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" |
12b5dbc0 | 33 | __version__ = "0.14.0" |
71aaa3f7 | 34 | __all__ = [ |
320644e2 PP |
35 | "__author__", |
36 | "__version__", | |
71aaa3f7 | 37 | "ByteOrder", |
320644e2 | 38 | "LabelsT", |
71aaa3f7 PP |
39 | "parse", |
40 | "ParseError", | |
41 | "ParseResult", | |
e57a18e1 | 42 | "TextLocation", |
e57a18e1 | 43 | "VariablesT", |
71aaa3f7 PP |
44 | ] |
45 | ||
46 | import re | |
47 | import abc | |
48 | import ast | |
49 | import sys | |
320644e2 | 50 | import copy |
71aaa3f7 | 51 | import enum |
05f81895 | 52 | import math |
71aaa3f7 | 53 | import struct |
e57a18e1 PP |
54 | import typing |
55 | from typing import Any, Set, Dict, List, Union, Pattern, Callable, NoReturn, Optional | |
71aaa3f7 PP |
56 | |
57 | ||
58 | # Text location (line and column numbers). | |
e57a18e1 | 59 | class TextLocation: |
71aaa3f7 PP |
60 | @classmethod |
61 | def _create(cls, line_no: int, col_no: int): | |
62 | self = cls.__new__(cls) | |
63 | self._init(line_no, col_no) | |
64 | return self | |
65 | ||
66 | def __init__(*args, **kwargs): # type: ignore | |
67 | raise NotImplementedError | |
68 | ||
69 | def _init(self, line_no: int, col_no: int): | |
70 | self._line_no = line_no | |
71 | self._col_no = col_no | |
72 | ||
73 | # Line number. | |
74 | @property | |
75 | def line_no(self): | |
76 | return self._line_no | |
77 | ||
78 | # Column number. | |
79 | @property | |
80 | def col_no(self): | |
81 | return self._col_no | |
82 | ||
2adf4336 | 83 | def __repr__(self): |
e57a18e1 | 84 | return "TextLocation({}, {})".format(self._line_no, self._col_no) |
2adf4336 | 85 | |
71aaa3f7 PP |
86 | |
87 | # Any item. | |
88 | class _Item: | |
e57a18e1 | 89 | def __init__(self, text_loc: TextLocation): |
71aaa3f7 PP |
90 | self._text_loc = text_loc |
91 | ||
92 | # Source text location. | |
93 | @property | |
94 | def text_loc(self): | |
95 | return self._text_loc | |
96 | ||
2adf4336 PP |
97 | |
98 | # Scalar item. | |
99 | class _ScalarItem(_Item): | |
71aaa3f7 PP |
100 | # Returns the size, in bytes, of this item. |
101 | @property | |
102 | @abc.abstractmethod | |
103 | def size(self) -> int: | |
104 | ... | |
105 | ||
106 | ||
107 | # A repeatable item. | |
2adf4336 | 108 | class _RepableItem: |
71aaa3f7 PP |
109 | pass |
110 | ||
111 | ||
112 | # Single byte. | |
2adf4336 | 113 | class _Byte(_ScalarItem, _RepableItem): |
e57a18e1 | 114 | def __init__(self, val: int, text_loc: TextLocation): |
71aaa3f7 PP |
115 | super().__init__(text_loc) |
116 | self._val = val | |
117 | ||
118 | # Byte value. | |
119 | @property | |
120 | def val(self): | |
121 | return self._val | |
122 | ||
123 | @property | |
124 | def size(self): | |
125 | return 1 | |
126 | ||
127 | def __repr__(self): | |
676f6189 | 128 | return "_Byte({}, {})".format(hex(self._val), repr(self._text_loc)) |
71aaa3f7 PP |
129 | |
130 | ||
131 | # String. | |
2adf4336 | 132 | class _Str(_ScalarItem, _RepableItem): |
e57a18e1 | 133 | def __init__(self, data: bytes, text_loc: TextLocation): |
71aaa3f7 PP |
134 | super().__init__(text_loc) |
135 | self._data = data | |
136 | ||
137 | # Encoded bytes. | |
138 | @property | |
139 | def data(self): | |
140 | return self._data | |
141 | ||
142 | @property | |
143 | def size(self): | |
144 | return len(self._data) | |
145 | ||
146 | def __repr__(self): | |
676f6189 | 147 | return "_Str({}, {})".format(repr(self._data), repr(self._text_loc)) |
71aaa3f7 PP |
148 | |
149 | ||
150 | # Byte order. | |
151 | @enum.unique | |
152 | class ByteOrder(enum.Enum): | |
153 | # Big endian. | |
154 | BE = "be" | |
155 | ||
156 | # Little endian. | |
157 | LE = "le" | |
158 | ||
159 | ||
2adf4336 PP |
160 | # Byte order setting. |
161 | class _SetBo(_Item): | |
e57a18e1 | 162 | def __init__(self, bo: ByteOrder, text_loc: TextLocation): |
0e8e3169 | 163 | super().__init__(text_loc) |
71aaa3f7 PP |
164 | self._bo = bo |
165 | ||
166 | @property | |
167 | def bo(self): | |
168 | return self._bo | |
169 | ||
2adf4336 | 170 | def __repr__(self): |
676f6189 | 171 | return "_SetBo({}, {})".format(repr(self._bo), repr(self._text_loc)) |
71aaa3f7 PP |
172 | |
173 | ||
174 | # Label. | |
175 | class _Label(_Item): | |
e57a18e1 | 176 | def __init__(self, name: str, text_loc: TextLocation): |
71aaa3f7 PP |
177 | super().__init__(text_loc) |
178 | self._name = name | |
179 | ||
180 | # Label name. | |
181 | @property | |
182 | def name(self): | |
183 | return self._name | |
184 | ||
71aaa3f7 | 185 | def __repr__(self): |
676f6189 | 186 | return "_Label({}, {})".format(repr(self._name), repr(self._text_loc)) |
71aaa3f7 PP |
187 | |
188 | ||
2adf4336 PP |
189 | # Offset setting. |
190 | class _SetOffset(_Item): | |
e57a18e1 | 191 | def __init__(self, val: int, text_loc: TextLocation): |
71aaa3f7 PP |
192 | super().__init__(text_loc) |
193 | self._val = val | |
194 | ||
676f6189 | 195 | # Offset value (bytes). |
71aaa3f7 PP |
196 | @property |
197 | def val(self): | |
198 | return self._val | |
199 | ||
71aaa3f7 | 200 | def __repr__(self): |
676f6189 PP |
201 | return "_SetOffset({}, {})".format(repr(self._val), repr(self._text_loc)) |
202 | ||
203 | ||
204 | # Offset alignment. | |
205 | class _AlignOffset(_Item): | |
e57a18e1 | 206 | def __init__(self, val: int, pad_val: int, text_loc: TextLocation): |
676f6189 PP |
207 | super().__init__(text_loc) |
208 | self._val = val | |
209 | self._pad_val = pad_val | |
210 | ||
211 | # Alignment value (bits). | |
212 | @property | |
213 | def val(self): | |
214 | return self._val | |
215 | ||
216 | # Padding byte value. | |
217 | @property | |
218 | def pad_val(self): | |
219 | return self._pad_val | |
220 | ||
221 | def __repr__(self): | |
222 | return "_AlignOffset({}, {}, {})".format( | |
223 | repr(self._val), repr(self._pad_val), repr(self._text_loc) | |
224 | ) | |
71aaa3f7 PP |
225 | |
226 | ||
227 | # Mixin of containing an AST expression and its string. | |
228 | class _ExprMixin: | |
229 | def __init__(self, expr_str: str, expr: ast.Expression): | |
230 | self._expr_str = expr_str | |
231 | self._expr = expr | |
232 | ||
233 | # Expression string. | |
234 | @property | |
235 | def expr_str(self): | |
236 | return self._expr_str | |
237 | ||
238 | # Expression node to evaluate. | |
239 | @property | |
240 | def expr(self): | |
241 | return self._expr | |
242 | ||
243 | ||
25ca454b PP |
244 | # Fill until some offset. |
245 | class _FillUntil(_Item, _ExprMixin): | |
246 | def __init__( | |
247 | self, expr_str: str, expr: ast.Expression, pad_val: int, text_loc: TextLocation | |
248 | ): | |
249 | super().__init__(text_loc) | |
250 | _ExprMixin.__init__(self, expr_str, expr) | |
251 | self._pad_val = pad_val | |
252 | ||
253 | # Padding byte value. | |
254 | @property | |
255 | def pad_val(self): | |
256 | return self._pad_val | |
257 | ||
258 | def __repr__(self): | |
259 | return "_FillUntil({}, {}, {}, {})".format( | |
260 | repr(self._expr_str), | |
261 | repr(self._expr), | |
262 | repr(self._pad_val), | |
263 | repr(self._text_loc), | |
264 | ) | |
265 | ||
266 | ||
2adf4336 PP |
267 | # Variable assignment. |
268 | class _VarAssign(_Item, _ExprMixin): | |
71aaa3f7 | 269 | def __init__( |
e57a18e1 | 270 | self, name: str, expr_str: str, expr: ast.Expression, text_loc: TextLocation |
71aaa3f7 PP |
271 | ): |
272 | super().__init__(text_loc) | |
273 | _ExprMixin.__init__(self, expr_str, expr) | |
274 | self._name = name | |
275 | ||
276 | # Name. | |
277 | @property | |
278 | def name(self): | |
279 | return self._name | |
280 | ||
71aaa3f7 | 281 | def __repr__(self): |
2adf4336 | 282 | return "_VarAssign({}, {}, {}, {})".format( |
676f6189 PP |
283 | repr(self._name), |
284 | repr(self._expr_str), | |
285 | repr(self._expr), | |
286 | repr(self._text_loc), | |
71aaa3f7 PP |
287 | ) |
288 | ||
289 | ||
269f6eb3 PP |
290 | # Fixed-length number, possibly needing more than one byte. |
291 | class _FlNum(_ScalarItem, _RepableItem, _ExprMixin): | |
71aaa3f7 | 292 | def __init__( |
e57a18e1 | 293 | self, expr_str: str, expr: ast.Expression, len: int, text_loc: TextLocation |
71aaa3f7 PP |
294 | ): |
295 | super().__init__(text_loc) | |
296 | _ExprMixin.__init__(self, expr_str, expr) | |
297 | self._len = len | |
298 | ||
299 | # Length (bits). | |
300 | @property | |
301 | def len(self): | |
302 | return self._len | |
303 | ||
304 | @property | |
305 | def size(self): | |
306 | return self._len // 8 | |
307 | ||
308 | def __repr__(self): | |
269f6eb3 | 309 | return "_FlNum({}, {}, {}, {})".format( |
676f6189 PP |
310 | repr(self._expr_str), |
311 | repr(self._expr), | |
312 | repr(self._len), | |
313 | repr(self._text_loc), | |
71aaa3f7 PP |
314 | ) |
315 | ||
316 | ||
05f81895 PP |
317 | # LEB128 integer. |
318 | class _Leb128Int(_Item, _RepableItem, _ExprMixin): | |
e57a18e1 | 319 | def __init__(self, expr_str: str, expr: ast.Expression, text_loc: TextLocation): |
05f81895 PP |
320 | super().__init__(text_loc) |
321 | _ExprMixin.__init__(self, expr_str, expr) | |
322 | ||
323 | def __repr__(self): | |
324 | return "{}({}, {}, {})".format( | |
325 | self.__class__.__name__, | |
326 | repr(self._expr_str), | |
327 | repr(self._expr), | |
676f6189 | 328 | repr(self._text_loc), |
05f81895 PP |
329 | ) |
330 | ||
331 | ||
332 | # Unsigned LEB128 integer. | |
333 | class _ULeb128Int(_Leb128Int, _RepableItem, _ExprMixin): | |
334 | pass | |
335 | ||
336 | ||
337 | # Signed LEB128 integer. | |
338 | class _SLeb128Int(_Leb128Int, _RepableItem, _ExprMixin): | |
339 | pass | |
340 | ||
341 | ||
71aaa3f7 | 342 | # Group of items. |
2adf4336 | 343 | class _Group(_Item, _RepableItem): |
e57a18e1 | 344 | def __init__(self, items: List[_Item], text_loc: TextLocation): |
71aaa3f7 PP |
345 | super().__init__(text_loc) |
346 | self._items = items | |
71aaa3f7 PP |
347 | |
348 | # Contained items. | |
349 | @property | |
350 | def items(self): | |
351 | return self._items | |
352 | ||
71aaa3f7 | 353 | def __repr__(self): |
676f6189 | 354 | return "_Group({}, {})".format(repr(self._items), repr(self._text_loc)) |
71aaa3f7 PP |
355 | |
356 | ||
357 | # Repetition item. | |
2adf4336 PP |
358 | class _Rep(_Item, _ExprMixin): |
359 | def __init__( | |
e57a18e1 | 360 | self, item: _Item, expr_str: str, expr: ast.Expression, text_loc: TextLocation |
2adf4336 | 361 | ): |
71aaa3f7 | 362 | super().__init__(text_loc) |
2adf4336 | 363 | _ExprMixin.__init__(self, expr_str, expr) |
71aaa3f7 | 364 | self._item = item |
71aaa3f7 PP |
365 | |
366 | # Item to repeat. | |
367 | @property | |
368 | def item(self): | |
369 | return self._item | |
370 | ||
71aaa3f7 | 371 | def __repr__(self): |
2adf4336 | 372 | return "_Rep({}, {}, {}, {})".format( |
676f6189 PP |
373 | repr(self._item), |
374 | repr(self._expr_str), | |
375 | repr(self._expr), | |
376 | repr(self._text_loc), | |
71aaa3f7 PP |
377 | ) |
378 | ||
379 | ||
27d52a19 PP |
380 | # Conditional item. |
381 | class _Cond(_Item, _ExprMixin): | |
382 | def __init__( | |
12b5dbc0 PP |
383 | self, |
384 | true_item: _Item, | |
385 | false_item: _Item, | |
