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" |
261c5ecf | 33 | __version__ = "0.10.0" |
71aaa3f7 PP |
34 | __all__ = [ |
35 | "ByteOrder", | |
36 | "parse", | |
37 | "ParseError", | |
38 | "ParseResult", | |
e57a18e1 PP |
39 | "TextLocation", |
40 | "LabelsT", | |
41 | "VariablesT", | |
71aaa3f7 PP |
42 | "__author__", |
43 | "__version__", | |
44 | ] | |
45 | ||
46 | import re | |
47 | import abc | |
48 | import ast | |
49 | import sys | |
50 | import enum | |
05f81895 | 51 | import math |
71aaa3f7 | 52 | import struct |
e57a18e1 PP |
53 | import typing |
54 | from typing import Any, Set, Dict, List, Union, Pattern, Callable, NoReturn, Optional | |
71aaa3f7 PP |
55 | |
56 | ||
57 | # Text location (line and column numbers). | |
e57a18e1 | 58 | class TextLocation: |
71aaa3f7 PP |
59 | @classmethod |
60 | def _create(cls, line_no: int, col_no: int): | |
61 | self = cls.__new__(cls) | |
62 | self._init(line_no, col_no) | |
63 | return self | |
64 | ||
65 | def __init__(*args, **kwargs): # type: ignore | |
66 | raise NotImplementedError | |
67 | ||
68 | def _init(self, line_no: int, col_no: int): | |
69 | self._line_no = line_no | |
70 | self._col_no = col_no | |
71 | ||
72 | # Line number. | |
73 | @property | |
74 | def line_no(self): | |
75 | return self._line_no | |
76 | ||
77 | # Column number. | |
78 | @property | |
79 | def col_no(self): | |
80 | return self._col_no | |
81 | ||
2adf4336 | 82 | def __repr__(self): |
e57a18e1 | 83 | return "TextLocation({}, {})".format(self._line_no, self._col_no) |
2adf4336 | 84 | |
71aaa3f7 PP |
85 | |
86 | # Any item. | |
87 | class _Item: | |
e57a18e1 | 88 | def __init__(self, text_loc: TextLocation): |
71aaa3f7 PP |
89 | self._text_loc = text_loc |
90 | ||
91 | # Source text location. | |
92 | @property | |
93 | def text_loc(self): | |
94 | return self._text_loc | |
95 | ||
2adf4336 PP |
96 | |
97 | # Scalar item. | |
98 | class _ScalarItem(_Item): | |
71aaa3f7 PP |
99 | # Returns the size, in bytes, of this item. |
100 | @property | |
101 | @abc.abstractmethod | |
102 | def size(self) -> int: | |
103 | ... | |
104 | ||
105 | ||
106 | # A repeatable item. | |
2adf4336 | 107 | class _RepableItem: |
71aaa3f7 PP |
108 | pass |
109 | ||
110 | ||
111 | # Single byte. | |
2adf4336 | 112 | class _Byte(_ScalarItem, _RepableItem): |
e57a18e1 | 113 | def __init__(self, val: int, text_loc: TextLocation): |
71aaa3f7 PP |
114 | super().__init__(text_loc) |
115 | self._val = val | |
116 | ||
117 | # Byte value. | |
118 | @property | |
119 | def val(self): | |
120 | return self._val | |
121 | ||
122 | @property | |
123 | def size(self): | |
124 | return 1 | |
125 | ||
126 | def __repr__(self): | |
676f6189 | 127 | return "_Byte({}, {})".format(hex(self._val), repr(self._text_loc)) |
71aaa3f7 PP |
128 | |
129 | ||
130 | # String. | |
2adf4336 | 131 | class _Str(_ScalarItem, _RepableItem): |
e57a18e1 | 132 | def __init__(self, data: bytes, text_loc: TextLocation): |
71aaa3f7 PP |
133 | super().__init__(text_loc) |
134 | self._data = data | |
135 | ||
136 | # Encoded bytes. | |
137 | @property | |
138 | def data(self): | |
