbd84e74 PP |
1509 | self._expect_pat(self._inner_expr_suffix_pat, "Expecting `}`") |
1510 | ||
1511 | return self._ast_expr_from_str(expr_str, expr_text_loc) | |
1512 | ||
25ca454b PP |
1513 | # Patterns for _try_parse_fill_until() |
1514 | _fill_until_prefix_pat = re.compile(r"\+") | |
1515 | _fill_until_pad_val_prefix_pat = re.compile(r"~") | |
676f6189 | 1516 | |
25ca454b PP |
1517 | # Tries to parse a filling, returning a filling item on success. |
1518 | def _try_parse_fill_until(self): | |
1519 | begin_text_loc = self._text_loc | |
1520 | ||
1521 | # Match prefix | |
1522 | if self._try_parse_pat(self._fill_until_prefix_pat) is None: | |
1523 | # No match | |
1524 | return | |
1525 | ||
1526 | # Expect expression | |
ba11fb1d | 1527 | self._skip_ws_and_comments() |
dbd84e74 | 1528 | expr_str, expr = self._expect_expr(accept_const_int=True) |
25ca454b PP |
1529 | |
1530 | # Padding value | |
1531 | pad_val = self._parse_pad_val() | |
676f6189 PP |
1532 | |
1533 | # Return item | |
25ca454b | 1534 | return _FillUntil(expr_str, expr, pad_val, begin_text_loc) |
676f6189 | 1535 | |
27d52a19 PP |
1536 | # Parses the multiplier expression of a repetition (block or |
1537 | # post-item) and returns the expression string and AST node. | |
1538 | def _expect_rep_mul_expr(self): | |
dbd84e74 | 1539 | return self._expect_expr(accept_const_int=True) |
27d52a19 PP |
1540 | |
1541 | # Common block end pattern | |
320644e2 | 1542 | _block_end_pat = re.compile(r"!end\b") |
27d52a19 | 1543 | |
e57a18e1 | 1544 | # Pattern for _try_parse_rep_block() |
320644e2 | 1545 | _rep_block_prefix_pat = re.compile(r"!r(?:epeat)?\b") |
e57a18e1 PP |
1546 | |
1547 | # Tries to parse a repetition block, returning a repetition item on | |
1548 | # success. | |
1549 | def _try_parse_rep_block(self): | |
1550 | begin_text_loc = self._text_loc | |
1551 | ||
1552 | # Match prefix | |
1553 | if self._try_parse_pat(self._rep_block_prefix_pat) is None: | |
1554 | # No match | |
1555 | return | |
1556 | ||
1557 | # Expect expression | |
1558 | self._skip_ws_and_comments() | |
1559 | expr_str, expr = self._expect_rep_mul_expr() | |
1560 | ||
1561 | # Parse items | |
ba11fb1d | 1562 | self._skip_ws_and_comments_and_syms() |
e57a18e1 PP |
1563 | items_text_loc = self._text_loc |
1564 | items = self._parse_items() | |
1565 | ||
1566 | # Expect end of block | |
ba11fb1d | 1567 | self._skip_ws_and_comments_and_syms() |
e57a18e1 | 1568 | self._expect_pat( |
27d52a19 | 1569 | self._block_end_pat, "Expecting an item or `!end` (end of repetition block)" |
e57a18e1 PP |
1570 | ) |
1571 | ||
1572 | # Return item | |
1573 | return _Rep(_Group(items, items_text_loc), expr_str, expr, begin_text_loc) | |
1574 | ||
27d52a19 | 1575 | # Pattern for _try_parse_cond_block() |
320644e2 | 1576 | _cond_block_prefix_pat = re.compile(r"!if\b") |
12b5dbc0 | 1577 | _cond_block_else_pat = re.compile(r"!else\b") |
27d52a19 PP |
1578 | |
1579 | # Tries to parse a conditional block, returning a conditional item | |
1580 | # on success. | |
1581 | def _try_parse_cond_block(self): | |
1582 | begin_text_loc = self._text_loc | |
1583 | ||
1584 | # Match prefix | |
1585 | if self._try_parse_pat(self._cond_block_prefix_pat) is None: | |
1586 | # No match | |
1587 | return | |
1588 | ||
1589 | # Expect expression | |
1590 | self._skip_ws_and_comments() | |
dbd84e74 | 1591 | expr_str, expr = self._expect_expr() |
27d52a19 | 1592 | |
12b5dbc0 | 1593 | # Parse "true" items |
ba11fb1d | 1594 | self._skip_ws_and_comments_and_syms() |
12b5dbc0 PP |
1595 | true_items_text_loc = self._text_loc |
1596 | true_items = self._parse_items() | |
1597 | false_items = [] # type: List[_Item] | |
1598 | false_items_text_loc = begin_text_loc | |
27d52a19 | 1599 | |
12b5dbc0 | 1600 | # `!else`? |
ba11fb1d | 1601 | self._skip_ws_and_comments_and_syms() |
12b5dbc0 PP |
1602 | |
1603 | if self._try_parse_pat(self._cond_block_else_pat) is not None: | |
1604 | # Parse "false" items | |
ba11fb1d | 1605 | self._skip_ws_and_comments_and_syms() |
12b5dbc0 PP |
1606 | false_items_text_loc = self._text_loc |
1607 | false_items = self._parse_items() | |
1608 | ||
1609 | # Expect end of block | |
27d52a19 PP |
1610 | self._expect_pat( |
1611 | self._block_end_pat, | |
12b5dbc0 | 1612 | "Expecting an item, `!else`, or `!end` (end of conditional block)", |
27d52a19 PP |
1613 | ) |
1614 | ||
1615 | # Return item | |
12b5dbc0 PP |
1616 | return _Cond( |
1617 | _Group(true_items, true_items_text_loc), | |
1618 | _Group(false_items, false_items_text_loc), | |
1619 | expr_str, | |
1620 | expr, | |
1621 | begin_text_loc, | |
1622 | ) | |
27d52a19 | 1623 | |
320644e2 PP |
1624 | # Common left parenthesis pattern |
1625 | _left_paren_pat = re.compile(r"\(") | |
1626 | ||
1627 | # Patterns for _try_parse_macro_def() and _try_parse_macro_exp() | |
1628 | _macro_params_comma_pat = re.compile(",") | |
1629 | ||
1630 | # Patterns for _try_parse_macro_def() | |
1631 | _macro_def_prefix_pat = re.compile(r"!m(?:acro)?\b") | |
1632 | ||
1633 | # Tries to parse a macro definition, adding it to `self._macro_defs` | |
1634 | # and returning `True` on success. | |
1635 | def _try_parse_macro_def(self): | |
1636 | begin_text_loc = self._text_loc | |
1637 | ||
1638 | # Match prefix | |
1639 | if self._try_parse_pat(self._macro_def_prefix_pat) is None: | |
1640 | # No match | |
1641 | return False | |
1642 | ||
1643 | # Expect a name | |
ba11fb1d | 1644 | self._skip_ws_and_comments() |
320644e2 PP |
1645 | name_text_loc = self._text_loc |
1646 | m = self._expect_pat(_py_name_pat, "Expecting a valid macro name") | |
1647 | ||
1648 | # Validate name | |
1649 | name = m.group(0) | |
1650 | ||
1651 | if name in self._macro_defs: | |
1652 | _raise_error("Duplicate macro named `{}`".format(name), name_text_loc) | |
1653 | ||
1654 | # Expect `(` | |
ba11fb1d | 1655 | self._skip_ws_and_comments() |
320644e2 PP |
1656 | self._expect_pat(self._left_paren_pat, "Expecting `(`") |
1657 | ||
1658 | # Try to parse comma-separated parameter names | |
1659 | param_names = [] # type: List[str] | |
1660 | expect_comma = False | |
1661 | ||
1662 | while True: | |
ba11fb1d | 1663 | self._skip_ws_and_comments() |
320644e2 PP |
1664 | |
1665 | # End? | |
1666 | if self._try_parse_pat(self._right_paren_pat) is not None: | |
1667 | # End | |
1668 | break | |
1669 | ||
1670 | # Comma? | |
1671 | if expect_comma: | |
1672 | self._expect_pat(self._macro_params_comma_pat, "Expecting `,`") | |
1673 | ||
1674 | # Expect parameter name | |
ba11fb1d | 1675 | self._skip_ws_and_comments() |
320644e2 PP |
1676 | param_text_loc = self._text_loc |
1677 | m = self._expect_pat(_py_name_pat, "Expecting valid parameter name") | |