386 | expr_str: str, | |
387 | expr: ast.Expression, | |
388 | text_loc: TextLocation, | |
27d52a19 PP |
389 | ): |
390 | super().__init__(text_loc) | |
391 | _ExprMixin.__init__(self, expr_str, expr) | |
12b5dbc0 PP |
392 | self._true_item = true_item |
393 | self._false_item = false_item | |
27d52a19 | 394 | |
12b5dbc0 | 395 | # Item when condition is true. |
27d52a19 | 396 | @property |
12b5dbc0 PP |
397 | def true_item(self): |
398 | return self._true_item | |
399 | ||
400 | # Item when condition is false. | |
401 | @property | |
402 | def false_item(self): | |
403 | return self._false_item | |
27d52a19 PP |
404 | |
405 | def __repr__(self): | |
12b5dbc0 PP |
406 | return "_Cond({}, {}, {}, {}, {})".format( |
407 | repr(self._true_item), | |
408 | repr(self._false_item), | |
27d52a19 PP |
409 | repr(self._expr_str), |
410 | repr(self._expr), | |
411 | repr(self._text_loc), | |
412 | ) | |
413 | ||
414 | ||
320644e2 PP |
415 | # Macro definition item. |
416 | class _MacroDef(_Item): | |
417 | def __init__( | |
418 | self, name: str, param_names: List[str], group: _Group, text_loc: TextLocation | |
419 | ): | |
420 | super().__init__(text_loc) | |
421 | self._name = name | |
422 | self._param_names = param_names | |
423 | self._group = group | |
424 | ||
425 | # Name. | |
426 | @property | |
427 | def name(self): | |
428 | return self._name | |
429 | ||
430 | # Parameters. | |
431 | @property | |
432 | def param_names(self): | |
433 | return self._param_names | |
434 | ||
435 | # Contained items. | |
436 | @property | |
437 | def group(self): | |
438 | return self._group | |
439 | ||
440 | def __repr__(self): | |
441 | return "_MacroDef({}, {}, {}, {})".format( | |
442 | repr(self._name), | |
443 | repr(self._param_names), | |
444 | repr(self._group), | |
445 | repr(self._text_loc), | |
446 | ) | |
447 | ||
448 | ||
449 | # Macro expansion parameter. | |
450 | class _MacroExpParam: | |
451 | def __init__(self, expr_str: str, expr: ast.Expression, text_loc: TextLocation): | |
452 | self._expr_str = expr_str | |
453 | self._expr = expr | |
454 | self._text_loc = text_loc | |
455 | ||
456 | # Expression string. | |
457 | @property | |
458 | def expr_str(self): | |
459 | return self._expr_str | |
460 | ||
461 | # Expression. | |
462 | @property | |
463 | def expr(self): | |
464 | return self._expr | |
465 | ||
466 | # Source text location. | |
467 | @property | |
468 | def text_loc(self): | |
469 | return self._text_loc | |
470 | ||
471 | def __repr__(self): | |
472 | return "_MacroExpParam({}, {}, {})".format( | |
473 | repr(self._expr_str), repr(self._expr), repr(self._text_loc) | |
474 | ) | |
475 | ||
476 | ||
477 | # Macro expansion item. | |
478 | class _MacroExp(_Item, _RepableItem): | |
479 | def __init__( | |
480 | self, | |
481 | name: str, | |
482 | params: List[_MacroExpParam], | |
483 | text_loc: TextLocation, | |
484 | ): | |
485 | super().__init__(text_loc) | |
486 | self._name = name | |
487 | self._params = params | |
488 | ||
489 | # Name. | |
490 | @property | |
491 | def name(self): | |
492 | return self._name | |
493 | ||
494 | # Parameters. | |
495 | @property | |
496 | def params(self): | |
497 | return self._params | |
498 | ||
499 | def __repr__(self): | |
500 | return "_MacroExp({}, {}, {})".format( | |
501 | repr(self._name), | |
502 | repr(self._params), | |
503 | repr(self._text_loc), | |
504 | ) | |
2adf4336 PP |
505 | |
506 | ||
71aaa3f7 PP |
507 | # A parsing error containing a message and a text location. |
508 | class ParseError(RuntimeError): | |
509 | @classmethod | |
e57a18e1 | 510 | def _create(cls, msg: str, text_loc: TextLocation): |
71aaa3f7 PP |
511 | self = cls.__new__(cls) |
512 | self._init(msg, text_loc) | |
513 | return self | |
514 | ||
515 | def __init__(self, *args, **kwargs): # type: ignore | |
516 | raise NotImplementedError | |
517 | ||
e57a18e1 | 518 | def _init(self, msg: str, text_loc: TextLocation): |
71aaa3f7 PP |
519 | super().__init__(msg) |
520 | self._text_loc = text_loc | |
521 | ||
522 | # Source text location. | |
523 | @property | |
524 | def text_loc(self): | |
525 | return self._text_loc | |
526 | ||
527 | ||
528 | # Raises a parsing error, forwarding the parameters to the constructor. | |
e57a18e1 | 529 | def _raise_error(msg: str, text_loc: TextLocation) -> NoReturn: |
71aaa3f7 PP |
530 | raise ParseError._create(msg, text_loc) # pyright: ignore[reportPrivateUsage] |
531 | ||
532 | ||
e57a18e1 PP |
533 | # Variables dictionary type (for type hints). |
534 | VariablesT = Dict[str, Union[int, float]] | |
535 | ||
536 | ||
537 | # Labels dictionary type (for type hints). | |
538 | LabelsT = Dict[str, int] | |
71aaa3f7 PP |
539 | |
540 | ||
541 | # Python name pattern. | |
542 | _py_name_pat = re.compile(r"[a-zA-Z_][a-zA-Z0-9_]*") | |
543 | ||
544 | ||
320644e2 PP |
545 | # Macro definition dictionary. |
546 | _MacroDefsT = Dict[str, _MacroDef] | |
547 | ||
548 | ||
71aaa3f7 PP |
549 | # Normand parser. |
550 | # | |
551 | # The constructor accepts a Normand input. After building, use the `res` | |
552 | # property to get the resulting main group. | |
553 | class _Parser: | |
554 | # Builds a parser to parse the Normand input `normand`, parsing | |
555 | # immediately. | |
e57a18e1 | 556 | def __init__(self, normand: str, variables: VariablesT, labels: LabelsT): |
71aaa3f7 PP |
557 | self._normand = normand |
558 | self._at = 0 | |
559 | self._line_no = 1 | |
560 | self._col_no = 1 | |
561 | self._label_names = set(labels.keys()) | |
562 | self._var_names = set(variables.keys()) | |
320644e2 | 563 | self._macro_defs = {} # type: _MacroDefsT |
71aaa3f7 PP |
564 | self._parse() |
565 | ||
566 | # Result (main group). | |
567 | @property | |
568 | def res(self): | |
569 | return self._res | |
570 | ||
320644e2 PP |
571 | # Macro definitions. |
572 | @property | |
573 | def macro_defs(self): | |
574 | return self._macro_defs | |
575 | ||
71aaa3f7 PP |
576 | # Current text location. |
577 | @property | |
578 | def _text_loc(self): | |
e57a18e1 | 579 | return TextLocation._create( # pyright: ignore[reportPrivateUsage] |
71aaa3f7 PP |
580 | self._line_no, self._col_no |
581 | ) | |
582 | ||
583 | # Returns `True` if this parser is done parsing. | |
584 | def _is_done(self): | |
585 | return self._at == len(self._normand) | |
586 | ||
587 | # Returns `True` if this parser isn't done parsing. | |
588 | def _isnt_done(self): | |
589 | return not self._is_done() | |
590 | ||
591 | # Raises a parse error, creating it using the message `msg` and the | |
592 | # current text location. | |
593 | def _raise_error(self, msg: str) -> NoReturn: | |
594 | _raise_error(msg, self._text_loc) | |
595 | ||
596 | # Tries to make the pattern `pat` match the current substring, | |
597 | # returning the match object and updating `self._at`, | |
598 | # `self._line_no`, and `self._col_no` on success. | |
599 | def _try_parse_pat(self, pat: Pattern[str]): | |
600 | m = pat.match(self._normand, self._at) | |
601 | ||
602 | if m is None: | |
603 | return | |
604 | ||
605 | # Skip matched string | |
606 | self._at += len(m.group(0)) | |
607 | ||
608 | # Update line number | |
609 | self._line_no += m.group(0).count("\n") | |
610 | ||
611 | # Update column number | |
612 | for i in reversed(range(self._at)): | |
613 | if self._normand[i] == "\n" or i == 0: | |
614 | if i == 0: | |
615 | self._col_no = self._at + 1 | |
616 | else: | |
617 | self._col_no = self._at - i | |
618 | ||
619 | break | |
620 | ||
621 | # Return match object | |
622 | return m | |
623 | ||
624 | # Expects the pattern `pat` to match the current substring, | |
625 | # returning the match object and updating `self._at`, | |
626 | # `self._line_no`, and `self._col_no` on success, or raising a parse | |
627 | # error with the message `error_msg` on error. | |
628 | def _expect_pat(self, pat: Pattern[str], error_msg: str): | |
629 | # Match | |
630 | m = self._try_parse_pat(pat) | |
631 | ||
632 | if m is None: | |
633 | # No match: error | |
634 | self._raise_error(error_msg) | |
635 | ||
636 | # Return match object | |
637 | return m | |
638 | ||
639 | # Pattern for _skip_ws_and_comments() | |
640 | _ws_or_syms_or_comments_pat = re.compile( | |
25ca454b | 641 | r"(?:[\s/\\?&:;.,[\]_=|-]|#[^#]*?(?:\n|#))*" |
71aaa3f7 PP |
642 | ) |
643 | ||
644 | # Skips as many whitespaces, insignificant symbol characters, and | |
645 | # comments as possible. | |
646 | def _skip_ws_and_comments(self): | |
647 | self._try_parse_pat(self._ws_or_syms_or_comments_pat) | |
648 | ||
320644e2 PP |
649 | # Pattern for _skip_ws() |
650 | _ws_pat = re.compile(r"\s*") | |
651 | ||
652 | # Skips as many whitespaces as possible. | |
653 | def _skip_ws(self): | |
654 | self._try_parse_pat(self._ws_pat) | |
655 | ||
71aaa3f7 PP |
656 | # Pattern for _try_parse_hex_byte() |
657 | _nibble_pat = re.compile(r"[A-Fa-f0-9]") | |
658 | ||
659 | # Tries to parse a hexadecimal byte, returning a byte item on | |
660 | # success. | |
661 | def _try_parse_hex_byte(self): | |
0e8e3169 PP |
662 | begin_text_loc = self._text_loc |
663 | ||
71aaa3f7 PP |
664 | # Match initial nibble |
665 | m_high = self._try_parse_pat(self._nibble_pat) | |
666 | ||
667 | if m_high is None: | |
668 | # No match | |
669 | return | |
670 | ||
671 | # Expect another nibble | |
672 | self._skip_ws_and_comments() | |
673 | m_low = self._expect_pat( | |
674 | self._nibble_pat, "Expecting another hexadecimal nibble" | |
675 | ) | |
676 | ||
677 | # Return item | |
0e8e3169 | 678 | return _Byte(int(m_high.group(0) + m_low.group(0), 16), begin_text_loc) |
71aaa3f7 PP |
679 | |
680 | # Patterns for _try_parse_bin_byte() | |
681 | _bin_byte_bit_pat = re.compile(r"[01]") | |
682 | _bin_byte_prefix_pat = re.compile(r"%") | |
683 | ||
684 | # Tries to parse a binary byte, returning a byte item on success. | |
685 | def _try_parse_bin_byte(self): | |
0e8e3169 PP |
686 | begin_text_loc = self._text_loc |
687 | ||
71aaa3f7 PP |
688 | # Match prefix |
689 | if self._try_parse_pat(self._bin_byte_prefix_pat) is None: | |
690 | # No match | |
691 | return | |
692 | ||
693 | # Expect eight bits | |
694 | bits = [] # type: List[str] | |
695 | ||
696 | for _ in range(8): | |
697 | self._skip_ws_and_comments() | |
698 | m = self._expect_pat(self._bin_byte_bit_pat, "Expecting a bit (`0` or `1`)") | |
699 | bits.append(m.group(0)) | |
700 | ||
701 | # Return item | |
0e8e3169 | 702 | return _Byte(int("".join(bits), 2), begin_text_loc) |
71aaa3f7 PP |
703 | |
704 | # Patterns for _try_parse_dec_byte() | |
320644e2 | 705 | _dec_byte_prefix_pat = re.compile(r"\$") |
71aaa3f7 PP |
706 | _dec_byte_val_pat = re.compile(r"(?P<neg>-?)(?P<val>\d+)") |
707 | ||
708 | # Tries to parse a decimal byte, returning a byte item on success. | |
709 | def _try_parse_dec_byte(self): | |
0e8e3169 PP |
710 | begin_text_loc = self._text_loc |
711 | ||
71aaa3f7 PP |
712 | # Match prefix |
713 | if self._try_parse_pat(self._dec_byte_prefix_pat) is None: | |
714 | # No match | |
715 | return | |
716 | ||
717 | # Expect the value | |