139 | return self._data | |
140 | ||
141 | @property | |
142 | def size(self): | |
143 | return len(self._data) | |
144 | ||
145 | def __repr__(self): | |
676f6189 | 146 | return "_Str({}, {})".format(repr(self._data), repr(self._text_loc)) |
71aaa3f7 PP |
147 | |
148 | ||
149 | # Byte order. | |
150 | @enum.unique | |
151 | class ByteOrder(enum.Enum): | |
152 | # Big endian. | |
153 | BE = "be" | |
154 | ||
155 | # Little endian. | |
156 | LE = "le" | |
157 | ||
158 | ||
2adf4336 PP |
159 | # Byte order setting. |
160 | class _SetBo(_Item): | |
e57a18e1 | 161 | def __init__(self, bo: ByteOrder, text_loc: TextLocation): |
0e8e3169 | 162 | super().__init__(text_loc) |
71aaa3f7 PP |
163 | self._bo = bo |
164 | ||
165 | @property | |
166 | def bo(self): | |
167 | return self._bo | |
168 | ||
2adf4336 | 169 | def __repr__(self): |
676f6189 | 170 | return "_SetBo({}, {})".format(repr(self._bo), repr(self._text_loc)) |
71aaa3f7 PP |
171 | |
172 | ||
173 | # Label. | |
174 | class _Label(_Item): | |
e57a18e1 | 175 | def __init__(self, name: str, text_loc: TextLocation): |
71aaa3f7 PP |
176 | super().__init__(text_loc) |
177 | self._name = name | |
178 | ||
179 | # Label name. | |
180 | @property | |
181 | def name(self): | |
182 | return self._name | |
183 | ||
71aaa3f7 | 184 | def __repr__(self): |
676f6189 | 185 | return "_Label({}, {})".format(repr(self._name), repr(self._text_loc)) |
71aaa3f7 PP |
186 | |
187 | ||
2adf4336 PP |
188 | # Offset setting. |
189 | class _SetOffset(_Item): | |
e57a18e1 | 190 | def __init__(self, val: int, text_loc: TextLocation): |
71aaa3f7 PP |
191 | super().__init__(text_loc) |
192 | self._val = val | |
193 | ||
676f6189 | 194 | # Offset value (bytes). |
71aaa3f7 PP |
195 | @property |
196 | def val(self): | |
197 | return self._val | |
198 | ||
71aaa3f7 | 199 | def __repr__(self): |
676f6189 PP |
200 | return "_SetOffset({}, {})".format(repr(self._val), repr(self._text_loc)) |
201 | ||
202 | ||
203 | # Offset alignment. | |
204 | class _AlignOffset(_Item): | |
e57a18e1 | 205 | def __init__(self, val: int, pad_val: int, text_loc: TextLocation): |
676f6189 PP |
206 | super().__init__(text_loc) |
207 | self._val = val | |
208 | self._pad_val = pad_val | |
209 | ||
210 | # Alignment value (bits). | |
211 | @property | |
212 | def val(self): | |
213 | return self._val | |
214 | ||
215 | # Padding byte value. | |
216 | @property | |
217 | def pad_val(self): | |
218 | return self._pad_val | |
219 | ||
220 | def __repr__(self): | |
221 | return "_AlignOffset({}, {}, {})".format( | |
222 | repr(self._val), repr(self._pad_val), repr(self._text_loc) | |
223 | ) | |
71aaa3f7 PP |
224 | |
225 | ||
226 | # Mixin of containing an AST expression and its string. | |
227 | class _ExprMixin: | |
228 | def __init__(self, expr_str: str, expr: ast.Expression): | |
229 | self._expr_str = expr_str | |
230 | self._expr = expr | |
231 | ||
232 | # Expression string. | |
233 | @property | |
234 | def expr_str(self): | |
235 | return self._expr_str | |
236 | ||
237 | # Expression node to evaluate. | |
238 | @property | |
239 | def expr(self): | |