1678 | ||
1679 | if m.group(0) in param_names: | |
1680 | _raise_error( | |
1681 | "Duplicate macro parameter named `{}`".format(m.group(0)), | |
1682 | param_text_loc, | |
1683 | ) | |
1684 | ||
1685 | param_names.append(m.group(0)) | |
1686 | expect_comma = True | |
1687 | ||
1688 | # Expect items | |
ba11fb1d | 1689 | self._skip_ws_and_comments_and_syms() |
320644e2 PP |
1690 | items_text_loc = self._text_loc |
1691 | old_var_names = self._var_names.copy() | |
1692 | old_label_names = self._label_names.copy() | |
1693 | self._var_names = set() # type: Set[str] | |
1694 | self._label_names = set() # type: Set[str] | |
1695 | items = self._parse_items() | |
1696 | self._var_names = old_var_names | |
1697 | self._label_names = old_label_names | |
1698 | ||
1699 | # Expect suffix | |
1700 | self._expect_pat( | |
1701 | self._block_end_pat, "Expecting an item or `!end` (end of macro block)" | |
1702 | ) | |
1703 | ||
1704 | # Register macro | |
1705 | self._macro_defs[name] = _MacroDef( | |
1706 | name, param_names, _Group(items, items_text_loc), begin_text_loc | |
1707 | ) | |
1708 | ||
1709 | return True | |
1710 | ||
1711 | # Patterns for _try_parse_macro_exp() | |
1712 | _macro_exp_prefix_pat = re.compile(r"m\b") | |
1713 | _macro_exp_colon_pat = re.compile(r":") | |
1714 | ||
1715 | # Tries to parse a macro expansion, returning a macro expansion item | |
1716 | # on success. | |
1717 | def _try_parse_macro_exp(self): | |
1718 | begin_text_loc = self._text_loc | |
1719 | ||
1720 | # Match prefix | |
1721 | if self._try_parse_pat(self._macro_exp_prefix_pat) is None: | |
1722 | # No match | |
1723 | return | |
1724 | ||
1725 | # Expect `:` | |
ba11fb1d | 1726 | self._skip_ws_and_comments() |
320644e2 PP |
1727 | self._expect_pat(self._macro_exp_colon_pat, "Expecting `:`") |
1728 | ||
1729 | # Expect a macro name | |
ba11fb1d | 1730 | self._skip_ws_and_comments() |
320644e2 PP |
1731 | name_text_loc = self._text_loc |
1732 | m = self._expect_pat(_py_name_pat, "Expecting a valid macro name") | |
1733 | ||
1734 | # Validate name | |
1735 | name = m.group(0) | |
1736 | macro_def = self._macro_defs.get(name) | |
1737 | ||
1738 | if macro_def is None: | |
1739 | _raise_error("Unknown macro name `{}`".format(name), name_text_loc) | |
1740 | ||
1741 | # Expect `(` | |
ba11fb1d | 1742 | self._skip_ws_and_comments() |
320644e2 PP |
1743 | self._expect_pat(self._left_paren_pat, "Expecting `(`") |
1744 | ||
1745 | # Try to parse comma-separated parameter values | |
1746 | params_text_loc = self._text_loc | |
1747 | params = [] # type: List[_MacroExpParam] | |
1748 | expect_comma = False | |
1749 | ||
1750 | while True: | |
ba11fb1d | 1751 | self._skip_ws_and_comments() |
320644e2 PP |
1752 | |
1753 | # End? | |
1754 | if self._try_parse_pat(self._right_paren_pat) is not None: | |
1755 | # End | |
1756 | break | |
1757 | ||
7a7b31e8 | 1758 | # Expect a value |
320644e2 PP |
1759 | if expect_comma: |
1760 | self._expect_pat(self._macro_params_comma_pat, "Expecting `,`") | |
1761 | ||
ba11fb1d | 1762 | self._skip_ws_and_comments() |
320644e2 PP |
1763 | param_text_loc = self._text_loc |
1764 | params.append( | |
1765 | _MacroExpParam( | |
dbd84e74 PP |
1766 | *self._expect_expr( |
1767 | accept_const_int=True, | |
1768 | allow_neg_int=True, | |
1769 | accept_const_float=True, | |
7a7b31e8 | 1770 | accept_lit_str=True, |
dbd84e74 | 1771 | ), |
6dd69a2a | 1772 | text_loc=param_text_loc |
320644e2 PP |
1773 | ) |
1774 | ) | |
1775 | expect_comma = True | |
1776 | ||
1777 | # Validate parameter values | |
1778 | if len(params) != len(macro_def.param_names): | |
1779 | sing_plur = "" if len(params) == 1 else "s" | |
1780 | _raise_error( | |
1781 | "Macro expansion passes {} parameter{} while the definition expects {}".format( | |
1782 | len(params), sing_plur, len(macro_def.param_names) | |
1783 | ), | |
1784 | params_text_loc, | |
1785 | ) | |
1786 | ||
1787 | # Return item | |
1788 | return _MacroExp(name, params, begin_text_loc) | |
1789 | ||
71aaa3f7 PP |
1790 | # Tries to parse a base item (anything except a repetition), |
1791 | # returning it on success. | |
1792 | def _try_parse_base_item(self): | |
1793 | # Byte item? | |
1794 | item = self._try_parse_byte() | |
1795 | ||
1796 | if item is not None: | |
1797 | return item | |
1798 | ||
1799 | # String item? | |
1800 | item = self._try_parse_str() | |
1801 | ||
1802 | if item is not None: | |
1803 | return item | |
1804 | ||
2adf4336 PP |
1805 | # Value, variable assignment, or byte order setting item? |
1806 | item = self._try_parse_val_or_var_assign_or_set_bo() | |
71aaa3f7 PP |
1807 | |
1808 | if item is not None: | |
1809 | return item | |
1810 | ||
2adf4336 PP |
1811 | # Label or offset setting item? |
1812 | item = self._try_parse_label_or_set_offset() | |
71aaa3f7 | 1813 | |
676f6189 PP |
1814 | if item is not None: |
1815 | return item | |
1816 | ||
1817 | # Offset alignment item? | |
1818 | item = self._try_parse_align_offset() | |
1819 | ||
25ca454b PP |
1820 | if item is not None: |
1821 | return item | |
1822 | ||
1823 | # Filling item? | |
1824 | item = self._try_parse_fill_until() | |
1825 | ||
71aaa3f7 PP |
1826 | if item is not None: |
1827 | return item | |
1828 | ||
1829 | # Group item? | |
1830 | item = self._try_parse_group() | |
1831 | ||
1832 | if item is not None: | |
1833 | return item | |
1834 | ||
320644e2 | 1835 | # Repetition block item? |
e57a18e1 | 1836 | item = self._try_parse_rep_block() |
71aaa3f7 | 1837 | |
e57a18e1 PP |
1838 | if item is not None: |
1839 | return item | |
1840 | ||
27d52a19 PP |
1841 | # Conditional block item? |
1842 | item = self._try_parse_cond_block() | |
1843 | ||
1844 | if item is not None: | |
1845 | return item | |
1846 | ||
320644e2 PP |
1847 | # Macro expansion? |
1848 | item = self._try_parse_macro_exp() | |
1849 | ||
1850 | if item is not None: | |
1851 | return item | |
1852 | ||
e57a18e1 PP |
1853 | # Pattern for _try_parse_rep_post() |
1854 | _rep_post_prefix_pat = re.compile(r"\*") | |
1855 | ||
1856 | # Tries to parse a post-item repetition, returning the expression | |
1857 | # string and AST expression node on success. | |
1858 | def _try_parse_rep_post(self): | |
71aaa3f7 | 1859 | # Match prefix |
e57a18e1 | 1860 | if self._try_parse_pat(self._rep_post_prefix_pat) is None: |
71aaa3f7 | 1861 | # No match |
2adf4336 | 1862 | return |
71aaa3f7 | 1863 | |
e57a18e1 | 1864 | # Return expression string and AST expression |
71aaa3f7 | 1865 | self._skip_ws_and_comments() |
e57a18e1 | 1866 | return self._expect_rep_mul_expr() |
71aaa3f7 | 1867 | |
1ca7b5e1 PP |
1868 | # Tries to parse an item, possibly followed by a repetition, |