320644e2 | 718 | self._skip_ws() |
71aaa3f7 PP |
719 | m = self._expect_pat(self._dec_byte_val_pat, "Expecting a decimal constant") |
720 | ||
721 | # Compute value | |
722 | val = int(m.group("val")) * (-1 if m.group("neg") == "-" else 1) | |
723 | ||
724 | # Validate | |
725 | if val < -128 or val > 255: | |
0e8e3169 | 726 | _raise_error("Invalid decimal byte value {}".format(val), begin_text_loc) |
71aaa3f7 PP |
727 | |
728 | # Two's complement | |
05f81895 | 729 | val %= 256 |
71aaa3f7 PP |
730 | |
731 | # Return item | |
0e8e3169 | 732 | return _Byte(val, begin_text_loc) |
71aaa3f7 PP |
733 | |
734 | # Tries to parse a byte, returning a byte item on success. | |
735 | def _try_parse_byte(self): | |
736 | # Hexadecimal | |
737 | item = self._try_parse_hex_byte() | |
738 | ||
739 | if item is not None: | |
740 | return item | |
741 | ||
742 | # Binary | |
743 | item = self._try_parse_bin_byte() | |
744 | ||
745 | if item is not None: | |
746 | return item | |
747 | ||
748 | # Decimal | |
749 | item = self._try_parse_dec_byte() | |
750 | ||
751 | if item is not None: | |
752 | return item | |
753 | ||
754 | # Patterns for _try_parse_str() | |
755 | _str_prefix_pat = re.compile(r'(?:u(?P<len>16|32)(?P<bo>be|le))?\s*"') | |
756 | _str_suffix_pat = re.compile(r'"') | |
757 | _str_str_pat = re.compile(r'(?:(?:\\.)|[^"])*') | |
758 | ||
759 | # Strings corresponding to escape sequence characters | |
760 | _str_escape_seq_strs = { | |
761 | "0": "\0", | |
762 | "a": "\a", | |
763 | "b": "\b", | |
764 | "e": "\x1b", | |
765 | "f": "\f", | |
766 | "n": "\n", | |
767 | "r": "\r", | |
768 | "t": "\t", | |
769 | "v": "\v", | |
770 | "\\": "\\", | |
771 | '"': '"', | |
772 | } | |
773 | ||
774 | # Tries to parse a string, returning a string item on success. | |
775 | def _try_parse_str(self): | |
0e8e3169 PP |
776 | begin_text_loc = self._text_loc |
777 | ||
71aaa3f7 PP |
778 | # Match prefix |
779 | m = self._try_parse_pat(self._str_prefix_pat) | |
780 | ||
781 | if m is None: | |
782 | # No match | |
783 | return | |
784 | ||
785 | # Get encoding | |
786 | encoding = "utf8" | |
787 | ||
788 | if m.group("len") is not None: | |
789 | encoding = "utf_{}_{}".format(m.group("len"), m.group("bo")) | |
790 | ||
791 | # Actual string | |
792 | m = self._expect_pat(self._str_str_pat, "Expecting a literal string") | |
793 | ||
794 | # Expect end of string | |
795 | self._expect_pat(self._str_suffix_pat, 'Expecting `"` (end of literal string)') | |
796 | ||
797 | # Replace escape sequences | |
798 | val = m.group(0) | |
799 | ||
800 | for ec in '0abefnrtv"\\': | |
801 | val = val.replace(r"\{}".format(ec), self._str_escape_seq_strs[ec]) | |
802 | ||
803 | # Encode | |
804 | data = val.encode(encoding) | |
805 | ||
806 | # Return item | |
0e8e3169 | 807 | return _Str(data, begin_text_loc) |
71aaa3f7 | 808 | |
320644e2 PP |
809 | # Common right parenthesis pattern |
810 | _right_paren_pat = re.compile(r"\)") | |
811 | ||
71aaa3f7 | 812 | # Patterns for _try_parse_group() |
320644e2 | 813 | _group_prefix_pat = re.compile(r"\(|!g(?:roup)?\b") |
71aaa3f7 PP |
814 | |
815 | # Tries to parse a group, returning a group item on success. | |
816 | def _try_parse_group(self): | |
0e8e3169 PP |
817 | begin_text_loc = self._text_loc |
818 | ||
71aaa3f7 | 819 | # Match prefix |
261c5ecf PP |
820 | m_open = self._try_parse_pat(self._group_prefix_pat) |
821 | ||
822 | if m_open is None: | |
71aaa3f7 PP |
823 | # No match |
824 | return | |
825 | ||
826 | # Parse items | |
827 | items = self._parse_items() | |
828 | ||
829 | # Expect end of group | |
830 | self._skip_ws_and_comments() | |
261c5ecf PP |
831 | |
832 | if m_open.group(0) == "(": | |
320644e2 | 833 | pat = self._right_paren_pat |
261c5ecf PP |
834 | exp = ")" |
835 | else: | |
836 | pat = self._block_end_pat | |
837 | exp = "!end" | |
838 | ||
839 | self._expect_pat(pat, "Expecting an item or `{}` (end of group)".format(exp)) | |
71aaa3f7 PP |
840 | |
841 | # Return item | |
0e8e3169 | 842 | return _Group(items, begin_text_loc) |
71aaa3f7 PP |
843 | |
844 | # Returns a stripped expression string and an AST expression node | |
845 | # from the expression string `expr_str` at text location `text_loc`. | |
e57a18e1 | 846 | def _ast_expr_from_str(self, expr_str: str, text_loc: TextLocation): |
71aaa3f7 PP |
847 | # Create an expression node from the expression string |
848 | expr_str = expr_str.strip().replace("\n", " ") | |
849 | ||
850 | try: | |
851 | expr = ast.parse(expr_str, mode="eval") | |
852 | except SyntaxError: | |
853 | _raise_error( | |
854 | "Invalid expression `{}`: invalid syntax".format(expr_str), | |
855 | text_loc, | |
856 | ) | |
857 | ||
858 | return expr_str, expr | |
859 | ||
269f6eb3 | 860 | # Patterns for _try_parse_num_and_attr() |
05f81895 | 861 | _val_expr_pat = re.compile(r"([^}:]+):\s*") |
269f6eb3 | 862 | _fl_num_len_attr_pat = re.compile(r"8|16|24|32|40|48|56|64") |
05f81895 | 863 | _leb128_int_attr_pat = re.compile(r"(u|s)leb128") |
71aaa3f7 | 864 | |
05f81895 PP |
865 | # Tries to parse a value and attribute (fixed length in bits or |
866 | # `leb128`), returning a value item on success. | |
269f6eb3 | 867 | def _try_parse_num_and_attr(self): |
71aaa3f7 PP |
868 | begin_text_loc = self._text_loc |
869 | ||
870 | # Match | |
871 | m_expr = self._try_parse_pat(self._val_expr_pat) | |
872 | ||
873 | if m_expr is None: | |
874 | # No match | |
875 | return | |
876 | ||
71aaa3f7 PP |
877 | # Create an expression node from the expression string |
878 | expr_str, expr = self._ast_expr_from_str(m_expr.group(1), begin_text_loc) | |
879 | ||
05f81895 | 880 | # Length? |
269f6eb3 | 881 | m_attr = self._try_parse_pat(self._fl_num_len_attr_pat) |
05f81895 PP |
882 | |
883 | if m_attr is None: | |
884 | # LEB128? | |
885 | m_attr = self._try_parse_pat(self._leb128_int_attr_pat) | |
886 | ||
887 | if m_attr is None: | |
888 | # At this point it's invalid | |
889 | self._raise_error( | |
890 | "Expecting a length (multiple of eight bits), `uleb128`, or `sleb128`" | |
891 | ) | |
892 | ||
893 | # Return LEB128 integer item | |
894 | cls = _ULeb128Int if m_attr.group(1) == "u" else _SLeb128Int | |
895 | return cls(expr_str, expr, begin_text_loc) | |
896 | else: | |
269f6eb3 PP |
897 | # Return fixed-length number item |
898 | return _FlNum( | |
05f81895 PP |
899 | expr_str, |
900 | expr, | |
901 | int(m_attr.group(0)), | |
902 | begin_text_loc, | |
903 | ) | |
71aaa3f7 | 904 | |
320644e2 PP |
905 | # Patterns for _try_parse_var_assign() |
906 | _var_assign_name_equal_pat = re.compile(r"({})\s*=".format(_py_name_pat.pattern)) | |
907 | _var_assign_expr_pat = re.compile(r"[^}]+") | |
71aaa3f7 | 908 | |
2adf4336 PP |
909 | # Tries to parse a variable assignment, returning a variable |
910 | # assignment item on success. | |
911 | def _try_parse_var_assign(self): | |
71aaa3f7 PP |
912 | begin_text_loc = self._text_loc |
913 | ||
914 | # Match | |
320644e2 | 915 | m = self._try_parse_pat(self._var_assign_name_equal_pat) |
71aaa3f7 PP |
916 | |
917 | if m is None: | |
918 | # No match | |
919 | return | |
920 | ||
921 | # Validate name | |
320644e2 | 922 | name = m.group(1) |
71aaa3f7 PP |
923 | |
924 | if name == _icitte_name: | |
0e8e3169 PP |
925 | _raise_error( |
926 | "`{}` is a reserved variable name".format(_icitte_name), begin_text_loc | |
927 | ) | |
71aaa3f7 PP |
928 | |
929 | if name in self._label_names: | |
0e8e3169 | 930 | _raise_error("Existing label named `{}`".format(name), begin_text_loc) |
71aaa3f7 | 931 | |
320644e2 PP |
932 | # Expect an expression |
933 | self._skip_ws() | |
934 | m = self._expect_pat(self._var_assign_expr_pat, "Expecting an expression") | |
71aaa3f7 PP |
935 | |
936 | # Create an expression node from the expression string | |
320644e2 PP |
937 | expr_str, expr = self._ast_expr_from_str(m.group(0), begin_text_loc) |
938 | ||
939 | # Add to known variable names | |
940 | self._var_names.add(name) | |
71aaa3f7 PP |
941 | |
942 | # Return item | |
2adf4336 | 943 | return _VarAssign( |
71aaa3f7 PP |
944 | name, |
945 | expr_str, | |
946 | expr, | |
0e8e3169 | 947 | begin_text_loc, |
71aaa3f7 PP |
948 | ) |
949 | ||
2adf4336 | 950 | # Pattern for _try_parse_set_bo() |
71aaa3f7 PP |
951 | _bo_pat = re.compile(r"[bl]e") |
952 | ||
2adf4336 PP |
953 | # Tries to parse a byte order name, returning a byte order setting |
954 | # item on success. | |
955 | def _try_parse_set_bo(self): | |
0e8e3169 PP |
956 | begin_text_loc = self._text_loc |
957 | ||
71aaa3f7 PP |
958 | # Match |
959 | m = self._try_parse_pat(self._bo_pat) | |
960 | ||
961 | if m is None: | |
962 | # No match | |
963 | return | |
964 | ||
965 | # Return corresponding item | |
966 | if m.group(0) == "be": | |
2adf4336 | 967 | return _SetBo(ByteOrder.BE, begin_text_loc) |
71aaa3f7 PP |
968 | else: |
969 | assert m.group(0) == "le" | |
2adf4336 | 970 | return _SetBo(ByteOrder.LE, begin_text_loc) |
71aaa3f7 PP |
971 | |
972 | # Patterns for _try_parse_val_or_bo() | |
320644e2 PP |
973 | _val_var_assign_set_bo_prefix_pat = re.compile(r"\{") |
974 | _val_var_assign_set_bo_suffix_pat = re.compile(r"\}") | |
71aaa3f7 | 975 | |
2adf4336 PP |
976 | # Tries to parse a value, a variable assignment, or a byte order |
977 | # setting, returning an item on success. | |
978 | def _try_parse_val_or_var_assign_or_set_bo(self): | |
71aaa3f7 | 979 | # Match prefix |
2adf4336 | 980 | if self._try_parse_pat(self._val_var_assign_set_bo_prefix_pat) is None: |
71aaa3f7 PP |
981 | # No match |
982 | return | |
983 | ||
320644e2 PP |
984 | self._skip_ws() |
985 | ||
2adf4336 PP |
986 | # Variable assignment item? |
987 | item = self._try_parse_var_assign() | |
71aaa3f7 PP |
988 | |
989 | if item is None: | |
269f6eb3 PP |
990 | # Number item? |
991 | item = self._try_parse_num_and_attr() | |
71aaa3f7 PP |
992 | |
993 | if item is None: | |
2adf4336 PP |
994 | # Byte order setting item? |
995 | item = self._try_parse_set_bo() | |
71aaa3f7 PP |
996 | |
997 | if item is None: | |
998 | # At this point it's invalid | |
2adf4336 | 999 | self._raise_error( |
269f6eb3 | 1000 | "Expecting a fixed-length number, a variable assignment, or a byte order setting" |
2adf4336 | 1001 | ) |
71aaa3f7 PP |
1002 | |
1003 | # Expect suffix | |
320644e2 | 1004 | self._skip_ws() |
2adf4336 | 1005 | self._expect_pat(self._val_var_assign_set_bo_suffix_pat, "Expecting `}`") |
71aaa3f7 PP |
1006 | return item |
1007 | ||
fc21bb27 PP |
1008 | # Returns a normalized version (so as to be parseable by int()) of |
1009 | # the constant integer string `s`, possibly negative, dealing with | |
1010 | # any radix suffix. | |
1011 | @staticmethod | |
1012 | def _norm_const_int(s: str): | |
1013 | neg = "" | |
1014 | pos = s | |
1015 | ||
1016 | if s.startswith("-"): | |
1017 | neg = "-" | |
1018 | pos = s[1:] | |
1019 | ||
1020 | for r in "xXoObB": | |
1021 | if pos.startswith("0" + r): | |
1022 | # Already correct | |
1023 | return s | |
1024 | ||
1025 | # Try suffix | |
1026 | asm_suf_base = { | |
1027 | "h": "x", | |
1028 | "H": "x", | |