240 | return self._expr | |
241 | ||
242 | ||
2adf4336 PP |
243 | # Variable assignment. |
244 | class _VarAssign(_Item, _ExprMixin): | |
71aaa3f7 | 245 | def __init__( |
e57a18e1 | 246 | self, name: str, expr_str: str, expr: ast.Expression, text_loc: TextLocation |
71aaa3f7 PP |
247 | ): |
248 | super().__init__(text_loc) | |
249 | _ExprMixin.__init__(self, expr_str, expr) | |
250 | self._name = name | |
251 | ||
252 | # Name. | |
253 | @property | |
254 | def name(self): | |
255 | return self._name | |
256 | ||
71aaa3f7 | 257 | def __repr__(self): |
2adf4336 | 258 | return "_VarAssign({}, {}, {}, {})".format( |
676f6189 PP |
259 | repr(self._name), |
260 | repr(self._expr_str), | |
261 | repr(self._expr), | |
262 | repr(self._text_loc), | |
71aaa3f7 PP |
263 | ) |
264 | ||
265 | ||
269f6eb3 PP |
266 | # Fixed-length number, possibly needing more than one byte. |
267 | class _FlNum(_ScalarItem, _RepableItem, _ExprMixin): | |
71aaa3f7 | 268 | def __init__( |
e57a18e1 | 269 | self, expr_str: str, expr: ast.Expression, len: int, text_loc: TextLocation |
71aaa3f7 PP |
270 | ): |
271 | super().__init__(text_loc) | |
272 | _ExprMixin.__init__(self, expr_str, expr) | |
273 | self._len = len | |
274 | ||
275 | # Length (bits). | |
276 | @property | |
277 | def len(self): | |
278 | return self._len | |
279 | ||
280 | @property | |
281 | def size(self): | |
282 | return self._len // 8 | |
283 | ||
284 | def __repr__(self): | |
269f6eb3 | 285 | return "_FlNum({}, {}, {}, {})".format( |
676f6189 PP |
286 | repr(self._expr_str), |
287 | repr(self._expr), | |
288 | repr(self._len), | |
289 | repr(self._text_loc), | |
71aaa3f7 PP |
290 | ) |
291 | ||
292 | ||
05f81895 PP |
293 | # LEB128 integer. |
294 | class _Leb128Int(_Item, _RepableItem, _ExprMixin): | |
e57a18e1 | 295 | def __init__(self, expr_str: str, expr: ast.Expression, text_loc: TextLocation): |
05f81895 PP |
296 | super().__init__(text_loc) |
297 | _ExprMixin.__init__(self, expr_str, expr) | |
298 | ||
299 | def __repr__(self): | |
300 | return "{}({}, {}, {})".format( | |
301 | self.__class__.__name__, | |
302 | repr(self._expr_str), | |
303 | repr(self._expr), | |
676f6189 | 304 | repr(self._text_loc), |
05f81895 PP |
305 | ) |
306 | ||
307 | ||
308 | # Unsigned LEB128 integer. | |
309 | class _ULeb128Int(_Leb128Int, _RepableItem, _ExprMixin): | |
310 | pass | |
311 | ||
312 | ||
313 | # Signed LEB128 integer. | |
314 | class _SLeb128Int(_Leb128Int, _RepableItem, _ExprMixin): | |
315 | pass | |
316 | ||
317 | ||
71aaa3f7 | 318 | # Group of items. |
2adf4336 | 319 | class _Group(_Item, _RepableItem): |
e57a18e1 | 320 | def __init__(self, items: List[_Item], text_loc: TextLocation): |
71aaa3f7 PP |
321 | super().__init__(text_loc) |
322 | self._items = items | |
71aaa3f7 PP |
323 | |
324 | # Contained items. | |
325 | @property | |
326 | def items(self): | |
327 | return self._items | |
328 | ||
71aaa3f7 | 329 | def __repr__(self): |
676f6189 | 330 | return "_Group({}, {})".format(repr(self._items), repr(self._text_loc)) |
71aaa3f7 PP |
331 | |
332 | ||
333 | # Repetition item. | |
2adf4336 PP |