1869 | # returning `True` on success. | |
1870 | # | |
1871 | # Appends any parsed item to `items`. | |
1872 | def _try_append_item(self, items: List[_Item]): | |
ba11fb1d | 1873 | self._skip_ws_and_comments_and_syms() |
71aaa3f7 | 1874 | |
320644e2 | 1875 | # Base item |
71aaa3f7 PP |
1876 | item = self._try_parse_base_item() |
1877 | ||
1878 | if item is None: | |
320644e2 | 1879 | return |
71aaa3f7 PP |
1880 | |
1881 | # Parse repetition if the base item is repeatable | |
1882 | if isinstance(item, _RepableItem): | |
0e8e3169 PP |
1883 | self._skip_ws_and_comments() |
1884 | rep_text_loc = self._text_loc | |
e57a18e1 | 1885 | rep_ret = self._try_parse_rep_post() |
71aaa3f7 | 1886 | |
2adf4336 | 1887 | if rep_ret is not None: |
6dd69a2a | 1888 | item = _Rep(item, *rep_ret, text_loc=rep_text_loc) |
71aaa3f7 | 1889 | |
1ca7b5e1 PP |
1890 | items.append(item) |
1891 | return True | |
71aaa3f7 PP |
1892 | |
1893 | # Parses and returns items, skipping whitespaces, insignificant | |
1894 | # symbols, and comments when allowed, and stopping at the first | |
1895 | # unknown character. | |
320644e2 PP |
1896 | # |
1897 | # Accepts and registers macro definitions if `accept_macro_defs` | |
1898 | # is `True`. | |
1899 | def _parse_items(self, accept_macro_defs: bool = False) -> List[_Item]: | |
71aaa3f7 PP |
1900 | items = [] # type: List[_Item] |
1901 | ||
1902 | while self._isnt_done(): | |
1ca7b5e1 PP |
1903 | # Try to append item |
1904 | if not self._try_append_item(items): | |
320644e2 PP |
1905 | if accept_macro_defs and self._try_parse_macro_def(): |
1906 | continue | |
1907 | ||
1ca7b5e1 PP |
1908 | # Unknown at this point |
1909 | break | |
71aaa3f7 PP |
1910 | |
1911 | return items | |
1912 | ||
1913 | # Parses the whole Normand input, setting `self._res` to the main | |
1914 | # group item on success. | |
1915 | def _parse(self): | |
1916 | if len(self._normand.strip()) == 0: | |
1917 | # Special case to make sure there's something to consume | |
1918 | self._res = _Group([], self._text_loc) | |
1919 | return | |
1920 | ||
1921 | # Parse first level items | |
320644e2 | 1922 | items = self._parse_items(True) |
71aaa3f7 PP |
1923 | |
1924 | # Make sure there's nothing left | |
ba11fb1d | 1925 | self._skip_ws_and_comments_and_syms() |
71aaa3f7 PP |
1926 | |
1927 | if self._isnt_done(): | |
1928 | self._raise_error( | |
1929 | "Unexpected character `{}`".format(self._normand[self._at]) | |
1930 | ) | |
1931 | ||
1932 | # Set main group item | |
1933 | self._res = _Group(items, self._text_loc) | |
1934 | ||
1935 | ||
1936 | # The return type of parse(). | |
1937 | class ParseResult: | |
1938 | @classmethod | |
1939 | def _create( | |
1940 | cls, | |
1941 | data: bytearray, | |
e57a18e1 PP |
1942 | variables: VariablesT, |
1943 | labels: LabelsT, | |
71aaa3f7 PP |
1944 | offset: int, |
1945 | bo: Optional[ByteOrder], | |
1946 | ): | |
1947 | self = cls.__new__(cls) | |
1948 | self._init(data, variables, labels, offset, bo) | |
1949 | return self | |
1950 | ||
1951 | def __init__(self, *args, **kwargs): # type: ignore | |
1952 | raise NotImplementedError | |
1953 | ||
1954 | def _init( | |
1955 | self, | |
1956 | data: bytearray, | |
e57a18e1 PP |
1957 | variables: VariablesT, |
1958 | labels: LabelsT, | |
71aaa3f7 PP |
1959 | offset: int, |
1960 | bo: Optional[ByteOrder], | |
1961 | ): | |
1962 | self._data = data | |
1963 | self._vars = variables | |
1964 | self._labels = labels | |
1965 | self._offset = offset | |
1966 | self._bo = bo | |
1967 | ||
1968 | # Generated data. | |
1969 | @property | |
1970 | def data(self): | |
1971 | return self._data | |
1972 | ||
1973 | # Dictionary of updated variable names to their last computed value. | |
1974 | @property | |
1975 | def variables(self): | |
1976 | return self._vars | |
1977 | ||
1978 | # Dictionary of updated main group label names to their computed | |
1979 | # value. | |
1980 | @property | |
1981 | def labels(self): | |
1982 | return self._labels | |
1983 | ||
1984 | # Updated offset. | |
1985 | @property | |
1986 | def offset(self): | |
1987 | return self._offset | |
1988 | ||
1989 | # Updated byte order. | |
1990 | @property | |
1991 | def byte_order(self): | |
1992 | return self._bo | |
1993 | ||
1994 | ||
1995 | # Raises a parse error for the item `item`, creating it using the | |
1996 | # message `msg`. | |
1997 | def _raise_error_for_item(msg: str, item: _Item) -> NoReturn: | |
1998 | _raise_error(msg, item.text_loc) | |
1999 | ||
2000 | ||
2001 | # The `ICITTE` reserved name. | |
2002 | _icitte_name = "ICITTE" | |
2003 | ||
2004 | ||
2adf4336 PP |
2005 | # Base node visitor. |
2006 | # | |
2007 | # Calls the _visit_name() method for each name node which isn't the name | |
2008 | # of a call. | |
2009 | class _NodeVisitor(ast.NodeVisitor): | |
2010 | def __init__(self): | |
71aaa3f7 PP |
2011 | self._parent_is_call = False |
2012 | ||
2013 | def generic_visit(self, node: ast.AST): | |
2014 | if type(node) is ast.Call: | |
2015 | self._parent_is_call = True | |
2016 | elif type(node) is ast.Name and not self._parent_is_call: | |
2adf4336 | 2017 | self._visit_name(node.id) |
71aaa3f7 PP |
2018 | |
2019 | super().generic_visit(node) | |
2020 | self._parent_is_call = False | |
2021 | ||
2adf4336 PP |
2022 | @abc.abstractmethod |
2023 | def _visit_name(self, name: str): | |
2024 | ... | |
2025 | ||
71aaa3f7 | 2026 | |
2adf4336 PP |
2027 | # Expression validator: validates that all the names within the |
2028 | # expression are allowed. | |
2029 | class _ExprValidator(_NodeVisitor): | |
320644e2 | 2030 | def __init__(self, expr_str: str, text_loc: TextLocation, allowed_names: Set[str]): |
2adf4336 | 2031 | super().__init__() |
320644e2 PP |
2032 | self._expr_str = expr_str |
2033 | self._text_loc = text_loc | |
2adf4336 | 2034 | self._allowed_names = allowed_names |
2adf4336 PP |
2035 | |
2036 | def _visit_name(self, name: str): | |
2037 | # Make sure the name refers to a known and reachable | |
2038 | # variable/label name. | |
e57a18e1 | 2039 | if name != _icitte_name and name not in self._allowed_names: |
2adf4336 | 2040 | msg = "Illegal (unknown or unreachable) variable/label name `{}` in expression `{}`".format( |
320644e2 | 2041 | name, self._expr_str |
2adf4336 PP |
2042 | ) |
2043 | ||
05f81895 | 2044 | allowed_names = self._allowed_names.copy() |
e57a18e1 | 2045 | allowed_names.add(_icitte_name) |
2adf4336 | 2046 | |
05f81895 | 2047 | if len(allowed_names) > 0: |
2adf4336 PP |
2048 | allowed_names_str = ", ".join( |
2049 | sorted(["`{}`".format(name) for name in allowed_names]) | |
2050 | ) | |
2051 | msg += "; the legal names are {{{}}}".format(allowed_names_str) | |
2052 | ||