1029 | "q": "o", | |
1030 | "Q": "o", | |
1031 | "o": "o", | |
1032 | "O": "o", | |
1033 | "b": "b", | |
1034 | "B": "B", | |
1035 | } | |
1036 | ||
1037 | for suf in asm_suf_base: | |
1038 | if pos[-1] == suf: | |
1039 | s = "{}0{}{}".format(neg, asm_suf_base[suf], pos.rstrip(suf)) | |
1040 | ||
1041 | return s | |
1042 | ||
320644e2 | 1043 | # Common constant integer patterns |
fc21bb27 PP |
1044 | _pos_const_int_pat = re.compile( |
1045 | 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+" | |
1046 | ) | |
320644e2 | 1047 | _const_int_pat = re.compile(r"(?P<neg>-)?(?:{})".format(_pos_const_int_pat.pattern)) |
71aaa3f7 | 1048 | |
2adf4336 PP |
1049 | # Tries to parse an offset setting value (after the initial `<`), |
1050 | # returning an offset item on success. | |
1051 | def _try_parse_set_offset_val(self): | |
0e8e3169 PP |
1052 | begin_text_loc = self._text_loc |
1053 | ||
71aaa3f7 PP |
1054 | # Match |
1055 | m = self._try_parse_pat(self._pos_const_int_pat) | |
1056 | ||
1057 | if m is None: | |
1058 | # No match | |
1059 | return | |
1060 | ||
1061 | # Return item | |
fc21bb27 | 1062 | return _SetOffset(int(self._norm_const_int(m.group(0)), 0), begin_text_loc) |
71aaa3f7 PP |
1063 | |
1064 | # Tries to parse a label name (after the initial `<`), returning a | |
1065 | # label item on success. | |
1066 | def _try_parse_label_name(self): | |
0e8e3169 PP |
1067 | begin_text_loc = self._text_loc |
1068 | ||
71aaa3f7 PP |
1069 | # Match |
1070 | m = self._try_parse_pat(_py_name_pat) | |
1071 | ||
1072 | if m is None: | |
1073 | # No match | |
1074 | return | |
1075 | ||
1076 | # Validate | |
1077 | name = m.group(0) | |
1078 | ||
1079 | if name == _icitte_name: | |
0e8e3169 PP |
1080 | _raise_error( |
1081 | "`{}` is a reserved label name".format(_icitte_name), begin_text_loc | |
1082 | ) | |
71aaa3f7 PP |
1083 | |
1084 | if name in self._label_names: | |
0e8e3169 | 1085 | _raise_error("Duplicate label name `{}`".format(name), begin_text_loc) |
71aaa3f7 PP |
1086 | |
1087 | if name in self._var_names: | |
0e8e3169 | 1088 | _raise_error("Existing variable named `{}`".format(name), begin_text_loc) |
71aaa3f7 PP |
1089 | |
1090 | # Add to known label names | |
1091 | self._label_names.add(name) | |
1092 | ||
1093 | # Return item | |
0e8e3169 | 1094 | return _Label(name, begin_text_loc) |
71aaa3f7 | 1095 | |
2adf4336 | 1096 | # Patterns for _try_parse_label_or_set_offset() |
320644e2 PP |
1097 | _label_set_offset_prefix_pat = re.compile(r"<") |
1098 | _label_set_offset_suffix_pat = re.compile(r">") | |
71aaa3f7 | 1099 | |
2adf4336 PP |
1100 | # Tries to parse a label or an offset setting, returning an item on |
1101 | # success. | |
1102 | def _try_parse_label_or_set_offset(self): | |
71aaa3f7 | 1103 | # Match prefix |
2adf4336 | 1104 | if self._try_parse_pat(self._label_set_offset_prefix_pat) is None: |
71aaa3f7 PP |
1105 | # No match |
1106 | return | |
1107 | ||
2adf4336 | 1108 | # Offset setting item? |
320644e2 | 1109 | self._skip_ws() |
2adf4336 | 1110 | item = self._try_parse_set_offset_val() |
71aaa3f7 PP |
1111 | |
1112 | if item is None: | |
1113 | # Label item? | |
1114 | item = self._try_parse_label_name() | |
1115 | ||
1116 | if item is None: | |
1117 | # At this point it's invalid | |
2adf4336 | 1118 | self._raise_error("Expecting a label name or an offset setting value") |
71aaa3f7 PP |
1119 | |
1120 | # Expect suffix | |
320644e2 | 1121 | self._skip_ws() |
2adf4336 | 1122 | self._expect_pat(self._label_set_offset_suffix_pat, "Expecting `>`") |
71aaa3f7 PP |
1123 | return item |
1124 | ||
25ca454b PP |
1125 | # Pattern for _parse_pad_val() |
1126 | _pad_val_prefix_pat = re.compile(r"~") | |
1127 | ||
1128 | # Tries to parse a padding value, returning the padding value, or 0 | |
1129 | # if none. | |
1130 | def _parse_pad_val(self): | |
1131 | # Padding value? | |
1132 | self._skip_ws() | |
1133 | pad_val = 0 | |
1134 | ||
1135 | if self._try_parse_pat(self._pad_val_prefix_pat) is not None: | |
1136 | self._skip_ws() | |
1137 | pad_val_text_loc = self._text_loc | |
1138 | m = self._expect_pat( | |
1139 | self._pos_const_int_pat, | |
1140 | "Expecting a positive constant integer (byte value)", | |
1141 | ) | |
1142 | ||
1143 | # Validate | |
fc21bb27 | 1144 | pad_val = int(self._norm_const_int(m.group(0)), 0) |
25ca454b PP |
1145 | |
1146 | if pad_val > 255: | |
1147 | _raise_error( | |
1148 | "Invalid padding byte value {}".format(pad_val), | |
1149 | pad_val_text_loc, | |
1150 | ) | |
1151 | ||
1152 | return pad_val | |
1153 | ||
676f6189 | 1154 | # Patterns for _try_parse_align_offset() |
320644e2 PP |
1155 | _align_offset_prefix_pat = re.compile(r"@") |
1156 | _align_offset_val_pat = re.compile(r"\d+") | |
676f6189 PP |
1157 | |
1158 | # Tries to parse an offset alignment, returning an offset alignment | |
1159 | # item on success. | |
1160 | def _try_parse_align_offset(self): | |
1161 | begin_text_loc = self._text_loc | |
1162 | ||
1163 | # Match prefix | |
1164 | if self._try_parse_pat(self._align_offset_prefix_pat) is None: | |
1165 | # No match | |
1166 | return | |
1167 | ||
320644e2 | 1168 | # Expect an alignment |
25ca454b | 1169 | self._skip_ws() |
676f6189 PP |
1170 | align_text_loc = self._text_loc |
1171 | m = self._expect_pat( | |
1172 | self._align_offset_val_pat, | |
1173 | "Expecting an alignment (positive multiple of eight bits)", | |
1174 | ) | |
1175 | ||
1176 | # Validate alignment | |
320644e2 | 1177 | val = int(m.group(0)) |
676f6189 PP |
1178 | |
1179 | if val <= 0 or (val % 8) != 0: | |
1180 | _raise_error( | |
1181 | "Invalid alignment value {} (not a positive multiple of eight)".format( | |
1182 | val | |
1183 | ), | |
1184 | align_text_loc, | |
1185 | ) | |
1186 | ||
25ca454b PP |
1187 | # Padding value |
1188 | pad_val = self._parse_pad_val() | |
676f6189 | 1189 | |
25ca454b PP |
1190 | # Return item |
1191 | return _AlignOffset(val, pad_val, begin_text_loc) | |
676f6189 | 1192 | |
25ca454b PP |
1193 | # Patterns for _try_parse_fill_until() |
1194 | _fill_until_prefix_pat = re.compile(r"\+") | |
1195 | _fill_until_pad_val_prefix_pat = re.compile(r"~") | |
676f6189 | 1196 | |
25ca454b PP |
1197 | # Tries to parse a filling, returning a filling item on success. |
1198 | def _try_parse_fill_until(self): | |
1199 | begin_text_loc = self._text_loc | |
1200 | ||
1201 | # Match prefix | |
1202 | if self._try_parse_pat(self._fill_until_prefix_pat) is None: | |
1203 | # No match | |
1204 | return | |
1205 | ||
1206 | # Expect expression | |
1207 | self._skip_ws() | |
1208 | expr_str, expr = self._expect_const_int_name_expr(True) | |
1209 | ||
1210 | # Padding value | |
1211 | pad_val = self._parse_pad_val() | |
676f6189 PP |
1212 | |
1213 | # Return item | |
25ca454b | 1214 | return _FillUntil(expr_str, expr, pad_val, begin_text_loc) |
676f6189 | 1215 | |
e57a18e1 | 1216 | # Patterns for _expect_rep_mul_expr() |
320644e2 PP |
1217 | _inner_expr_prefix_pat = re.compile(r"\{") |
1218 | _inner_expr_pat = re.compile(r"[^}]+") | |
1219 | _inner_expr_suffix_pat = re.compile(r"\}") | |
1220 | ||
1221 | # Parses a constant integer if `accept_const_int` is `True` | |
1222 | # (possibly negative if `allow_neg` is `True`), a name, or an | |
1223 | # expression within `{` and `}`. | |
1224 | def _expect_const_int_name_expr( | |
1225 | self, accept_const_int: bool, allow_neg: bool = False | |
1226 | ): | |
e57a18e1 PP |
1227 | expr_text_loc = self._text_loc |
1228 | ||
1229 | # Constant integer? | |
27d52a19 PP |
1230 | m = None |
1231 | ||
320644e2 PP |
1232 | if accept_const_int: |
1233 | m = self._try_parse_pat(self._const_int_pat) | |
e57a18e1 PP |
1234 | |
1235 | if m is None: | |
1236 | # Name? | |
1237 | m = self._try_parse_pat(_py_name_pat) | |
1238 | ||
1239 | if m is None: | |
1240 | # Expression? | |
320644e2 PP |
1241 | if self._try_parse_pat(self._inner_expr_prefix_pat) is None: |
1242 | pos_msg = "" if allow_neg else "positive " | |
1243 | ||
1244 | if accept_const_int: | |
1245 | mid_msg = "a {}constant integer, a name, or `{{`".format( | |
1246 | pos_msg | |
1247 | ) | |
27d52a19 PP |
1248 | else: |
1249 | mid_msg = "a name or `{`" | |
1250 | ||
e57a18e1 | 1251 | # At this point it's invalid |
27d52a19 | 1252 | self._raise_error("Expecting {}".format(mid_msg)) |
e57a18e1 PP |
1253 | |
1254 | # Expect an expression | |
320644e2 | 1255 | self._skip_ws() |
e57a18e1 | 1256 | expr_text_loc = self._text_loc |
320644e2 | 1257 | m = self._expect_pat(self._inner_expr_pat, "Expecting an expression") |
e57a18e1 PP |
1258 | expr_str = m.group(0) |
1259 | ||
1260 | # Expect `}` | |
320644e2 PP |
1261 | self._skip_ws() |
1262 | self._expect_pat(self._inner_expr_suffix_pat, "Expecting `}`") | |
e57a18e1 PP |
1263 | else: |
1264 | expr_str = m.group(0) | |
1265 | else: | |
320644e2 PP |
1266 | if m.group("neg") == "-" and not allow_neg: |
1267 | _raise_error("Expecting a positive constant integer", expr_text_loc) | |
1268 | ||
fc21bb27 | 1269 | expr_str = self._norm_const_int(m.group(0)) |
e57a18e1 PP |
1270 | |
1271 | return self._ast_expr_from_str(expr_str, expr_text_loc) | |
1272 | ||
27d52a19 PP |
1273 | # Parses the multiplier expression of a repetition (block or |
1274 | # post-item) and returns the expression string and AST node. | |
1275 | def _expect_rep_mul_expr(self): | |
320644e2 | 1276 | return self._expect_const_int_name_expr(True) |
27d52a19 PP |
1277 | |
1278 | # Common block end pattern | |
320644e2 | 1279 | _block_end_pat = re.compile(r"!end\b") |
27d52a19 | 1280 | |
e57a18e1 | 1281 | # Pattern for _try_parse_rep_block() |
320644e2 | 1282 | _rep_block_prefix_pat = re.compile(r"!r(?:epeat)?\b") |
e57a18e1 PP |
1283 | |
1284 | # Tries to parse a repetition block, returning a repetition item on | |
1285 | # success. | |
1286 | def _try_parse_rep_block(self): | |
1287 | begin_text_loc = self._text_loc | |
1288 | ||
1289 | # Match prefix | |
1290 | if self._try_parse_pat(self._rep_block_prefix_pat) is None: | |
1291 | # No match | |
1292 | return | |
1293 | ||
1294 | # Expect expression | |
1295 | self._skip_ws_and_comments() | |
1296 | expr_str, expr = self._expect_rep_mul_expr() | |
1297 | ||
1298 | # Parse items | |
1299 | self._skip_ws_and_comments() | |
1300 | items_text_loc = self._text_loc | |
1301 | items = self._parse_items() | |
1302 | ||
1303 | # Expect end of block | |
1304 | self._skip_ws_and_comments() | |
1305 | self._expect_pat( | |
27d52a19 | 1306 | self._block_end_pat, "Expecting an item or `!end` (end of repetition block)" |
e57a18e1 PP |
1307 | ) |
1308 | ||
1309 | # Return item | |
1310 | return _Rep(_Group(items, items_text_loc), expr_str, expr, begin_text_loc) | |
1311 | ||
27d52a19 | 1312 | # Pattern for _try_parse_cond_block() |
320644e2 | 1313 | _cond_block_prefix_pat = re.compile(r"!if\b") |
12b5dbc0 | 1314 | _cond_block_else_pat = re.compile(r"!else\b") |
27d52a19 PP |
1315 | |
1316 | # Tries to parse a conditional block, returning a conditional item | |
1317 | # on success. | |
1318 | def _try_parse_cond_block(self): | |
1319 | begin_text_loc = self._text_loc | |