334 | class _Rep(_Item, _ExprMixin): |
335 | def __init__( | |
e57a18e1 | 336 | self, item: _Item, expr_str: str, expr: ast.Expression, text_loc: TextLocation |
2adf4336 | 337 | ): |
71aaa3f7 | 338 | super().__init__(text_loc) |
2adf4336 | 339 | _ExprMixin.__init__(self, expr_str, expr) |
71aaa3f7 | 340 | self._item = item |
71aaa3f7 PP |
341 | |
342 | # Item to repeat. | |
343 | @property | |
344 | def item(self): | |
345 | return self._item | |
346 | ||
71aaa3f7 | 347 | def __repr__(self): |
2adf4336 | 348 | return "_Rep({}, {}, {}, {})".format( |
676f6189 PP |
349 | repr(self._item), |
350 | repr(self._expr_str), | |
351 | repr(self._expr), | |
352 | repr(self._text_loc), | |
71aaa3f7 PP |
353 | ) |
354 | ||
355 | ||
27d52a19 PP |
356 | # Conditional item. |
357 | class _Cond(_Item, _ExprMixin): | |
358 | def __init__( | |
359 | self, item: _Item, expr_str: str, expr: ast.Expression, text_loc: TextLocation | |
360 | ): | |
361 | super().__init__(text_loc) | |
362 | _ExprMixin.__init__(self, expr_str, expr) | |
363 | self._item = item | |
364 | ||
365 | # Conditional item. | |
366 | @property | |
367 | def item(self): | |
368 | return self._item | |
369 | ||
370 | def __repr__(self): | |
371 | return "_Cond({}, {}, {}, {})".format( | |
372 | repr(self._item), | |
373 | repr(self._expr_str), | |
374 | repr(self._expr), | |
375 | repr(self._text_loc), | |
376 | ) | |
377 | ||
378 | ||
2adf4336 | 379 | # Expression item type. |
27d52a19 | 380 | _ExprItemT = Union[_FlNum, _Leb128Int, _VarAssign, _Rep, _Cond] |
2adf4336 PP |
381 | |
382 | ||
71aaa3f7 PP |
383 | # A parsing error containing a message and a text location. |
384 | class ParseError(RuntimeError): | |
385 | @classmethod | |
e57a18e1 | 386 | def _create(cls, msg: str, text_loc: TextLocation): |
71aaa3f7 PP |
387 | self = cls.__new__(cls) |
388 | self._init(msg, text_loc) | |
389 | return self | |
390 | ||
391 | def __init__(self, *args, **kwargs): # type: ignore | |
392 | raise NotImplementedError | |
393 | ||
e57a18e1 | 394 | def _init(self, msg: str, text_loc: TextLocation): |
71aaa3f7 PP |
395 | super().__init__(msg) |
396 | self._text_loc = text_loc | |
397 | ||
398 | # Source text location. | |
399 | @property | |
400 | def text_loc(self): | |
401 | return self._text_loc | |
402 | ||
403 | ||
404 | # Raises a parsing error, forwarding the parameters to the constructor. | |
e57a18e1 | 405 | def _raise_error(msg: str, text_loc: TextLocation) -> NoReturn: |
71aaa3f7 PP |
406 | raise ParseError._create(msg, text_loc) # pyright: ignore[reportPrivateUsage] |
407 | ||
408 | ||
e57a18e1 PP |
409 | # Variables dictionary type (for type hints). |
410 | VariablesT = Dict[str, Union[int, float]] | |
411 | ||
412 | ||
413 | # Labels dictionary type (for type hints). | |
414 | LabelsT = Dict[str, int] | |
71aaa3f7 PP |
415 | |
416 | ||
417 | # Python name pattern. | |
418 | _py_name_pat = re.compile(r"[a-zA-Z_][a-zA-Z0-9_]*") | |
419 | ||
420 | ||
421 | # Normand parser. | |
422 | # | |
423 | # The constructor accepts a Normand input. After building, use the `res` | |