2053 | _raise_error( | |
2054 | msg, | |
320644e2 | 2055 | self._text_loc, |
2adf4336 PP |
2056 | ) |
2057 | ||
2058 | ||
2adf4336 PP |
2059 | # Generator state. |
2060 | class _GenState: | |
2061 | def __init__( | |
1b8aa84a | 2062 | self, |
e57a18e1 PP |
2063 | variables: VariablesT, |
2064 | labels: LabelsT, | |
1b8aa84a PP |
2065 | offset: int, |
2066 | bo: Optional[ByteOrder], | |
2adf4336 PP |
2067 | ): |
2068 | self.variables = variables.copy() | |
2069 | self.labels = labels.copy() | |
2070 | self.offset = offset | |
2071 | self.bo = bo | |
71aaa3f7 | 2072 | |
320644e2 PP |
2073 | def __repr__(self): |
2074 | return "_GenState({}, {}, {}, {})".format( | |
2075 | repr(self.variables), repr(self.labels), repr(self.offset), repr(self.bo) | |
2076 | ) | |
2077 | ||
2078 | ||
2079 | # Fixed-length number item instance. | |
2080 | class _FlNumItemInst: | |
f5dcb24c PP |
2081 | def __init__( |
2082 | self, | |
2083 | item: _FlNum, | |
2084 | offset_in_data: int, | |
2085 | state: _GenState, | |
2086 | parse_error_msgs: List[ParseErrorMessage], | |
2087 | ): | |
320644e2 PP |
2088 | self._item = item |
2089 | self._offset_in_data = offset_in_data | |
2090 | self._state = state | |
f5dcb24c | 2091 | self._parse_error_msgs = parse_error_msgs |
320644e2 PP |
2092 | |
2093 | @property | |
2094 | def item(self): | |
2095 | return self._item | |
2096 | ||
2097 | @property | |
2098 | def offset_in_data(self): | |
2099 | return self._offset_in_data | |
2100 | ||
2101 | @property | |
2102 | def state(self): | |
2103 | return self._state | |
2104 | ||
f5dcb24c PP |
2105 | @property |
2106 | def parse_error_msgs(self): | |
2107 | return self._parse_error_msgs | |
2108 | ||
71aaa3f7 | 2109 | |
2adf4336 | 2110 | # Generator of data and final state from a group item. |
71aaa3f7 PP |
2111 | # |
2112 | # Generation happens in memory at construction time. After building, use | |
2113 | # the `data`, `variables`, `labels`, `offset`, and `bo` properties to | |
2114 | # get the resulting context. | |
2adf4336 PP |
2115 | # |
2116 | # The steps of generation are: | |
2117 | # | |
320644e2 PP |
2118 | # 1. Handle each item in prefix order. |
2119 | # | |
2120 | # The handlers append bytes to `self._data` and update some current | |
2121 | # state object (`_GenState` instance). | |
2122 | # | |
2123 | # When handling a fixed-length number item, try to evaluate its | |
2124 | # expression using the current state. If this fails, then it might be | |
2125 | # because the expression refers to a "future" label: save the current | |
2126 | # offset in `self._data` (generated data) and a snapshot of the | |
2127 | # current state within `self._fl_num_item_insts` (`_FlNumItemInst` | |
f5dcb24c PP |
2128 | # object). _gen_fl_num_item_insts() will deal with this later. A |
2129 | # `_FlNumItemInst` instance also contains a snapshot of the current | |
2130 | # parsing error messages (`self._parse_error_msgs`) which need to be | |
2131 | # taken into account when handling the instance later. | |
2adf4336 | 2132 | # |
320644e2 PP |
2133 | # When handling the items of a group, keep a map of immediate label |
2134 | # names to their offset. Then, after having processed all the items, | |
2135 | # update the relevant saved state snapshots in | |
2136 | # `self._fl_num_item_insts` with those immediate label values. | |
2137 | # _gen_fl_num_item_insts() will deal with this later. | |
2adf4336 | 2138 | # |
320644e2 PP |
2139 | # 2. Handle all the fixed-length number item instances of which the |
2140 | # expression evaluation failed before. | |
2adf4336 | 2141 | # |
320644e2 PP |
2142 | # At this point, `self._fl_num_item_insts` contains everything that's |
2143 | # needed to evaluate the expressions, including the values of | |
2144 | # "future" labels from the point of view of some fixed-length number | |
2145 | # item instance. | |
2adf4336 | 2146 | # |
f5dcb24c PP |
2147 | # If an evaluation fails at this point, then it's a user error. Add |
2148 | # to the parsing error all the saved parsing error messages of the | |
2149 | # instance. Those additional messages add precious context to the | |
2150 | # error. | |
71aaa3f7 PP |
2151 | class _Gen: |
2152 | def __init__( | |
2153 | self, | |
2154 | group: _Group, | |
320644e2 | 2155 | macro_defs: _MacroDefsT, |
e57a18e1 PP |
2156 | variables: VariablesT, |
2157 | labels: LabelsT, | |
71aaa3f7 PP |
2158 | offset: int, |
2159 | bo: Optional[ByteOrder], | |
2160 | ): | |
320644e2 PP |
2161 | self._macro_defs = macro_defs |
2162 | self._fl_num_item_insts = [] # type: List[_FlNumItemInst] | |
f5dcb24c | 2163 | self._parse_error_msgs = [] # type: List[ParseErrorMessage] |
2adf4336 | 2164 | self._gen(group, _GenState(variables, labels, offset, bo)) |
71aaa3f7 PP |
2165 | |
2166 | # Generated bytes. | |
2167 | @property | |
2168 | def data(self): | |
2169 | return self._data | |
2170 | ||
2171 | # Updated variables. | |
2172 | @property | |
2173 | def variables(self): | |
2adf4336 | 2174 | return self._final_state.variables |
71aaa3f7 PP |
2175 | |
2176 | # Updated main group labels. | |
2177 | @property | |
2178 | def labels(self): | |
2adf4336 | 2179 | return self._final_state.labels |
71aaa3f7 PP |
2180 | |
2181 | # Updated offset. | |
2182 | @property | |
2183 | def offset(self): | |
2adf4336 | 2184 | return self._final_state.offset |
71aaa3f7 PP |
2185 | |
2186 | # Updated byte order. | |
2187 | @property | |
2188 | def bo(self): | |
2adf4336 PP |
2189 | return self._final_state.bo |
2190 | ||
320644e2 PP |
2191 | # Evaluates the expression `expr` of which the original string is |
2192 | # `expr_str` at the location `text_loc` considering the current | |
2adf4336 PP |
2193 | # generation state `state`. |
2194 | # | |
7a7b31e8 | 2195 | # If `accept_float` is `True`, then the type of the result may be |
269f6eb3 | 2196 | # `float` too. |
7a7b31e8 PP |
2197 | # |
2198 | # If `accept_str` is `True`, then the type of the result may be | |
2199 | # `str` too. | |
2adf4336 | 2200 | @staticmethod |
320644e2 PP |
2201 | def _eval_expr( |
2202 | expr_str: str, | |
2203 | expr: ast.Expression, | |
2204 | text_loc: TextLocation, | |
269f6eb3 | 2205 | state: _GenState, |
7a7b31e8 PP |
2206 | accept_float: bool = False, |
2207 | accept_str: bool = False, | |
269f6eb3 | 2208 | ): |
e57a18e1 PP |
2209 | syms = {} # type: VariablesT |
2210 | syms.update(state.labels) | |
71aaa3f7 | 2211 | |
e57a18e1 PP |
2212 | # Set the `ICITTE` name to the current offset |
2213 | syms[_icitte_name] = state.offset | |
71aaa3f7 PP |
2214 | |
2215 | # Add the current variables | |
2adf4336 | 2216 | syms.update(state.variables) |
71aaa3f7 PP |
2217 | |
2218 | # Validate the node and its children | |