1320 | ||
1321 | # Match prefix | |
1322 | if self._try_parse_pat(self._cond_block_prefix_pat) is None: | |
1323 | # No match | |
1324 | return | |
1325 | ||
1326 | # Expect expression | |
1327 | self._skip_ws_and_comments() | |
320644e2 | 1328 | expr_str, expr = self._expect_const_int_name_expr(False) |
27d52a19 | 1329 | |
12b5dbc0 | 1330 | # Parse "true" items |
27d52a19 | 1331 | self._skip_ws_and_comments() |
12b5dbc0 PP |
1332 | true_items_text_loc = self._text_loc |
1333 | true_items = self._parse_items() | |
1334 | false_items = [] # type: List[_Item] | |
1335 | false_items_text_loc = begin_text_loc | |
27d52a19 | 1336 | |
12b5dbc0 | 1337 | # `!else`? |
27d52a19 | 1338 | self._skip_ws_and_comments() |
12b5dbc0 PP |
1339 | |
1340 | if self._try_parse_pat(self._cond_block_else_pat) is not None: | |
1341 | # Parse "false" items | |
1342 | self._skip_ws_and_comments() | |
1343 | false_items_text_loc = self._text_loc | |
1344 | false_items = self._parse_items() | |
1345 | ||
1346 | # Expect end of block | |
27d52a19 PP |
1347 | self._expect_pat( |
1348 | self._block_end_pat, | |
12b5dbc0 | 1349 | "Expecting an item, `!else`, or `!end` (end of conditional block)", |
27d52a19 PP |
1350 | ) |
1351 | ||
1352 | # Return item | |
12b5dbc0 PP |
1353 | return _Cond( |
1354 | _Group(true_items, true_items_text_loc), | |
1355 | _Group(false_items, false_items_text_loc), | |
1356 | expr_str, | |
1357 | expr, | |
1358 | begin_text_loc, | |
1359 | ) | |
27d52a19 | 1360 | |
320644e2 PP |
1361 | # Common left parenthesis pattern |
1362 | _left_paren_pat = re.compile(r"\(") | |
1363 | ||
1364 | # Patterns for _try_parse_macro_def() and _try_parse_macro_exp() | |
1365 | _macro_params_comma_pat = re.compile(",") | |
1366 | ||
1367 | # Patterns for _try_parse_macro_def() | |
1368 | _macro_def_prefix_pat = re.compile(r"!m(?:acro)?\b") | |
1369 | ||
1370 | # Tries to parse a macro definition, adding it to `self._macro_defs` | |
1371 | # and returning `True` on success. | |
1372 | def _try_parse_macro_def(self): | |
1373 | begin_text_loc = self._text_loc | |
1374 | ||
1375 | # Match prefix | |
1376 | if self._try_parse_pat(self._macro_def_prefix_pat) is None: | |
1377 | # No match | |
1378 | return False | |
1379 | ||
1380 | # Expect a name | |
1381 | self._skip_ws() | |
1382 | name_text_loc = self._text_loc | |
1383 | m = self._expect_pat(_py_name_pat, "Expecting a valid macro name") | |
1384 | ||
1385 | # Validate name | |
1386 | name = m.group(0) | |
1387 | ||
1388 | if name in self._macro_defs: | |
1389 | _raise_error("Duplicate macro named `{}`".format(name), name_text_loc) | |
1390 | ||
1391 | # Expect `(` | |
1392 | self._skip_ws() | |
1393 | self._expect_pat(self._left_paren_pat, "Expecting `(`") | |
1394 | ||
1395 | # Try to parse comma-separated parameter names | |
1396 | param_names = [] # type: List[str] | |
1397 | expect_comma = False | |
1398 | ||
1399 | while True: | |
1400 | self._skip_ws() | |
1401 | ||
1402 | # End? | |
1403 | if self._try_parse_pat(self._right_paren_pat) is not None: | |
1404 | # End | |
1405 | break | |
1406 | ||
1407 | # Comma? | |
1408 | if expect_comma: | |
1409 | self._expect_pat(self._macro_params_comma_pat, "Expecting `,`") | |
1410 | ||
1411 | # Expect parameter name | |
1412 | self._skip_ws() | |
1413 | param_text_loc = self._text_loc | |
1414 | m = self._expect_pat(_py_name_pat, "Expecting valid parameter name") | |
1415 | ||
1416 | if m.group(0) in param_names: | |
1417 | _raise_error( | |
1418 | "Duplicate macro parameter named `{}`".format(m.group(0)), | |
1419 | param_text_loc, | |
1420 | ) | |
1421 | ||
1422 | param_names.append(m.group(0)) | |
1423 | expect_comma = True | |
1424 | ||
1425 | # Expect items | |
1426 | self._skip_ws_and_comments() | |
1427 | items_text_loc = self._text_loc | |
1428 | old_var_names = self._var_names.copy() | |
1429 | old_label_names = self._label_names.copy() | |
1430 | self._var_names = set() # type: Set[str] | |
1431 | self._label_names = set() # type: Set[str] | |
1432 | items = self._parse_items() | |
1433 | self._var_names = old_var_names | |
1434 | self._label_names = old_label_names | |
1435 | ||
1436 | # Expect suffix | |
1437 | self._expect_pat( | |
1438 | self._block_end_pat, "Expecting an item or `!end` (end of macro block)" | |
1439 | ) | |
1440 | ||
1441 | # Register macro | |
1442 | self._macro_defs[name] = _MacroDef( | |
1443 | name, param_names, _Group(items, items_text_loc), begin_text_loc | |
1444 | ) | |
1445 | ||
1446 | return True | |
1447 | ||
1448 | # Patterns for _try_parse_macro_exp() | |
1449 | _macro_exp_prefix_pat = re.compile(r"m\b") | |
1450 | _macro_exp_colon_pat = re.compile(r":") | |
1451 | ||
1452 | # Tries to parse a macro expansion, returning a macro expansion item | |
1453 | # on success. | |
1454 | def _try_parse_macro_exp(self): | |
1455 | begin_text_loc = self._text_loc | |
1456 | ||
1457 | # Match prefix | |
1458 | if self._try_parse_pat(self._macro_exp_prefix_pat) is None: | |
1459 | # No match | |
1460 | return | |
1461 | ||
1462 | # Expect `:` | |
1463 | self._skip_ws() | |
1464 | self._expect_pat(self._macro_exp_colon_pat, "Expecting `:`") | |
1465 | ||
1466 | # Expect a macro name | |
1467 | self._skip_ws() | |
1468 | name_text_loc = self._text_loc | |
1469 | m = self._expect_pat(_py_name_pat, "Expecting a valid macro name") | |
1470 | ||
1471 | # Validate name | |
1472 | name = m.group(0) | |
1473 | macro_def = self._macro_defs.get(name) | |
1474 | ||
1475 | if macro_def is None: | |
1476 | _raise_error("Unknown macro name `{}`".format(name), name_text_loc) | |
1477 | ||
1478 | # Expect `(` | |
1479 | self._skip_ws() | |
1480 | self._expect_pat(self._left_paren_pat, "Expecting `(`") | |
1481 | ||
1482 | # Try to parse comma-separated parameter values | |
1483 | params_text_loc = self._text_loc | |
1484 | params = [] # type: List[_MacroExpParam] | |
1485 | expect_comma = False | |
1486 | ||
1487 | while True: | |
1488 | self._skip_ws() | |
1489 | ||
1490 | # End? | |
1491 | if self._try_parse_pat(self._right_paren_pat) is not None: | |
1492 | # End | |
1493 | break | |
1494 | ||
1495 | # Expect a Value | |
1496 | if expect_comma: | |
1497 | self._expect_pat(self._macro_params_comma_pat, "Expecting `,`") | |
1498 | ||
1499 | self._skip_ws() | |
1500 | param_text_loc = self._text_loc | |
1501 | params.append( | |
1502 | _MacroExpParam( | |
1503 | *self._expect_const_int_name_expr(True, True), param_text_loc | |
1504 | ) | |
1505 | ) | |
1506 | expect_comma = True | |
1507 | ||
1508 | # Validate parameter values | |
1509 | if len(params) != len(macro_def.param_names): | |
1510 | sing_plur = "" if len(params) == 1 else "s" | |
1511 | _raise_error( | |
1512 | "Macro expansion passes {} parameter{} while the definition expects {}".format( | |
1513 | len(params), sing_plur, len(macro_def.param_names) | |
1514 | ), | |
1515 | params_text_loc, | |
1516 | ) | |
1517 | ||
1518 | # Return item | |
1519 | return _MacroExp(name, params, begin_text_loc) | |
1520 | ||
71aaa3f7 PP |
1521 | # Tries to parse a base item (anything except a repetition), |
1522 | # returning it on success. | |
1523 | def _try_parse_base_item(self): | |
1524 | # Byte item? | |
1525 | item = self._try_parse_byte() | |
1526 | ||
1527 | if item is not None: | |
1528 | return item | |
1529 | ||
1530 | # String item? | |
1531 | item = self._try_parse_str() | |
1532 | ||
1533 | if item is not None: | |
1534 | return item | |
1535 | ||
2adf4336 PP |
1536 | # Value, variable assignment, or byte order setting item? |
1537 | item = self._try_parse_val_or_var_assign_or_set_bo() | |
71aaa3f7 PP |
1538 | |
1539 | if item is not None: | |
1540 | return item | |
1541 | ||
2adf4336 PP |
1542 | # Label or offset setting item? |
1543 | item = self._try_parse_label_or_set_offset() | |
71aaa3f7 | 1544 | |
676f6189 PP |
1545 | if item is not None: |
1546 | return item | |
1547 | ||
1548 | # Offset alignment item? | |
1549 | item = self._try_parse_align_offset() | |
1550 | ||
25ca454b PP |
1551 | if item is not None: |
1552 | return item | |
1553 | ||
1554 | # Filling item? | |
1555 | item = self._try_parse_fill_until() | |
1556 | ||
71aaa3f7 PP |
1557 | if item is not None: |
1558 | return item | |
1559 | ||
1560 | # Group item? | |
1561 | item = self._try_parse_group() | |
1562 | ||
1563 | if item is not None: | |
1564 | return item | |
1565 | ||
320644e2 | 1566 | # Repetition block item? |
e57a18e1 | 1567 | item = self._try_parse_rep_block() |
71aaa3f7 | 1568 | |
e57a18e1 PP |
1569 | if item is not None: |
1570 | return item | |
1571 | ||
27d52a19 PP |
1572 | # Conditional block item? |
1573 | item = self._try_parse_cond_block() | |
1574 | ||
1575 | if item is not None: | |
1576 | return item | |
1577 | ||
320644e2 PP |
1578 | # Macro expansion? |
1579 | item = self._try_parse_macro_exp() | |
1580 | ||
1581 | if item is not None: | |
1582 | return item | |
1583 | ||
e57a18e1 PP |
1584 | # Pattern for _try_parse_rep_post() |
1585 | _rep_post_prefix_pat = re.compile(r"\*") | |
1586 | ||
1587 | # Tries to parse a post-item repetition, returning the expression | |
1588 | # string and AST expression node on success. | |
1589 | def _try_parse_rep_post(self): | |
71aaa3f7 | 1590 | # Match prefix |
e57a18e1 | 1591 | if self._try_parse_pat(self._rep_post_prefix_pat) is None: |
71aaa3f7 | 1592 | # No match |
2adf4336 | 1593 | return |
71aaa3f7 | 1594 | |
e57a18e1 | 1595 | # Return expression string and AST expression |
71aaa3f7 | 1596 | self._skip_ws_and_comments() |
e57a18e1 | 1597 | return self._expect_rep_mul_expr() |
71aaa3f7 | 1598 | |
1ca7b5e1 PP |
1599 | # Tries to parse an item, possibly followed by a repetition, |
1600 | # returning `True` on success. | |
1601 | # | |
1602 | # Appends any parsed item to `items`. | |
1603 | def _try_append_item(self, items: List[_Item]): | |
71aaa3f7 PP |
1604 | self._skip_ws_and_comments() |
1605 | ||
320644e2 | 1606 | # Base item |
71aaa3f7 PP |
1607 | item = self._try_parse_base_item() |
1608 | ||
1609 | if item is None: | |
320644e2 | 1610 | return |
71aaa3f7 PP |
1611 | |
1612 | # Parse repetition if the base item is repeatable | |
1613 | if isinstance(item, _RepableItem): | |
0e8e3169 PP |
1614 | self._skip_ws_and_comments() |
1615 | rep_text_loc = self._text_loc | |
e57a18e1 | 1616 | rep_ret = self._try_parse_rep_post() |
71aaa3f7 | 1617 | |
2adf4336 | 1618 | if rep_ret is not None: |
320644e2 | 1619 | item = _Rep(item, *rep_ret, rep_text_loc) |
71aaa3f7 | 1620 | |
1ca7b5e1 PP |
1621 | items.append(item) |
1622 | return True | |
71aaa3f7 PP |
1623 | |
1624 | # Parses and returns items, skipping whitespaces, insignificant | |
1625 | # symbols, and comments when allowed, and stopping at the first | |
1626 | # unknown character. | |
320644e2 PP |
1627 | # |
1628 | # Accepts and registers macro definitions if `accept_macro_defs` | |
1629 | # is `True`. | |
1630 | def _parse_items(self, accept_macro_defs: bool = False) -> List[_Item]: | |
71aaa3f7 PP |
1631 | items = [] # type: List[_Item] |
1632 | ||
1633 | while self._isnt_done(): | |