424 | # property to get the resulting main group. | |
425 | class _Parser: | |
426 | # Builds a parser to parse the Normand input `normand`, parsing | |
427 | # immediately. | |
e57a18e1 | 428 | def __init__(self, normand: str, variables: VariablesT, labels: LabelsT): |
71aaa3f7 PP |
429 | self._normand = normand |
430 | self._at = 0 | |
431 | self._line_no = 1 | |
432 | self._col_no = 1 | |
433 | self._label_names = set(labels.keys()) | |
434 | self._var_names = set(variables.keys()) | |
435 | self._parse() | |
436 | ||
437 | # Result (main group). | |
438 | @property | |
439 | def res(self): | |
440 | return self._res | |
441 | ||
442 | # Current text location. | |
443 | @property | |
444 | def _text_loc(self): | |
e57a18e1 | 445 | return TextLocation._create( # pyright: ignore[reportPrivateUsage] |
71aaa3f7 PP |
446 | self._line_no, self._col_no |
447 | ) | |
448 | ||
449 | # Returns `True` if this parser is done parsing. | |
450 | def _is_done(self): | |
451 | return self._at == len(self._normand) | |
452 | ||
453 | # Returns `True` if this parser isn't done parsing. | |
454 | def _isnt_done(self): | |
455 | return not self._is_done() | |
456 | ||
457 | # Raises a parse error, creating it using the message `msg` and the | |
458 | # current text location. | |
459 | def _raise_error(self, msg: str) -> NoReturn: | |
460 | _raise_error(msg, self._text_loc) | |
461 | ||
462 | # Tries to make the pattern `pat` match the current substring, | |
463 | # returning the match object and updating `self._at`, | |
464 | # `self._line_no`, and `self._col_no` on success. | |
465 | def _try_parse_pat(self, pat: Pattern[str]): | |
466 | m = pat.match(self._normand, self._at) | |
467 | ||
468 | if m is None: | |
469 | return | |
470 | ||
471 | # Skip matched string | |
472 | self._at += len(m.group(0)) | |
473 | ||
474 | # Update line number | |
475 | self._line_no += m.group(0).count("\n") | |
476 | ||
477 | # Update column number | |
478 | for i in reversed(range(self._at)): | |
479 | if self._normand[i] == "\n" or i == 0: | |
480 | if i == 0: | |
481 | self._col_no = self._at + 1 | |
482 | else: | |
483 | self._col_no = self._at - i | |
484 | ||
485 | break | |
486 | ||
487 | # Return match object | |
488 | return m | |
489 | ||
490 | # Expects the pattern `pat` to match the current substring, | |
491 | # returning the match object and updating `self._at`, | |
492 | # `self._line_no`, and `self._col_no` on success, or raising a parse | |
493 | # error with the message `error_msg` on error. | |
494 | def _expect_pat(self, pat: Pattern[str], error_msg: str): | |
495 | # Match | |
496 | m = self._try_parse_pat(pat) | |
497 | ||
498 | if m is None: | |
499 | # No match: error | |
500 | self._raise_error(error_msg) | |
501 | ||
502 | # Return match object | |
503 | return m | |
504 | ||
505 | # Pattern for _skip_ws_and_comments() | |
506 | _ws_or_syms_or_comments_pat = re.compile( | |
e57a18e1 | 507 | r"(?:[\s/\\?&:;.,+[\]_=|-]|#[^#]*?(?:\n|#))*" |
71aaa3f7 PP |
508 | ) |
509 | ||
510 | # Skips as many whitespaces, insignificant symbol characters, and | |
511 | # comments as possible. | |