320644e2 | 2219 | _ExprValidator(expr_str, text_loc, set(syms.keys())).visit(expr) |
71aaa3f7 PP |
2220 | |
2221 | # Compile and evaluate expression node | |
2222 | try: | |
320644e2 | 2223 | val = eval(compile(expr, "", "eval"), None, syms) |
71aaa3f7 | 2224 | except Exception as exc: |
320644e2 PP |
2225 | _raise_error( |
2226 | "Failed to evaluate expression `{}`: {}".format(expr_str, exc), | |
2227 | text_loc, | |
71aaa3f7 PP |
2228 | ) |
2229 | ||
27d52a19 PP |
2230 | # Convert `bool` result type to `int` to normalize |
2231 | if type(val) is bool: | |
2232 | val = int(val) | |
2233 | ||
269f6eb3 PP |
2234 | # Validate result type |
2235 | expected_types = {int} # type: Set[type] | |
269f6eb3 | 2236 | |
7a7b31e8 | 2237 | if accept_float: |
269f6eb3 | 2238 | expected_types.add(float) |
7a7b31e8 PP |
2239 | |
2240 | if accept_str: | |
2241 | expected_types.add(str) | |
269f6eb3 PP |
2242 | |
2243 | if type(val) not in expected_types: | |
7a7b31e8 PP |
2244 | expected_types_str = sorted( |
2245 | ["`{}`".format(t.__name__) for t in expected_types] | |
2246 | ) | |
2247 | ||
2248 | if len(expected_types_str) == 1: | |
2249 | msg_expected = expected_types_str[0] | |
2250 | elif len(expected_types_str) == 2: | |
2251 | msg_expected = " or ".join(expected_types_str) | |
2252 | else: | |
2253 | expected_types_str[-1] = "or {}".format(expected_types_str[-1]) | |
2254 | msg_expected = ", ".join(expected_types_str) | |
2255 | ||
320644e2 | 2256 | _raise_error( |
269f6eb3 | 2257 | "Invalid expression `{}`: expecting result type {}, not `{}`".format( |
7a7b31e8 | 2258 | expr_str, msg_expected, type(val).__name__ |
71aaa3f7 | 2259 | ), |
320644e2 | 2260 | text_loc, |
71aaa3f7 PP |
2261 | ) |
2262 | ||
2263 | return val | |
2264 | ||
7a7b31e8 PP |
2265 | # Forwards to _eval_expr() with the expression and text location of |
2266 | # `item`. | |
320644e2 PP |
2267 | @staticmethod |
2268 | def _eval_item_expr( | |
7a7b31e8 | 2269 | item: Union[_Cond, _FillUntil, _FlNum, _Leb128Int, _Rep, _Str, _VarAssign], |
320644e2 | 2270 | state: _GenState, |
7a7b31e8 PP |
2271 | accept_float: bool = False, |
2272 | accept_str: bool = False, | |
320644e2 PP |
2273 | ): |
2274 | return _Gen._eval_expr( | |
7a7b31e8 | 2275 | item.expr_str, item.expr, item.text_loc, state, accept_float, accept_str |
320644e2 PP |
2276 | ) |
2277 | ||
2278 | # Handles the byte item `item`. | |
2279 | def _handle_byte_item(self, item: _Byte, state: _GenState): | |
2280 | self._data.append(item.val) | |
2281 | state.offset += item.size | |
2282 | ||
7a7b31e8 PP |
2283 | # Handles the literal string item `item`. |
2284 | def _handle_lit_str_item(self, item: _LitStr, state: _GenState): | |
320644e2 PP |
2285 | self._data += item.data |
2286 | state.offset += item.size | |
2287 | ||
2288 | # Handles the byte order setting item `item`. | |
2289 | def _handle_set_bo_item(self, item: _SetBo, state: _GenState): | |
2290 | # Update current byte order | |
2291 | state.bo = item.bo | |
2292 | ||
2293 | # Handles the variable assignment item `item`. | |
2294 | def _handle_var_assign_item(self, item: _VarAssign, state: _GenState): | |
2295 | # Update variable | |
7a7b31e8 PP |
2296 | state.variables[item.name] = self._eval_item_expr( |
2297 | item, state, accept_float=True, accept_str=True | |
2298 | ) | |
320644e2 PP |
2299 | |
2300 | # Handles the fixed-length number item `item`. | |
2301 | def _handle_fl_num_item(self, item: _FlNum, state: _GenState): | |
2302 | # Validate current byte order | |
2303 | if state.bo is None and item.len > 8: | |
2304 | _raise_error_for_item( | |
2305 | "Current byte order isn't defined at first fixed-length number (`{}`) to encode on more than 8 bits".format( | |
2306 | item.expr_str | |
2307 | ), | |
2308 | item, | |
2309 | ) | |
2310 | ||
2311 | # Try an immediate evaluation. If it fails, then keep everything | |
2312 | # needed to (try to) generate the bytes of this item later. | |
2313 | try: | |
2314 | data = self._gen_fl_num_item_inst_data(item, state) | |
2315 | except Exception: | |
2316 | self._fl_num_item_insts.append( | |
f5dcb24c PP |
2317 | _FlNumItemInst( |
2318 | item, | |
2319 | len(self._data), | |
2320 | copy.deepcopy(state), | |
2321 | copy.deepcopy(self._parse_error_msgs), | |
2322 | ) | |
320644e2 PP |
2323 | ) |
2324 | ||
2325 | # Reserve space in `self._data` for this instance | |
2326 | data = bytes([0] * (item.len // 8)) | |
2327 | ||
2328 | # Append bytes | |
2329 | self._data += data | |
2330 | ||
2331 | # Update offset | |
2332 | state.offset += len(data) | |
2333 | ||
05f81895 PP |
2334 | # Returns the size, in bytes, required to encode the value `val` |
2335 | # with LEB128 (signed version if `is_signed` is `True`). | |
2336 | @staticmethod | |
2337 | def _leb128_size_for_val(val: int, is_signed: bool): | |
2338 | if val < 0: | |
2339 | # Equivalent upper bound. | |
2340 | # | |
2341 | # For example, if `val` is -128, then the full integer for | |
2342 | # this number of bits would be [-128, 127]. | |
2343 | val = -val - 1 | |
2344 | ||
2345 | # Number of bits (add one for the sign if needed) | |
2346 | bits = val.bit_length() + int(is_signed) | |
2347 | ||
2348 | if bits == 0: | |
2349 | bits = 1 | |
2350 | ||
2351 | # Seven bits per byte | |
2352 | return math.ceil(bits / 7) | |
2353 | ||
320644e2 PP |
2354 | # Handles the LEB128 integer item `item`. |
2355 | def _handle_leb128_int_item(self, item: _Leb128Int, state: _GenState): | |
2356 | # Compute value | |
7a7b31e8 | 2357 | val = self._eval_item_expr(item, state) |
676f6189 | 2358 | |
320644e2 PP |
2359 | # Size in bytes |
2360 | size = self._leb128_size_for_val(val, type(item) is _SLeb128Int) | |
05f81895 | 2361 | |
320644e2 PP |
2362 | # For each byte |
2363 | for _ in range(size): | |
2364 | # Seven LSBs, MSB of the byte set (continue) | |
2365 | self._data.append((val & 0x7F) | 0x80) | |
2366 | val >>= 7 | |
2adf4336 | 2367 | |
320644e2 PP |
2368 | # Clear MSB of last byte (stop) |
2369 | self._data[-1] &= ~0x80 | |
2adf4336 | 2370 | |
320644e2 PP |
2371 | # Update offset |
2372 | state.offset += size | |
27d52a19 | 2373 | |
7a7b31e8 PP |
2374 | # Handles the string item `item`. |
2375 | def _handle_str_item(self, item: _Str, state: _GenState): | |
2376 | # Compute value | |
2377 | val = str(self._eval_item_expr(item, state, accept_float=True, accept_str=True)) | |
2378 | ||
2379 | # Encode | |
2380 | data = _encode_str(val, item.codec, item.text_loc) | |
2381 | ||
2382 | # Add to data | |
2383 | self._data += data | |
2384 | ||
2385 | # Update offset | |
2386 | state.offset += len(data) | |
2387 | ||
320644e2 PP |
2388 | # Handles the group item `item`, removing the immediate labels from |