1ca7b5e1 PP |
1634 | # Try to append item |
1635 | if not self._try_append_item(items): | |
320644e2 PP |
1636 | if accept_macro_defs and self._try_parse_macro_def(): |
1637 | continue | |
1638 | ||
1ca7b5e1 PP |
1639 | # Unknown at this point |
1640 | break | |
71aaa3f7 PP |
1641 | |
1642 | return items | |
1643 | ||
1644 | # Parses the whole Normand input, setting `self._res` to the main | |
1645 | # group item on success. | |
1646 | def _parse(self): | |
1647 | if len(self._normand.strip()) == 0: | |
1648 | # Special case to make sure there's something to consume | |
1649 | self._res = _Group([], self._text_loc) | |
1650 | return | |
1651 | ||
1652 | # Parse first level items | |
320644e2 | 1653 | items = self._parse_items(True) |
71aaa3f7 PP |
1654 | |
1655 | # Make sure there's nothing left | |
1656 | self._skip_ws_and_comments() | |
1657 | ||
1658 | if self._isnt_done(): | |
1659 | self._raise_error( | |
1660 | "Unexpected character `{}`".format(self._normand[self._at]) | |
1661 | ) | |
1662 | ||
1663 | # Set main group item | |
1664 | self._res = _Group(items, self._text_loc) | |
1665 | ||
1666 | ||
1667 | # The return type of parse(). | |
1668 | class ParseResult: | |
1669 | @classmethod | |
1670 | def _create( | |
1671 | cls, | |
1672 | data: bytearray, | |
e57a18e1 PP |
1673 | variables: VariablesT, |
1674 | labels: LabelsT, | |
71aaa3f7 PP |
1675 | offset: int, |
1676 | bo: Optional[ByteOrder], | |
1677 | ): | |
1678 | self = cls.__new__(cls) | |
1679 | self._init(data, variables, labels, offset, bo) | |
1680 | return self | |
1681 | ||
1682 | def __init__(self, *args, **kwargs): # type: ignore | |
1683 | raise NotImplementedError | |
1684 | ||
1685 | def _init( | |
1686 | self, | |
1687 | data: bytearray, | |
e57a18e1 PP |
1688 | variables: VariablesT, |
1689 | labels: LabelsT, | |
71aaa3f7 PP |
1690 | offset: int, |
1691 | bo: Optional[ByteOrder], | |
1692 | ): | |
1693 | self._data = data | |
1694 | self._vars = variables | |
1695 | self._labels = labels | |
1696 | self._offset = offset | |
1697 | self._bo = bo | |
1698 | ||
1699 | # Generated data. | |
1700 | @property | |
1701 | def data(self): | |
1702 | return self._data | |
1703 | ||
1704 | # Dictionary of updated variable names to their last computed value. | |
1705 | @property | |
1706 | def variables(self): | |
1707 | return self._vars | |
1708 | ||
1709 | # Dictionary of updated main group label names to their computed | |
1710 | # value. | |
1711 | @property | |
1712 | def labels(self): | |
1713 | return self._labels | |
1714 | ||
1715 | # Updated offset. | |
1716 | @property | |
1717 | def offset(self): | |
1718 | return self._offset | |
1719 | ||
1720 | # Updated byte order. | |
1721 | @property | |
1722 | def byte_order(self): | |
1723 | return self._bo | |
1724 | ||
1725 | ||
1726 | # Raises a parse error for the item `item`, creating it using the | |
1727 | # message `msg`. | |
1728 | def _raise_error_for_item(msg: str, item: _Item) -> NoReturn: | |
1729 | _raise_error(msg, item.text_loc) | |
1730 | ||
1731 | ||
1732 | # The `ICITTE` reserved name. | |
1733 | _icitte_name = "ICITTE" | |
1734 | ||
1735 | ||
2adf4336 PP |
1736 | # Base node visitor. |
1737 | # | |
1738 | # Calls the _visit_name() method for each name node which isn't the name | |
1739 | # of a call. | |
1740 | class _NodeVisitor(ast.NodeVisitor): | |
1741 | def __init__(self): | |
71aaa3f7 PP |
1742 | self._parent_is_call = False |
1743 | ||
1744 | def generic_visit(self, node: ast.AST): | |
1745 | if type(node) is ast.Call: | |
1746 | self._parent_is_call = True | |
1747 | elif type(node) is ast.Name and not self._parent_is_call: | |
2adf4336 | 1748 | self._visit_name(node.id) |
71aaa3f7 PP |
1749 | |
1750 | super().generic_visit(node) | |
1751 | self._parent_is_call = False | |
1752 | ||
2adf4336 PP |
1753 | @abc.abstractmethod |
1754 | def _visit_name(self, name: str): | |
1755 | ... | |
1756 | ||
71aaa3f7 | 1757 | |
2adf4336 PP |
1758 | # Expression validator: validates that all the names within the |
1759 | # expression are allowed. | |
1760 | class _ExprValidator(_NodeVisitor): | |
320644e2 | 1761 | def __init__(self, expr_str: str, text_loc: TextLocation, allowed_names: Set[str]): |
2adf4336 | 1762 | super().__init__() |
320644e2 PP |
1763 | self._expr_str = expr_str |
1764 | self._text_loc = text_loc | |
2adf4336 | 1765 | self._allowed_names = allowed_names |
2adf4336 PP |
1766 | |
1767 | def _visit_name(self, name: str): | |
1768 | # Make sure the name refers to a known and reachable | |
1769 | # variable/label name. | |
e57a18e1 | 1770 | if name != _icitte_name and name not in self._allowed_names: |
2adf4336 | 1771 | msg = "Illegal (unknown or unreachable) variable/label name `{}` in expression `{}`".format( |
320644e2 | 1772 | name, self._expr_str |
2adf4336 PP |
1773 | ) |
1774 | ||
05f81895 | 1775 | allowed_names = self._allowed_names.copy() |
e57a18e1 | 1776 | allowed_names.add(_icitte_name) |
2adf4336 | 1777 | |
05f81895 | 1778 | if len(allowed_names) > 0: |
2adf4336 PP |
1779 | allowed_names_str = ", ".join( |
1780 | sorted(["`{}`".format(name) for name in allowed_names]) | |
1781 | ) | |
1782 | msg += "; the legal names are {{{}}}".format(allowed_names_str) | |
1783 | ||
1784 | _raise_error( | |
1785 | msg, | |
320644e2 | 1786 | self._text_loc, |
2adf4336 PP |
1787 | ) |
1788 | ||
1789 | ||
2adf4336 PP |
1790 | # Generator state. |
1791 | class _GenState: | |
1792 | def __init__( | |
1b8aa84a | 1793 | self, |
e57a18e1 PP |
1794 | variables: VariablesT, |
1795 | labels: LabelsT, | |
1b8aa84a PP |
1796 | offset: int, |
1797 | bo: Optional[ByteOrder], | |
2adf4336 PP |
1798 | ): |
1799 | self.variables = variables.copy() | |
1800 | self.labels = labels.copy() | |
1801 | self.offset = offset | |
1802 | self.bo = bo | |
71aaa3f7 | 1803 | |
320644e2 PP |
1804 | def __repr__(self): |
1805 | return "_GenState({}, {}, {}, {})".format( | |
1806 | repr(self.variables), repr(self.labels), repr(self.offset), repr(self.bo) | |
1807 | ) | |
1808 | ||
1809 | ||
1810 | # Fixed-length number item instance. | |
1811 | class _FlNumItemInst: | |
1812 | def __init__(self, item: _FlNum, offset_in_data: int, state: _GenState): | |
1813 | self._item = item | |
1814 | self._offset_in_data = offset_in_data | |
1815 | self._state = state | |
1816 | ||
1817 | @property | |
1818 | def item(self): | |
1819 | return self._item | |
1820 | ||
1821 | @property | |
1822 | def offset_in_data(self): | |
1823 | return self._offset_in_data | |
1824 | ||
1825 | @property | |
1826 | def state(self): | |
1827 | return self._state | |
1828 | ||
71aaa3f7 | 1829 | |
2adf4336 | 1830 | # Generator of data and final state from a group item. |
71aaa3f7 PP |
1831 | # |
1832 | # Generation happens in memory at construction time. After building, use | |
1833 | # the `data`, `variables`, `labels`, `offset`, and `bo` properties to | |
1834 | # get the resulting context. | |
2adf4336 PP |
1835 | # |
1836 | # The steps of generation are: | |
1837 | # | |
320644e2 PP |
1838 | # 1. Handle each item in prefix order. |
1839 | # | |
1840 | # The handlers append bytes to `self._data` and update some current | |
1841 | # state object (`_GenState` instance). | |
1842 | # | |
1843 | # When handling a fixed-length number item, try to evaluate its | |
1844 | # expression using the current state. If this fails, then it might be | |
1845 | # because the expression refers to a "future" label: save the current | |
1846 | # offset in `self._data` (generated data) and a snapshot of the | |
1847 | # current state within `self._fl_num_item_insts` (`_FlNumItemInst` | |
1848 | # object). _gen_fl_num_item_insts() will deal with this later. | |
2adf4336 | 1849 | # |
320644e2 PP |
1850 | # When handling the items of a group, keep a map of immediate label |
1851 | # names to their offset. Then, after having processed all the items, | |
1852 | # update the relevant saved state snapshots in | |
1853 | # `self._fl_num_item_insts` with those immediate label values. | |
1854 | # _gen_fl_num_item_insts() will deal with this later. | |
2adf4336 | 1855 | # |
320644e2 PP |
1856 | # 2. Handle all the fixed-length number item instances of which the |
1857 | # expression evaluation failed before. | |
2adf4336 | 1858 | # |
320644e2 PP |
1859 | # At this point, `self._fl_num_item_insts` contains everything that's |
1860 | # needed to evaluate the expressions, including the values of | |
1861 | # "future" labels from the point of view of some fixed-length number | |
1862 | # item instance. | |
2adf4336 | 1863 | # |
320644e2 | 1864 | # If an evaluation fails at this point, then it's a user error. |
71aaa3f7 PP |
1865 | class _Gen: |
1866 | def __init__( | |
1867 | self, | |
1868 | group: _Group, | |
320644e2 | 1869 | macro_defs: _MacroDefsT, |
e57a18e1 PP |
1870 | variables: VariablesT, |
1871 | labels: LabelsT, | |
71aaa3f7 PP |
1872 | offset: int, |
1873 | bo: Optional[ByteOrder], | |
1874 | ): | |
320644e2 PP |
1875 | self._macro_defs = macro_defs |
1876 | self._fl_num_item_insts = [] # type: List[_FlNumItemInst] | |
2adf4336 | 1877 | self._gen(group, _GenState(variables, labels, offset, bo)) |
71aaa3f7 PP |
1878 | |
1879 | # Generated bytes. | |
1880 | @property | |
1881 | def data(self): | |
1882 | return self._data | |
1883 | ||
1884 | # Updated variables. | |
1885 | @property | |
1886 | def variables(self): | |
2adf4336 | 1887 | return self._final_state.variables |
71aaa3f7 PP |
1888 | |
1889 | # Updated main group labels. | |
1890 | @property | |
1891 | def labels(self): | |
2adf4336 | 1892 | return self._final_state.labels |
71aaa3f7 PP |
1893 | |
1894 | # Updated offset. | |
1895 | @property | |
1896 | def offset(self): | |
2adf4336 | 1897 | return self._final_state.offset |
71aaa3f7 PP |
1898 | |
1899 | # Updated byte order. | |
1900 | @property | |
1901 | def bo(self): | |
2adf4336 PP |
1902 | return self._final_state.bo |
1903 | ||
320644e2 PP |
1904 | # Evaluates the expression `expr` of which the original string is |
1905 | # `expr_str` at the location `text_loc` considering the current | |
2adf4336 PP |
1906 | # generation state `state`. |
1907 | # | |
269f6eb3 PP |
1908 | # If `allow_float` is `True`, then the type of the result may be |
1909 | # `float` too. | |
2adf4336 | 1910 | @staticmethod |
320644e2 PP |
1911 | def _eval_expr( |
1912 | expr_str: str, | |
1913 | expr: ast.Expression, | |
1914 | text_loc: TextLocation, | |
269f6eb3 | 1915 | state: _GenState, |
269f6eb3 PP |
1916 | allow_float: bool = False, |
1917 | ): | |
e57a18e1 PP |
1918 | syms = {} # type: VariablesT |
1919 | syms.update(state.labels) | |
71aaa3f7 | 1920 | |
e57a18e1 PP |
1921 | # Set the `ICITTE` name to the current offset |
1922 | syms[_icitte_name] = state.offset | |
71aaa3f7 PP |
1923 | |
1924 | # Add the current variables | |
2adf4336 | 1925 | syms.update(state.variables) |
71aaa3f7 PP |
1926 | |
1927 | # Validate the node and its children | |
320644e2 | 1928 | _ExprValidator(expr_str, text_loc, set(syms.keys())).visit(expr) |
71aaa3f7 PP |