512 | def _skip_ws_and_comments(self): | |
513 | self._try_parse_pat(self._ws_or_syms_or_comments_pat) | |
514 | ||
515 | # Pattern for _try_parse_hex_byte() | |
516 | _nibble_pat = re.compile(r"[A-Fa-f0-9]") | |
517 | ||
518 | # Tries to parse a hexadecimal byte, returning a byte item on | |
519 | # success. | |
520 | def _try_parse_hex_byte(self): | |
0e8e3169 PP |
521 | begin_text_loc = self._text_loc |
522 | ||
71aaa3f7 PP |
523 | # Match initial nibble |
524 | m_high = self._try_parse_pat(self._nibble_pat) | |
525 | ||
526 | if m_high is None: | |
527 | # No match | |
528 | return | |
529 | ||
530 | # Expect another nibble | |
531 | self._skip_ws_and_comments() | |
532 | m_low = self._expect_pat( | |
533 | self._nibble_pat, "Expecting another hexadecimal nibble" | |
534 | ) | |
535 | ||
536 | # Return item | |
0e8e3169 | 537 | return _Byte(int(m_high.group(0) + m_low.group(0), 16), begin_text_loc) |
71aaa3f7 PP |
538 | |
539 | # Patterns for _try_parse_bin_byte() | |
540 | _bin_byte_bit_pat = re.compile(r"[01]") | |
541 | _bin_byte_prefix_pat = re.compile(r"%") | |
542 | ||
543 | # Tries to parse a binary byte, returning a byte item on success. | |
544 | def _try_parse_bin_byte(self): | |
0e8e3169 PP |
545 | begin_text_loc = self._text_loc |
546 | ||
71aaa3f7 PP |
547 | # Match prefix |
548 | if self._try_parse_pat(self._bin_byte_prefix_pat) is None: | |
549 | # No match | |
550 | return | |
551 | ||
552 | # Expect eight bits | |
553 | bits = [] # type: List[str] | |
554 | ||
555 | for _ in range(8): | |
556 | self._skip_ws_and_comments() | |
557 | m = self._expect_pat(self._bin_byte_bit_pat, "Expecting a bit (`0` or `1`)") | |
558 | bits.append(m.group(0)) | |
559 | ||
560 | # Return item | |
0e8e3169 | 561 | return _Byte(int("".join(bits), 2), begin_text_loc) |
71aaa3f7 PP |
562 | |
563 | # Patterns for _try_parse_dec_byte() | |
564 | _dec_byte_prefix_pat = re.compile(r"\$\s*") | |
565 | _dec_byte_val_pat = re.compile(r"(?P<neg>-?)(?P<val>\d+)") | |
566 | ||
567 | # Tries to parse a decimal byte, returning a byte item on success. | |
568 | def _try_parse_dec_byte(self): | |
0e8e3169 PP |
569 | begin_text_loc = self._text_loc |
570 | ||
71aaa3f7 PP |
571 | # Match prefix |
572 | if self._try_parse_pat(self._dec_byte_prefix_pat) is None: | |
573 | # No match | |
574 | return | |
575 | ||
576 | # Expect the value | |
577 | m = self._expect_pat(self._dec_byte_val_pat, "Expecting a decimal constant") | |
578 | ||
579 | # Compute value | |
580 | val = int(m.group("val")) * (-1 if m.group("neg") == "-" else 1) | |
581 | ||
582 | # Validate | |
583 | if val < -128 or val > 255: | |
0e8e3169 | 584 | _raise_error("Invalid decimal byte value {}".format(val), begin_text_loc) |
71aaa3f7 PP |
585 | |
586 | # Two's complement | |
05f81895 | 587 | val %= 256 |
71aaa3f7 PP |
588 | |
589 | # Return item | |
0e8e3169 | 590 | return _Byte(val, begin_text_loc) |
71aaa3f7 PP |
591 | |
592 | # Tries to parse a byte, returning a byte item on success. | |
593 | def _try_parse_byte(self): | |
594 | # Hexadecimal | |
595 | item = self._try_parse_hex_byte() | |
596 | ||