2389 | # `state` at the end if `remove_immediate_labels` is `True`. | |
2390 | def _handle_group_item( | |
2391 | self, item: _Group, state: _GenState, remove_immediate_labels: bool = True | |
2392 | ): | |
2393 | first_fl_num_item_inst_index = len(self._fl_num_item_insts) | |
2394 | immediate_labels = {} # type: LabelsT | |
27d52a19 | 2395 | |
320644e2 PP |
2396 | # Handle each item |
2397 | for subitem in item.items: | |
2398 | if type(subitem) is _Label: | |
2399 | # Add to local immediate labels | |
2400 | immediate_labels[subitem.name] = state.offset | |
2adf4336 | 2401 | |
320644e2 | 2402 | self._handle_item(subitem, state) |
2adf4336 | 2403 | |
320644e2 PP |
2404 | # Remove immediate labels from current state if needed |
2405 | if remove_immediate_labels: | |
2406 | for name in immediate_labels: | |
2407 | del state.labels[name] | |
2adf4336 | 2408 | |
320644e2 PP |
2409 | # Add all immediate labels to all state snapshots since |
2410 | # `first_fl_num_item_inst_index`. | |
2411 | for inst in self._fl_num_item_insts[first_fl_num_item_inst_index:]: | |
2412 | inst.state.labels.update(immediate_labels) | |
2adf4336 | 2413 | |
320644e2 PP |
2414 | # Handles the repetition item `item`. |
2415 | def _handle_rep_item(self, item: _Rep, state: _GenState): | |
2416 | # Compute the repetition count | |
2417 | mul = _Gen._eval_item_expr(item, state) | |
05f81895 | 2418 | |
320644e2 PP |
2419 | # Validate result |
2420 | if mul < 0: | |
2421 | _raise_error_for_item( | |
2422 | "Invalid expression `{}`: unexpected negative result {:,}".format( | |
2423 | item.expr_str, mul | |
2424 | ), | |
2425 | item, | |
2426 | ) | |
2adf4336 | 2427 | |
320644e2 PP |
2428 | # Generate item data `mul` times |
2429 | for _ in range(mul): | |
2430 | self._handle_item(item.item, state) | |
2adf4336 | 2431 | |
320644e2 | 2432 | # Handles the conditional item `item`. |
12b5dbc0 | 2433 | def _handle_cond_item(self, item: _Cond, state: _GenState): |
320644e2 PP |
2434 | # Compute the conditional value |
2435 | val = _Gen._eval_item_expr(item, state) | |
2adf4336 | 2436 | |
320644e2 PP |
2437 | # Generate item data if needed |
2438 | if val: | |
12b5dbc0 PP |
2439 | self._handle_item(item.true_item, state) |
2440 | else: | |
2441 | self._handle_item(item.false_item, state) | |
2adf4336 | 2442 | |
320644e2 PP |
2443 | # Evaluates the parameters of the macro expansion item `item` |
2444 | # considering the initial state `init_state` and returns a new state | |
2445 | # to handle the items of the macro. | |
2446 | def _eval_macro_exp_params(self, item: _MacroExp, init_state: _GenState): | |
2447 | # New state | |
2448 | exp_state = _GenState({}, {}, init_state.offset, init_state.bo) | |
2adf4336 | 2449 | |
320644e2 PP |
2450 | # Evaluate the parameter expressions |
2451 | macro_def = self._macro_defs[item.name] | |
2adf4336 | 2452 | |
320644e2 PP |
2453 | for param_name, param in zip(macro_def.param_names, item.params): |
2454 | exp_state.variables[param_name] = _Gen._eval_expr( | |
7a7b31e8 PP |
2455 | param.expr_str, |
2456 | param.expr, | |
2457 | param.text_loc, | |
2458 | init_state, | |
2459 | accept_float=True, | |
2460 | accept_str=True, | |
320644e2 | 2461 | ) |
2adf4336 | 2462 | |
320644e2 | 2463 | return exp_state |
2adf4336 | 2464 | |
320644e2 PP |
2465 | # Handles the macro expansion item `item`. |
2466 | def _handle_macro_exp_item(self, item: _MacroExp, state: _GenState): | |
f5dcb24c | 2467 | parse_error_msg_text = "While expanding the macro `{}`:".format(item.name) |
27d52a19 | 2468 | |
f5dcb24c PP |
2469 | try: |
2470 | # New state | |
2471 | exp_state = self._eval_macro_exp_params(item, state) | |
2472 | ||
2473 | # Process the contained group | |
2474 | init_data_size = len(self._data) | |
2475 | parse_error_msg = ( | |
2476 | ParseErrorMessage._create( # pyright: ignore[reportPrivateUsage] | |
2477 | parse_error_msg_text, item.text_loc | |
2478 | ) | |
2479 | ) | |
2480 | self._parse_error_msgs.append(parse_error_msg) | |
2481 | self._handle_item(self._macro_defs[item.name].group, exp_state) | |
2482 | self._parse_error_msgs.pop() | |
2483 | except ParseError as exc: | |
2484 | _augment_error(exc, parse_error_msg_text, item.text_loc) | |
27d52a19 | 2485 | |
320644e2 PP |
2486 | # Update state offset and return |
2487 | state.offset += len(self._data) - init_data_size | |
676f6189 | 2488 | |
320644e2 PP |
2489 | # Handles the offset setting item `item`. |
2490 | def _handle_set_offset_item(self, item: _SetOffset, state: _GenState): | |
676f6189 | 2491 | state.offset = item.val |
2adf4336 | 2492 | |
25ca454b | 2493 | # Handles the offset alignment item `item` (adds padding). |
320644e2 PP |
2494 | def _handle_align_offset_item(self, item: _AlignOffset, state: _GenState): |
2495 | init_offset = state.offset | |
2496 | align_bytes = item.val // 8 | |
2497 | state.offset = (state.offset + align_bytes - 1) // align_bytes * align_bytes | |
2498 | self._data += bytes([item.pad_val] * (state.offset - init_offset)) | |
2adf4336 | 2499 | |
25ca454b PP |
2500 | # Handles the filling item `item` (adds padding). |
2501 | def _handle_fill_until_item(self, item: _FillUntil, state: _GenState): | |
2502 | # Compute the new offset | |
2503 | new_offset = _Gen._eval_item_expr(item, state) | |
2504 | ||
2505 | # Validate the new offset | |
2506 | if new_offset < state.offset: | |
2507 | _raise_error_for_item( | |
2508 | "Invalid expression `{}`: new offset {:,} is less than current offset {:,}".format( | |
2509 | item.expr_str, new_offset, state.offset | |
2510 | ), | |
2511 | item, | |
2512 | ) | |
2513 | ||
2514 | # Fill | |
2515 | self._data += bytes([item.pad_val] * (new_offset - state.offset)) | |
2516 | ||
2517 | # Update offset | |
2518 | state.offset = new_offset | |
2519 | ||
320644e2 PP |
2520 | # Handles the label item `item`. |
2521 | def _handle_label_item(self, item: _Label, state: _GenState): | |
2522 | state.labels[item.name] = state.offset | |
2adf4336 | 2523 | |
320644e2 PP |
2524 | # Handles the item `item`, returning the updated next repetition |
2525 | # instance. | |
2526 | def _handle_item(self, item: _Item, state: _GenState): | |
2527 | return self._item_handlers[type(item)](item, state) | |
71aaa3f7 | 2528 | |
320644e2 PP |
2529 | # Generates the data for a fixed-length integer item instance having |
2530 | # the value `val` and returns it. | |
2531 | def _gen_fl_int_item_inst_data(self, val: int, item: _FlNum, state: _GenState): | |
71aaa3f7 PP |
2532 | # Validate range |
2533 | if val < -(2 ** (item.len - 1)) or val > 2**item.len - 1: | |
2534 | _raise_error_for_item( | |
320644e2 PP |