1929 | |
1930 | # Compile and evaluate expression node | |
1931 | try: | |
320644e2 | 1932 | val = eval(compile(expr, "", "eval"), None, syms) |
71aaa3f7 | 1933 | except Exception as exc: |
320644e2 PP |
1934 | _raise_error( |
1935 | "Failed to evaluate expression `{}`: {}".format(expr_str, exc), | |
1936 | text_loc, | |
71aaa3f7 PP |
1937 | ) |
1938 | ||
27d52a19 PP |
1939 | # Convert `bool` result type to `int` to normalize |
1940 | if type(val) is bool: | |
1941 | val = int(val) | |
1942 | ||
269f6eb3 PP |
1943 | # Validate result type |
1944 | expected_types = {int} # type: Set[type] | |
1945 | type_msg = "`int`" | |
1946 | ||
1947 | if allow_float: | |
1948 | expected_types.add(float) | |
1949 | type_msg += " or `float`" | |
1950 | ||
1951 | if type(val) not in expected_types: | |
320644e2 | 1952 | _raise_error( |
269f6eb3 | 1953 | "Invalid expression `{}`: expecting result type {}, not `{}`".format( |
320644e2 | 1954 | expr_str, type_msg, type(val).__name__ |
71aaa3f7 | 1955 | ), |
320644e2 | 1956 | text_loc, |
71aaa3f7 PP |
1957 | ) |
1958 | ||
1959 | return val | |
1960 | ||
320644e2 PP |
1961 | # Evaluates the expression of `item` considering the current |
1962 | # generation state `state`. | |
1963 | # | |
1964 | # If `allow_float` is `True`, then the type of the result may be | |
1965 | # `float` too. | |
1966 | @staticmethod | |
1967 | def _eval_item_expr( | |
25ca454b | 1968 | item: Union[_FlNum, _Leb128Int, _FillUntil, _VarAssign, _Rep, _Cond], |
320644e2 PP |
1969 | state: _GenState, |
1970 | allow_float: bool = False, | |
1971 | ): | |
1972 | return _Gen._eval_expr( | |
1973 | item.expr_str, item.expr, item.text_loc, state, allow_float | |
1974 | ) | |
1975 | ||
1976 | # Handles the byte item `item`. | |
1977 | def _handle_byte_item(self, item: _Byte, state: _GenState): | |
1978 | self._data.append(item.val) | |
1979 | state.offset += item.size | |
1980 | ||
1981 | # Handles the string item `item`. | |
1982 | def _handle_str_item(self, item: _Str, state: _GenState): | |
1983 | self._data += item.data | |
1984 | state.offset += item.size | |
1985 | ||
1986 | # Handles the byte order setting item `item`. | |
1987 | def _handle_set_bo_item(self, item: _SetBo, state: _GenState): | |
1988 | # Update current byte order | |
1989 | state.bo = item.bo | |
1990 | ||
1991 | # Handles the variable assignment item `item`. | |
1992 | def _handle_var_assign_item(self, item: _VarAssign, state: _GenState): | |
1993 | # Update variable | |
1994 | state.variables[item.name] = self._eval_item_expr(item, state, True) | |
1995 | ||
1996 | # Handles the fixed-length number item `item`. | |
1997 | def _handle_fl_num_item(self, item: _FlNum, state: _GenState): | |
1998 | # Validate current byte order | |
1999 | if state.bo is None and item.len > 8: | |
2000 | _raise_error_for_item( | |
2001 | "Current byte order isn't defined at first fixed-length number (`{}`) to encode on more than 8 bits".format( | |
2002 | item.expr_str | |
2003 | ), | |
2004 | item, | |
2005 | ) | |
2006 | ||
2007 | # Try an immediate evaluation. If it fails, then keep everything | |
2008 | # needed to (try to) generate the bytes of this item later. | |
2009 | try: | |
2010 | data = self._gen_fl_num_item_inst_data(item, state) | |
2011 | except Exception: | |
2012 | self._fl_num_item_insts.append( | |
2013 | _FlNumItemInst(item, len(self._data), copy.deepcopy(state)) | |
2014 | ) | |
2015 | ||
2016 | # Reserve space in `self._data` for this instance | |
2017 | data = bytes([0] * (item.len // 8)) | |
2018 | ||
2019 | # Append bytes | |
2020 | self._data += data | |
2021 | ||
2022 | # Update offset | |
2023 | state.offset += len(data) | |
2024 | ||
05f81895 PP |
2025 | # Returns the size, in bytes, required to encode the value `val` |
2026 | # with LEB128 (signed version if `is_signed` is `True`). | |
2027 | @staticmethod | |
2028 | def _leb128_size_for_val(val: int, is_signed: bool): | |
2029 | if val < 0: | |
2030 | # Equivalent upper bound. | |
2031 | # | |
2032 | # For example, if `val` is -128, then the full integer for | |
2033 | # this number of bits would be [-128, 127]. | |
2034 | val = -val - 1 | |
2035 | ||
2036 | # Number of bits (add one for the sign if needed) | |
2037 | bits = val.bit_length() + int(is_signed) | |
2038 | ||
2039 | if bits == 0: | |
2040 | bits = 1 | |
2041 | ||
2042 | # Seven bits per byte | |
2043 | return math.ceil(bits / 7) | |
2044 | ||
320644e2 PP |
2045 | # Handles the LEB128 integer item `item`. |
2046 | def _handle_leb128_int_item(self, item: _Leb128Int, state: _GenState): | |
2047 | # Compute value | |
2048 | val = self._eval_item_expr(item, state, False) | |
676f6189 | 2049 | |
320644e2 PP |
2050 | # Size in bytes |
2051 | size = self._leb128_size_for_val(val, type(item) is _SLeb128Int) | |
05f81895 | 2052 | |
320644e2 PP |
2053 | # For each byte |
2054 | for _ in range(size): | |
2055 | # Seven LSBs, MSB of the byte set (continue) | |
2056 | self._data.append((val & 0x7F) | 0x80) | |
2057 | val >>= 7 | |
2adf4336 | 2058 | |
320644e2 PP |
2059 | # Clear MSB of last byte (stop) |
2060 | self._data[-1] &= ~0x80 | |
2adf4336 | 2061 | |
320644e2 PP |
2062 | # Update offset |
2063 | state.offset += size | |
27d52a19 | 2064 | |
320644e2 PP |
2065 | # Handles the group item `item`, removing the immediate labels from |
2066 | # `state` at the end if `remove_immediate_labels` is `True`. | |
2067 | def _handle_group_item( | |
2068 | self, item: _Group, state: _GenState, remove_immediate_labels: bool = True | |
2069 | ): | |
2070 | first_fl_num_item_inst_index = len(self._fl_num_item_insts) | |
2071 | immediate_labels = {} # type: LabelsT | |
27d52a19 | 2072 | |
320644e2 PP |
2073 | # Handle each item |
2074 | for subitem in item.items: | |
2075 | if type(subitem) is _Label: | |
2076 | # Add to local immediate labels | |
2077 | immediate_labels[subitem.name] = state.offset | |
2adf4336 | 2078 | |
320644e2 | 2079 | self._handle_item(subitem, state) |
2adf4336 | 2080 | |
320644e2 PP |
2081 | # Remove immediate labels from current state if needed |
2082 | if remove_immediate_labels: | |
2083 | for name in immediate_labels: | |
2084 | del state.labels[name] | |
2adf4336 | 2085 | |
320644e2 PP |
2086 | # Add all immediate labels to all state snapshots since |
2087 | # `first_fl_num_item_inst_index`. | |
2088 | for inst in self._fl_num_item_insts[first_fl_num_item_inst_index:]: | |
2089 | inst.state.labels.update(immediate_labels) | |
2adf4336 | 2090 | |
320644e2 PP |
2091 | # Handles the repetition item `item`. |
2092 | def _handle_rep_item(self, item: _Rep, state: _GenState): | |
2093 | # Compute the repetition count | |
2094 | mul = _Gen._eval_item_expr(item, state) | |
05f81895 | 2095 | |
320644e2 PP |
2096 | # Validate result |
2097 | if mul < 0: | |
2098 | _raise_error_for_item( | |
2099 | "Invalid expression `{}`: unexpected negative result {:,}".format( | |
2100 | item.expr_str, mul | |
2101 | ), | |
2102 | item, | |
2103 | ) | |
2adf4336 | 2104 | |
320644e2 PP |
2105 | # Generate item data `mul` times |
2106 | for _ in range(mul): | |
2107 | self._handle_item(item.item, state) | |
2adf4336 | 2108 | |
320644e2 | 2109 | # Handles the conditional item `item`. |
12b5dbc0 | 2110 | def _handle_cond_item(self, item: _Cond, state: _GenState): |
320644e2 PP |
2111 | # Compute the conditional value |
2112 | val = _Gen._eval_item_expr(item, state) | |
2adf4336 | 2113 | |
320644e2 PP |
2114 | # Generate item data if needed |
2115 | if val: | |
12b5dbc0 PP |
2116 | self._handle_item(item.true_item, state) |
2117 | else: | |
2118 | self._handle_item(item.false_item, state) | |
2adf4336 | 2119 | |
320644e2 PP |
2120 | # Evaluates the parameters of the macro expansion item `item` |
2121 | # considering the initial state `init_state` and returns a new state | |
2122 | # to handle the items of the macro. | |
2123 | def _eval_macro_exp_params(self, item: _MacroExp, init_state: _GenState): | |
2124 | # New state | |
2125 | exp_state = _GenState({}, {}, init_state.offset, init_state.bo) | |
2adf4336 | 2126 | |
320644e2 PP |
2127 | # Evaluate the parameter expressions |
2128 | macro_def = self._macro_defs[item.name] | |
2adf4336 | 2129 | |
320644e2 PP |
2130 | for param_name, param in zip(macro_def.param_names, item.params): |
2131 | exp_state.variables[param_name] = _Gen._eval_expr( | |
2132 | param.expr_str, param.expr, param.text_loc, init_state, True | |
2133 | ) | |
2adf4336 | 2134 | |
320644e2 | 2135 | return exp_state |
2adf4336 | 2136 | |
320644e2 PP |
2137 | # Handles the macro expansion item `item`. |
2138 | def _handle_macro_exp_item(self, item: _MacroExp, state: _GenState): | |
2139 | # New state | |
2140 | exp_state = self._eval_macro_exp_params(item, state) | |
27d52a19 | 2141 | |
320644e2 PP |
2142 | # Process the contained group |
2143 | init_data_size = len(self._data) | |
2144 | self._handle_item(self._macro_defs[item.name].group, exp_state) | |
27d52a19 | 2145 | |
320644e2 PP |
2146 | # Update state offset and return |
2147 | state.offset += len(self._data) - init_data_size | |
676f6189 | 2148 | |
320644e2 PP |
2149 | # Handles the offset setting item `item`. |
2150 | def _handle_set_offset_item(self, item: _SetOffset, state: _GenState): | |
676f6189 | 2151 | state.offset = item.val |
2adf4336 | 2152 | |
25ca454b | 2153 | # Handles the offset alignment item `item` (adds padding). |
320644e2 PP |
2154 | def _handle_align_offset_item(self, item: _AlignOffset, state: _GenState): |
2155 | init_offset = state.offset | |
2156 | align_bytes = item.val // 8 | |
2157 | state.offset = (state.offset + align_bytes - 1) // align_bytes * align_bytes | |
2158 | self._data += bytes([item.pad_val] * (state.offset - init_offset)) | |
2adf4336 | 2159 | |
25ca454b PP |
2160 | # Handles the filling item `item` (adds padding). |
2161 | def _handle_fill_until_item(self, item: _FillUntil, state: _GenState): | |
2162 | # Compute the new offset | |
2163 | new_offset = _Gen._eval_item_expr(item, state) | |
2164 | ||
2165 | # Validate the new offset | |
2166 | if new_offset < state.offset: | |
2167 | _raise_error_for_item( | |
2168 | "Invalid expression `{}`: new offset {:,} is less than current offset {:,}".format( | |
2169 | item.expr_str, new_offset, state.offset | |
2170 | ), | |
2171 | item, | |
2172 | ) | |
2173 | ||
2174 | # Fill | |
2175 | self._data += bytes([item.pad_val] * (new_offset - state.offset)) | |
2176 | ||
2177 | # Update offset | |
2178 | state.offset = new_offset | |
2179 | ||
320644e2 PP |
2180 | # Handles the label item `item`. |
2181 | def _handle_label_item(self, item: _Label, state: _GenState): | |
2182 | state.labels[item.name] = state.offset | |
2adf4336 | 2183 | |
320644e2 PP |
2184 | # Handles the item `item`, returning the updated next repetition |
2185 | # instance. | |
2186 | def _handle_item(self, item: _Item, state: _GenState): | |
2187 | return self._item_handlers[type(item)](item, state) | |
71aaa3f7 | 2188 | |
320644e2 PP |
2189 | # Generates the data for a fixed-length integer item instance having |