597 | if item is not None: | |
598 | return item | |
599 | ||
600 | # Binary | |
601 | item = self._try_parse_bin_byte() | |
602 | ||
603 | if item is not None: | |
604 | return item | |
605 | ||
606 | # Decimal | |
607 | item = self._try_parse_dec_byte() | |
608 | ||
609 | if item is not None: | |
610 | return item | |
611 | ||
612 | # Patterns for _try_parse_str() | |
613 | _str_prefix_pat = re.compile(r'(?:u(?P<len>16|32)(?P<bo>be|le))?\s*"') | |
614 | _str_suffix_pat = re.compile(r'"') | |
615 | _str_str_pat = re.compile(r'(?:(?:\\.)|[^"])*') | |
616 | ||
617 | # Strings corresponding to escape sequence characters | |
618 | _str_escape_seq_strs = { | |
619 | "0": "\0", | |
620 | "a": "\a", | |
621 | "b": "\b", | |
622 | "e": "\x1b", | |
623 | "f": "\f", | |
624 | "n": "\n", | |
625 | "r": "\r", | |
626 | "t": "\t", | |
627 | "v": "\v", | |
628 | "\\": "\\", | |
629 | '"': '"', | |
630 | } | |
631 | ||
632 | # Tries to parse a string, returning a string item on success. | |
633 | def _try_parse_str(self): | |
0e8e3169 PP |
634 | begin_text_loc = self._text_loc |
635 | ||
71aaa3f7 PP |
636 | # Match prefix |
637 | m = self._try_parse_pat(self._str_prefix_pat) | |
638 | ||
639 | if m is None: | |
640 | # No match | |
641 | return | |
642 | ||
643 | # Get encoding | |
644 | encoding = "utf8" | |
645 | ||
646 | if m.group("len") is not None: | |
647 | encoding = "utf_{}_{}".format(m.group("len"), m.group("bo")) | |
648 | ||
649 | # Actual string | |
650 | m = self._expect_pat(self._str_str_pat, "Expecting a literal string") | |
651 | ||
652 | # Expect end of string | |
653 | self._expect_pat(self._str_suffix_pat, 'Expecting `"` (end of literal string)') | |
654 | ||
655 | # Replace escape sequences | |
656 | val = m.group(0) | |
657 | ||
658 | for ec in '0abefnrtv"\\': | |
659 | val = val.replace(r"\{}".format(ec), self._str_escape_seq_strs[ec]) | |
660 | ||
661 | # Encode | |
662 | data = val.encode(encoding) | |
663 | ||
664 | # Return item | |
0e8e3169 | 665 | return _Str(data, begin_text_loc) |
71aaa3f7 PP |
666 | |
667 | # Patterns for _try_parse_group() | |
261c5ecf PP |
668 | _group_prefix_pat = re.compile(r"\(|!g(roup)?\b") |
669 | _group_suffix_paren_pat = re.compile(r"\)") | |
71aaa3f7 PP |
670 | |
671 | # Tries to parse a group, returning a group item on success. | |
672 | def _try_parse_group(self): | |
0e8e3169 PP |
673 | begin_text_loc = self._text_loc |
674 | ||
71aaa3f7 | 675 | # Match prefix |
261c5ecf PP |
676 | m_open = self._try_parse_pat(self._group_prefix_pat) |
677 | ||
678 | if m_open is None: | |
71aaa3f7 PP |
679 | # No match |
680 | return | |
681 | ||
682 | # Parse items | |
683 | items = self._parse_items() | |
684 | ||
685 | # Expect end of group | |
686 | self._skip_ws_and_comments() | |
261c5ecf PP |
687 | |
688 | if m_open.group(0) == "(": | |
689 | pat = self._group_suffix_paren_pat | |
690 | exp = ")" | |
691 | else: | |
692 | pat = self._block_end_pat | |
693 | exp = "!end" | |
694 | ||
695 | self._expect_pat(pat, "Expecting an item or `{}` (end of group)".format(exp)) | |
71aaa3f7 PP |
696 | |
697 | # Return item | |