2535 | "Value {:,} is outside the {}-bit range when evaluating expression `{}`".format( |
2536 | val, item.len, item.expr_str | |
71aaa3f7 PP |
2537 | ), |
2538 | item, | |
2539 | ) | |
2540 | ||
2541 | # Encode result on 64 bits (to extend the sign bit whatever the | |
2542 | # value of `item.len`). | |
71aaa3f7 PP |
2543 | data = struct.pack( |
2544 | "{}{}".format( | |
2adf4336 | 2545 | ">" if state.bo in (None, ByteOrder.BE) else "<", |
71aaa3f7 PP |
2546 | "Q" if val >= 0 else "q", |
2547 | ), | |
2548 | val, | |
2549 | ) | |
2550 | ||
2551 | # Keep only the requested length | |
2552 | len_bytes = item.len // 8 | |
2553 | ||
2adf4336 | 2554 | if state.bo in (None, ByteOrder.BE): |
71aaa3f7 PP |
2555 | # Big endian: keep last bytes |
2556 | data = data[-len_bytes:] | |
2557 | else: | |
2558 | # Little endian: keep first bytes | |
2adf4336 | 2559 | assert state.bo == ByteOrder.LE |
71aaa3f7 PP |
2560 | data = data[:len_bytes] |
2561 | ||
320644e2 PP |
2562 | # Return data |
2563 | return data | |
269f6eb3 | 2564 | |
320644e2 PP |
2565 | # Generates the data for a fixed-length floating point number item |
2566 | # instance having the value `val` and returns it. | |
2567 | def _gen_fl_float_item_inst_data(self, val: float, item: _FlNum, state: _GenState): | |
269f6eb3 PP |
2568 | # Validate length |
2569 | if item.len not in (32, 64): | |
2570 | _raise_error_for_item( | |
2571 | "Invalid {}-bit length for a fixed-length floating point number (value {:,})".format( | |
2572 | item.len, val | |
2573 | ), | |
2574 | item, | |
2575 | ) | |
2576 | ||
320644e2 PP |
2577 | # Encode and return result |
2578 | return struct.pack( | |
269f6eb3 PP |
2579 | "{}{}".format( |
2580 | ">" if state.bo in (None, ByteOrder.BE) else "<", | |
2581 | "f" if item.len == 32 else "d", | |
2582 | ), | |
2583 | val, | |
2584 | ) | |
2585 | ||
320644e2 PP |
2586 | # Generates the data for a fixed-length number item instance and |
2587 | # returns it. | |
2588 | def _gen_fl_num_item_inst_data(self, item: _FlNum, state: _GenState): | |
269f6eb3 | 2589 | # Compute value |
e57a18e1 | 2590 | val = self._eval_item_expr(item, state, True) |
269f6eb3 | 2591 | |
269f6eb3 PP |
2592 | # Handle depending on type |
2593 | if type(val) is int: | |
320644e2 | 2594 | return self._gen_fl_int_item_inst_data(val, item, state) |
269f6eb3 PP |
2595 | else: |
2596 | assert type(val) is float | |
320644e2 | 2597 | return self._gen_fl_float_item_inst_data(val, item, state) |
05f81895 | 2598 | |
320644e2 PP |
2599 | # Generates the data for all the fixed-length number item instances |
2600 | # and writes it at the correct offset within `self._data`. | |
2601 | def _gen_fl_num_item_insts(self): | |
2602 | for inst in self._fl_num_item_insts: | |
2603 | # Generate bytes | |
f5dcb24c PP |
2604 | try: |
2605 | data = self._gen_fl_num_item_inst_data(inst.item, inst.state) | |
2606 | except ParseError as exc: | |
2607 | # Add all the saved parse error messages for this | |
2608 | # instance. | |
2609 | for msg in reversed(inst.parse_error_msgs): | |
2610 | _add_error_msg(exc, msg.text, msg.text_location) | |
2611 | ||
2612 | raise | |
05f81895 | 2613 | |
320644e2 PP |
2614 | # Insert bytes into `self._data` |
2615 | self._data[inst.offset_in_data : inst.offset_in_data + len(data)] = data | |
2adf4336 PP |
2616 | |
2617 | # Generates the data (`self._data`) and final state | |
2618 | # (`self._final_state`) from `group` and the initial state `state`. | |
2619 | def _gen(self, group: _Group, state: _GenState): | |
71aaa3f7 PP |
2620 | # Initial state |
2621 | self._data = bytearray() | |
71aaa3f7 PP |
2622 | |
2623 | # Item handlers | |
2624 | self._item_handlers = { | |
676f6189 | 2625 | _AlignOffset: self._handle_align_offset_item, |
71aaa3f7 | 2626 | _Byte: self._handle_byte_item, |
27d52a19 | 2627 | _Cond: self._handle_cond_item, |
25ca454b | 2628 | _FillUntil: self._handle_fill_until_item, |
269f6eb3 | 2629 | _FlNum: self._handle_fl_num_item, |
71aaa3f7 | 2630 | _Group: self._handle_group_item, |
2adf4336 | 2631 | _Label: self._handle_label_item, |
7a7b31e8 | 2632 | _LitStr: self._handle_lit_str_item, |
320644e2 | 2633 | _MacroExp: self._handle_macro_exp_item, |
71aaa3f7 | 2634 | _Rep: self._handle_rep_item, |
2adf4336 PP |
2635 | _SetBo: self._handle_set_bo_item, |
2636 | _SetOffset: self._handle_set_offset_item, | |
05f81895 | 2637 | _SLeb128Int: self._handle_leb128_int_item, |
2adf4336 | 2638 | _Str: self._handle_str_item, |
05f81895 | 2639 | _ULeb128Int: self._handle_leb128_int_item, |
2adf4336 | 2640 | _VarAssign: self._handle_var_assign_item, |
320644e2 | 2641 | } # type: Dict[type, Callable[[Any, _GenState], None]] |
2adf4336 PP |
2642 | |
2643 | # Handle the group item, _not_ removing the immediate labels | |
2644 | # because the `labels` property offers them. | |
320644e2 | 2645 | self._handle_group_item(group, state, False) |
2adf4336 PP |
2646 | |
2647 | # This is actually the final state | |
2648 | self._final_state = state | |
71aaa3f7 | 2649 | |
320644e2 PP |
2650 | # Generate all the fixed-length number bytes now that we know |
2651 | # their full state | |
2652 | self._gen_fl_num_item_insts() | |
2653 | ||
71aaa3f7 PP |
2654 | |
2655 | # Returns a `ParseResult` instance containing the bytes encoded by the | |
2656 | # input string `normand`. | |
2657 | # | |
2658 | # `init_variables` is a dictionary of initial variable names (valid | |
2659 | # Python names) to integral values. A variable name must not be the | |
2660 | # reserved name `ICITTE`. | |
2661 | # | |
2662 | # `init_labels` is a dictionary of initial label names (valid Python | |
2663 | # names) to integral values. A label name must not be the reserved name | |
2664 | # `ICITTE`. | |
2665 | # | |
2666 | # `init_offset` is the initial offset. | |
2667 | # | |
2668 | # `init_byte_order` is the initial byte order. | |
2669 | # | |
2670 | # Raises `ParseError` on any parsing error. | |
2671 | def parse( | |
2672 | normand: str, | |
e57a18e1 PP |
2673 | init_variables: Optional[VariablesT] = None, |
2674 | init_labels: Optional[LabelsT] = None, | |
71aaa3f7 PP |
2675 | init_offset: int = 0, |
2676 | init_byte_order: Optional[ByteOrder] = None, | |
2677 | ): | |
2678 | if init_variables is None: | |
2679 | init_variables = {} | |
2680 | ||
2681 | if init_labels is None: | |
2682 | init_labels = {} | |
2683 | ||
320644e2 | 2684 | parser = _Parser(normand, init_variables, init_labels) |
71aaa3f7 | 2685 | gen = _Gen( |
320644e2 PP |
2686 | parser.res, |
2687 | parser.macro_defs, | |
71aaa3f7 PP |
2688 | init_variables, |
2689 | init_labels, | |
2690 | init_offset, | |
2691 | init_byte_order, | |
2692 | ) | |
2693 | return ParseResult._create( # pyright: ignore[reportPrivateUsage] | |