2190 | # the value `val` and returns it. | |
2191 | def _gen_fl_int_item_inst_data(self, val: int, item: _FlNum, state: _GenState): | |
71aaa3f7 PP |
2192 | # Validate range |
2193 | if val < -(2 ** (item.len - 1)) or val > 2**item.len - 1: | |
2194 | _raise_error_for_item( | |
320644e2 PP |
2195 | "Value {:,} is outside the {}-bit range when evaluating expression `{}`".format( |
2196 | val, item.len, item.expr_str | |
71aaa3f7 PP |
2197 | ), |
2198 | item, | |
2199 | ) | |
2200 | ||
2201 | # Encode result on 64 bits (to extend the sign bit whatever the | |
2202 | # value of `item.len`). | |
71aaa3f7 PP |
2203 | data = struct.pack( |
2204 | "{}{}".format( | |
2adf4336 | 2205 | ">" if state.bo in (None, ByteOrder.BE) else "<", |
71aaa3f7 PP |
2206 | "Q" if val >= 0 else "q", |
2207 | ), | |
2208 | val, | |
2209 | ) | |
2210 | ||
2211 | # Keep only the requested length | |
2212 | len_bytes = item.len // 8 | |
2213 | ||
2adf4336 | 2214 | if state.bo in (None, ByteOrder.BE): |
71aaa3f7 PP |
2215 | # Big endian: keep last bytes |
2216 | data = data[-len_bytes:] | |
2217 | else: | |
2218 | # Little endian: keep first bytes | |
2adf4336 | 2219 | assert state.bo == ByteOrder.LE |
71aaa3f7 PP |
2220 | data = data[:len_bytes] |
2221 | ||
320644e2 PP |
2222 | # Return data |
2223 | return data | |
269f6eb3 | 2224 | |
320644e2 PP |
2225 | # Generates the data for a fixed-length floating point number item |
2226 | # instance having the value `val` and returns it. | |
2227 | def _gen_fl_float_item_inst_data(self, val: float, item: _FlNum, state: _GenState): | |
269f6eb3 PP |
2228 | # Validate length |
2229 | if item.len not in (32, 64): | |
2230 | _raise_error_for_item( | |
2231 | "Invalid {}-bit length for a fixed-length floating point number (value {:,})".format( | |
2232 | item.len, val | |
2233 | ), | |
2234 | item, | |
2235 | ) | |
2236 | ||
320644e2 PP |
2237 | # Encode and return result |
2238 | return struct.pack( | |
269f6eb3 PP |
2239 | "{}{}".format( |
2240 | ">" if state.bo in (None, ByteOrder.BE) else "<", | |
2241 | "f" if item.len == 32 else "d", | |
2242 | ), | |
2243 | val, | |
2244 | ) | |
2245 | ||
320644e2 PP |
2246 | # Generates the data for a fixed-length number item instance and |
2247 | # returns it. | |
2248 | def _gen_fl_num_item_inst_data(self, item: _FlNum, state: _GenState): | |
269f6eb3 | 2249 | # Compute value |
e57a18e1 | 2250 | val = self._eval_item_expr(item, state, True) |
269f6eb3 | 2251 | |
269f6eb3 PP |
2252 | # Handle depending on type |
2253 | if type(val) is int: | |
320644e2 | 2254 | return self._gen_fl_int_item_inst_data(val, item, state) |
269f6eb3 PP |
2255 | else: |
2256 | assert type(val) is float | |
320644e2 | 2257 | return self._gen_fl_float_item_inst_data(val, item, state) |
05f81895 | 2258 | |
320644e2 PP |
2259 | # Generates the data for all the fixed-length number item instances |
2260 | # and writes it at the correct offset within `self._data`. | |
2261 | def _gen_fl_num_item_insts(self): | |
2262 | for inst in self._fl_num_item_insts: | |
2263 | # Generate bytes | |
2264 | data = self._gen_fl_num_item_inst_data(inst.item, inst.state) | |
05f81895 | 2265 | |
320644e2 PP |
2266 | # Insert bytes into `self._data` |
2267 | self._data[inst.offset_in_data : inst.offset_in_data + len(data)] = data | |
2adf4336 PP |
2268 | |
2269 | # Generates the data (`self._data`) and final state | |
2270 | # (`self._final_state`) from `group` and the initial state `state`. | |
2271 | def _gen(self, group: _Group, state: _GenState): | |
71aaa3f7 PP |
2272 | # Initial state |
2273 | self._data = bytearray() | |
71aaa3f7 PP |
2274 | |
2275 | # Item handlers | |
2276 | self._item_handlers = { | |
676f6189 | 2277 | _AlignOffset: self._handle_align_offset_item, |
71aaa3f7 | 2278 | _Byte: self._handle_byte_item, |
27d52a19 | 2279 | _Cond: self._handle_cond_item, |
25ca454b | 2280 | _FillUntil: self._handle_fill_until_item, |
269f6eb3 | 2281 | _FlNum: self._handle_fl_num_item, |
71aaa3f7 | 2282 | _Group: self._handle_group_item, |
2adf4336 | 2283 | _Label: self._handle_label_item, |
320644e2 | 2284 | _MacroExp: self._handle_macro_exp_item, |
71aaa3f7 | 2285 | _Rep: self._handle_rep_item, |
2adf4336 PP |
2286 | _SetBo: self._handle_set_bo_item, |
2287 | _SetOffset: self._handle_set_offset_item, | |
05f81895 | 2288 | _SLeb128Int: self._handle_leb128_int_item, |
2adf4336 | 2289 | _Str: self._handle_str_item, |
05f81895 | 2290 | _ULeb128Int: self._handle_leb128_int_item, |
2adf4336 | 2291 | _VarAssign: self._handle_var_assign_item, |
320644e2 | 2292 | } # type: Dict[type, Callable[[Any, _GenState], None]] |
2adf4336 PP |
2293 | |
2294 | # Handle the group item, _not_ removing the immediate labels | |
2295 | # because the `labels` property offers them. | |
320644e2 | 2296 | self._handle_group_item(group, state, False) |
2adf4336 PP |
2297 | |
2298 | # This is actually the final state | |
2299 | self._final_state = state | |
71aaa3f7 | 2300 | |
320644e2 PP |
2301 | # Generate all the fixed-length number bytes now that we know |
2302 | # their full state | |
2303 | self._gen_fl_num_item_insts() | |
2304 | ||
71aaa3f7 PP |
2305 | |
2306 | # Returns a `ParseResult` instance containing the bytes encoded by the | |
2307 | # input string `normand`. | |
2308 | # | |
2309 | # `init_variables` is a dictionary of initial variable names (valid | |
2310 | # Python names) to integral values. A variable name must not be the | |
2311 | # reserved name `ICITTE`. | |
2312 | # | |
2313 | # `init_labels` is a dictionary of initial label names (valid Python | |
2314 | # names) to integral values. A label name must not be the reserved name | |
2315 | # `ICITTE`. | |
2316 | # | |
2317 | # `init_offset` is the initial offset. | |
2318 | # | |
2319 | # `init_byte_order` is the initial byte order. | |
2320 | # | |
2321 | # Raises `ParseError` on any parsing error. | |
2322 | def parse( | |
2323 | normand: str, | |
e57a18e1 PP |
2324 | init_variables: Optional[VariablesT] = None, |
2325 | init_labels: Optional[LabelsT] = None, | |
71aaa3f7 PP |
2326 | init_offset: int = 0, |
2327 | init_byte_order: Optional[ByteOrder] = None, | |
2328 | ): | |
2329 | if init_variables is None: | |
2330 | init_variables = {} | |
2331 | ||
2332 | if init_labels is None: | |
2333 | init_labels = {} | |
2334 | ||
320644e2 | 2335 | parser = _Parser(normand, init_variables, init_labels) |
71aaa3f7 | 2336 | gen = _Gen( |
320644e2 PP |
2337 | parser.res, |
2338 | parser.macro_defs, | |
71aaa3f7 PP |
2339 | init_variables, |
2340 | init_labels, | |
2341 | init_offset, | |
2342 | init_byte_order, | |
2343 | ) | |
2344 | return ParseResult._create( # pyright: ignore[reportPrivateUsage] | |
2345 | gen.data, gen.variables, gen.labels, gen.offset, gen.bo | |
2346 | ) | |
2347 | ||
2348 | ||
2349 | # Parses the command-line arguments. | |
2350 | def _parse_cli_args(): | |
2351 | import argparse | |
2352 | ||
2353 | # Build parser | |
2354 | ap = argparse.ArgumentParser() | |
2355 | ap.add_argument( | |
2356 | "--offset", | |
2357 | metavar="OFFSET", | |
2358 | action="store", | |
2359 | type=int, | |
2360 | default=0, | |
2361 | help="initial offset (positive)", | |
2362 | ) | |
2363 | ap.add_argument( | |
2364 | "-b", | |
2365 | "--byte-order", | |
2366 | metavar="BO", | |
2367 | choices=["be", "le"], | |
2368 | type=str, | |
2369 | help="initial byte order (`be` or `le`)", | |
2370 | ) | |
2371 | ap.add_argument( | |
2372 | "--var", | |
2373 | metavar="NAME=VAL", | |
2374 | action="append", | |
2375 | help="add an initial variable (may be repeated)", | |
2376 | ) | |
2377 | ap.add_argument( | |
2378 | "-l", | |
2379 | "--label", | |
2380 | metavar="NAME=VAL", | |
2381 | action="append", | |
2382 | help="add an initial label (may be repeated)", | |
2383 | ) | |
2384 | ap.add_argument( | |
2385 | "--version", action="version", version="Normand {}".format(__version__) | |
2386 | ) | |
2387 | ap.add_argument( | |
2388 | "path", | |
2389 | metavar="PATH", | |
2390 | action="store", | |
2391 | nargs="?", | |
2392 | help="input path (none means standard input)", | |
2393 | ) | |
2394 | ||
2395 | # Parse | |
2396 | return ap.parse_args() | |
2397 | ||
2398 | ||
2399 | # Raises a command-line error with the message `msg`. | |
2400 | def _raise_cli_error(msg: str) -> NoReturn: | |
2401 | raise RuntimeError("Command-line error: {}".format(msg)) | |
2402 | ||
2403 | ||
2404 | # Returns a dictionary of string to integers from the list of strings | |
2405 | # `args` containing `NAME=VAL` entries. | |
2406 | def _dict_from_arg(args: Optional[List[str]]): | |
e57a18e1 | 2407 | d = {} # type: LabelsT |
71aaa3f7 PP |
2408 | |
2409 | if args is None: | |
2410 | return d | |
2411 | ||
2412 | for arg in args: | |
2413 | m = re.match(r"({})=(\d+)$".format(_py_name_pat.pattern), arg) | |
2414 | ||
2415 | if m is None: | |
2416 | _raise_cli_error("Invalid assignment {}".format(arg)) | |
2417 | ||
2e1c1acd PP |
2418 | d[m.group(1)] = int(m.group(2)) |
2419 | ||
71aaa3f7 PP |
2420 | return d |
2421 | ||
2422 | ||
2423 | # CLI entry point without exception handling. | |
2424 | def _try_run_cli(): | |
2425 | import os.path | |
2426 | ||
2427 | # Parse arguments | |
2428 | args = _parse_cli_args() | |
2429 | ||
2430 | # Read input | |
2431 | if args.path is None: | |
2432 | normand = sys.stdin.read() | |
2433 | else: | |
2434 | with open(args.path) as f: | |
2435 | normand = f.read() | |
2436 | ||
2437 | # Variables and labels | |
e57a18e1 | 2438 | variables = typing.cast(VariablesT, _dict_from_arg(args.var)) |
71aaa3f7 PP |
2439 | labels = _dict_from_arg(args.label) |
2440 | ||
2441 | # Validate offset | |
2442 | if args.offset < 0: | |
2443 | _raise_cli_error("Invalid negative offset {}") | |
2444 | ||
2445 | # Validate and set byte order | |
2446 | bo = None # type: Optional[ByteOrder] | |
2447 | ||
2448 | if args.byte_order is not None: | |
2449 | if args.byte_order == "be": | |
2450 | bo = ByteOrder.BE | |
2451 | else: | |
2452 | assert args.byte_order == "le" | |
2453 | bo = ByteOrder.LE | |
2454 | ||
2455 | # Parse | |
2456 | try: | |
2457 | res = parse(normand, variables, labels, args.offset, bo) | |
2458 | except ParseError as exc: | |
2459 | prefix = "" | |
2460 | ||
2461 | if args.path is not None: | |
2462 | prefix = "{}:".format(os.path.abspath(args.path)) | |
2463 | ||
2464 | _fail( | |
2465 | "{}{}:{} - {}".format( | |
2466 | prefix, exc.text_loc.line_no, exc.text_loc.col_no, str(exc) | |
2467 | ) | |
2468 | ) | |
2469 | ||
2470 | ||
2471 | sys.stdout.buffer.write(res.data) | |
2472 | ||
2473 | ||
2474 | # Prints the exception message `msg` and exits with status 1. | |
2475 | def _fail(msg: str) -> NoReturn: | |
2476 | if not msg.endswith("."): | |
2477 | msg += "." | |
2478 | ||
2479 | print(msg, file=sys.stderr) | |
2480 | sys.exit(1) | |
2481 | ||
2482 | ||
2483 | # CLI entry point. | |
2484 | def _run_cli(): | |
2485 | try: | |
2486 | _try_run_cli() | |
2487 | except Exception as exc: | |
2488 | _fail(str(exc)) | |
2489 | ||
2490 | ||
2491 | if __name__ == "__main__": | |
2492 | _run_cli() |