2694 | gen.data, gen.variables, gen.labels, gen.offset, gen.bo | |
2695 | ) | |
2696 | ||
2697 | ||
f5dcb24c PP |
2698 | # Raises a command-line error with the message `msg`. |
2699 | def _raise_cli_error(msg: str) -> NoReturn: | |
2700 | raise RuntimeError("Command-line error: {}".format(msg)) | |
2701 | ||
2702 | ||
b2410769 PP |
2703 | # Returns the `int` or `float` value out of a CLI assignment value. |
2704 | def _val_from_assign_val_str(s: str, is_label: bool): | |
2705 | s = s.strip() | |
2706 | ||
2707 | # Floating point number? | |
2708 | if not is_label: | |
2709 | m = _const_float_pat.fullmatch(s) | |
2710 | ||
2711 | if m is not None: | |
2712 | return float(m.group(0)) | |
2713 | ||
2714 | # Integer? | |
2715 | m = _const_int_pat.fullmatch(s) | |
2716 | ||
2717 | if m is not None: | |
2718 | return int(_norm_const_int(m.group(0)), 0) | |
2719 | ||
2720 | exp = "an integer" if is_label else "a number" | |
2721 | _raise_cli_error("Invalid assignment value `{}`: expecting {}".format(s, exp)) | |
2722 | ||
2723 | ||
2724 | # Returns a dictionary of string to numbers from the list of strings | |
f5dcb24c | 2725 | # `args` containing `NAME=VAL` entries. |
7a7b31e8 | 2726 | def _dict_from_arg(args: Optional[List[str]], is_label: bool, is_str_only: bool): |
b2410769 | 2727 | d = {} # type: VariablesT |
f5dcb24c PP |
2728 | |
2729 | if args is None: | |
2730 | return d | |
2731 | ||
2732 | for arg in args: | |
7a7b31e8 | 2733 | m = re.match(r"({})\s*=\s*(.*)$".format(_py_name_pat.pattern), arg) |
f5dcb24c PP |
2734 | |
2735 | if m is None: | |
b2410769 | 2736 | _raise_cli_error("Invalid assignment `{}`".format(arg)) |
f5dcb24c | 2737 | |
7a7b31e8 PP |
2738 | if is_str_only: |
2739 | val = m.group(2) | |
2740 | else: | |
2741 | val = _val_from_assign_val_str(m.group(2), is_label) | |
2742 | ||
2743 | d[m.group(1)] = val | |
f5dcb24c PP |
2744 | |
2745 | return d | |
2746 | ||
2747 | ||
2748 | # Parses the command-line arguments and returns, in this order: | |
2749 | # | |
2750 | # 1. The input file path, or `None` if none. | |
2751 | # 2. The Normand input text. | |
2752 | # 3. The initial offset. | |
2753 | # 4. The initial byte order. | |
2754 | # 5. The initial variables. | |
2755 | # 6. The initial labels. | |
71aaa3f7 PP |
2756 | def _parse_cli_args(): |
2757 | import argparse | |
2758 | ||
2759 | # Build parser | |
2760 | ap = argparse.ArgumentParser() | |
2761 | ap.add_argument( | |
2762 | "--offset", | |
2763 | metavar="OFFSET", | |
2764 | action="store", | |
2765 | type=int, | |
2766 | default=0, | |
2767 | help="initial offset (positive)", | |
2768 | ) | |
2769 | ap.add_argument( | |
2770 | "-b", | |
2771 | "--byte-order", | |
2772 | metavar="BO", | |
2773 | choices=["be", "le"], | |
2774 | type=str, | |
2775 | help="initial byte order (`be` or `le`)", | |
2776 | ) | |
2777 | ap.add_argument( | |
b2410769 | 2778 | "-v", |
71aaa3f7 PP |
2779 | "--var", |
2780 | metavar="NAME=VAL", | |
2781 | action="append", | |
7a7b31e8 PP |
2782 | help="add an initial numeric variable (may be repeated)", |
2783 | ) | |
2784 | ap.add_argument( | |
2785 | "-s", | |
2786 | "--var-str", | |
2787 | metavar="NAME=VAL", | |
2788 | action="append", | |
2789 | help="add an initial string variable (may be repeated)", | |
71aaa3f7 PP |
2790 | ) |
2791 | ap.add_argument( | |
2792 | "-l", | |
2793 | "--label", | |
2794 | metavar="NAME=VAL", | |
2795 | action="append", | |
2796 | help="add an initial label (may be repeated)", | |
2797 | ) | |
2798 | ap.add_argument( | |
2799 | "--version", action="version", version="Normand {}".format(__version__) | |
2800 | ) | |
2801 | ap.add_argument( | |
2802 | "path", | |
2803 | metavar="PATH", | |
2804 | action="store", | |
2805 | nargs="?", | |
2806 | help="input path (none means standard input)", | |
2807 | ) | |
2808 | ||
2809 | # Parse | |
f5dcb24c | 2810 | args = ap.parse_args() |
71aaa3f7 PP |
2811 | |
2812 | # Read input | |
2813 | if args.path is None: | |
2814 | normand = sys.stdin.read() | |
2815 | else: | |
2816 | with open(args.path) as f: | |
2817 | normand = f.read() | |
2818 | ||
2819 | # Variables and labels | |
7a7b31e8 PP |
2820 | variables = _dict_from_arg(args.var, False, False) |
2821 | variables.update(_dict_from_arg(args.var_str, False, True)) | |
2822 | labels = _dict_from_arg(args.label, True, False) | |
71aaa3f7 PP |
2823 | |
2824 | # Validate offset | |
2825 | if args.offset < 0: | |
2826 | _raise_cli_error("Invalid negative offset {}") | |
2827 | ||
2828 | # Validate and set byte order | |
2829 | bo = None # type: Optional[ByteOrder] | |
2830 | ||
2831 | if args.byte_order is not None: | |
2832 | if args.byte_order == "be": | |
2833 | bo = ByteOrder.BE | |
2834 | else: | |
2835 | assert args.byte_order == "le" | |
2836 | bo = ByteOrder.LE | |
2837 | ||
f5dcb24c | 2838 | # Return input and initial state |
b2410769 | 2839 | return args.path, normand, args.offset, bo, variables, typing.cast(LabelsT, labels) |
71aaa3f7 | 2840 | |
71aaa3f7 | 2841 | |
f5dcb24c PP |
2842 | # CLI entry point without exception handling. |
2843 | def _run_cli_with_args( | |
2844 | normand: str, | |
2845 | offset: int, | |
2846 | bo: Optional[ByteOrder], | |
2847 | variables: VariablesT, | |
2848 | labels: LabelsT, | |
2849 | ): | |
2850 | sys.stdout.buffer.write(parse(normand, variables, labels, offset, bo).data) | |
71aaa3f7 PP |
2851 | |
2852 | ||
2853 | # Prints the exception message `msg` and exits with status 1. | |
2854 | def _fail(msg: str) -> NoReturn: | |
2855 | if not msg.endswith("."): | |
2856 | msg += "." | |
2857 | ||
f5dcb24c | 2858 | print(msg.strip(), file=sys.stderr) |
71aaa3f7 PP |
2859 | sys.exit(1) |
2860 | ||
2861 | ||
2862 | # CLI entry point. | |
2863 | def _run_cli(): | |
2864 | try: | |
f5dcb24c PP |
2865 | args = _parse_cli_args() |
2866 | except Exception as exc: | |
2867 | _fail(str(exc)) | |
2868 | ||
2869 | try: | |
2870 | _run_cli_with_args(*args[1:]) | |
2871 | except ParseError as exc: | |
2872 | import os.path | |
2873 | ||
2874 | prefix = "" if args[0] is None else "{}:".format(os.path.abspath(args[0])) | |
2875 | fail_msg = "" | |
2876 | ||
2877 | for msg in reversed(exc.messages): | |
2878 | fail_msg += "{}{}:{} - {}".format( | |
2879 | prefix, | |
2880 | msg.text_location.line_no, | |
2881 | msg.text_location.col_no, | |
2882 | msg.text, | |
2883 | ) | |
2884 | ||
2885 | if fail_msg[-1] not in ".:;": | |
2886 | fail_msg += "." | |
2887 | ||
2888 | fail_msg += "\n" | |
2889 | ||
2890 | _fail(fail_msg.strip()) | |
71aaa3f7 PP |
2891 | except Exception as exc: |
2892 | _fail(str(exc)) | |
2893 | ||
2894 | ||
2895 | if __name__ == "__main__": | |
2896 | _run_cli() |