0e8e3169 | 698 | return _Group(items, begin_text_loc) |
71aaa3f7 PP |
699 | |
700 | # Returns a stripped expression string and an AST expression node | |
701 | # from the expression string `expr_str` at text location `text_loc`. | |
e57a18e1 | 702 | def _ast_expr_from_str(self, expr_str: str, text_loc: TextLocation): |
71aaa3f7 PP |
703 | # Create an expression node from the expression string |
704 | expr_str = expr_str.strip().replace("\n", " ") | |
705 | ||
706 | try: | |
707 | expr = ast.parse(expr_str, mode="eval") | |
708 | except SyntaxError: | |
709 | _raise_error( | |
710 | "Invalid expression `{}`: invalid syntax".format(expr_str), | |
711 | text_loc, | |
712 | ) | |
713 | ||
714 | return expr_str, expr | |
715 | ||
269f6eb3 | 716 | # Patterns for _try_parse_num_and_attr() |
05f81895 | 717 | _val_expr_pat = re.compile(r"([^}:]+):\s*") |
269f6eb3 | 718 | _fl_num_len_attr_pat = re.compile(r"8|16|24|32|40|48|56|64") |
05f81895 | 719 | _leb128_int_attr_pat = re.compile(r"(u|s)leb128") |
71aaa3f7 | 720 | |
05f81895 PP |
721 | # Tries to parse a value and attribute (fixed length in bits or |
722 | # `leb128`), returning a value item on success. | |
269f6eb3 | 723 | def _try_parse_num_and_attr(self): |
71aaa3f7 PP |
724 | begin_text_loc = self._text_loc |
725 | ||
726 | # Match | |
727 | m_expr = self._try_parse_pat(self._val_expr_pat) | |
728 | ||
729 | if m_expr is None: | |
730 | # No match | |
731 | return | |
732 | ||
71aaa3f7 PP |
733 | # Create an expression node from the expression string |
734 | expr_str, expr = self._ast_expr_from_str(m_expr.group(1), begin_text_loc) | |
735 | ||
05f81895 | 736 | # Length? |
269f6eb3 | 737 | m_attr = self._try_parse_pat(self._fl_num_len_attr_pat) |
05f81895 PP |
738 | |
739 | if m_attr is None: | |
740 | # LEB128? | |
741 | m_attr = self._try_parse_pat(self._leb128_int_attr_pat) | |
742 | ||
743 | if m_attr is None: | |
744 | # At this point it's invalid | |
745 | self._raise_error( | |
746 | "Expecting a length (multiple of eight bits), `uleb128`, or `sleb128`" | |
747 | ) | |
748 | ||
749 | # Return LEB128 integer item | |
750 | cls = _ULeb128Int if m_attr.group(1) == "u" else _SLeb128Int | |
751 | return cls(expr_str, expr, begin_text_loc) | |
752 | else: | |
269f6eb3 PP |
753 | # Return fixed-length number item |
754 | return _FlNum( | |
05f81895 PP |
755 | expr_str, |
756 | expr, | |
757 | int(m_attr.group(0)), | |
758 | begin_text_loc, | |
759 | ) | |
71aaa3f7 | 760 | |
269f6eb3 | 761 | # Patterns for _try_parse_num_and_attr() |
2adf4336 | 762 | _var_assign_pat = re.compile( |
71aaa3f7 PP |
763 | r"(?P<name>{})\s*=\s*(?P<expr>[^}}]+)".format(_py_name_pat.pattern) |
764 | ) | |
765 | ||
2adf4336 PP |
766 | # Tries to parse a variable assignment, returning a variable |
767 | # assignment item on success. | |
768 | def _try_parse_var_assign(self): | |
71aaa3f7 PP |
769 | begin_text_loc = self._text_loc |
770 | ||
771 | # Match | |
2adf4336 | 772 | m = self._try_parse_pat(self._var_assign_pat) |
71aaa3f7 PP |
773 | |
774 | if m is None: | |
775 | # No match | |
776 | return | |
777 | ||
778 | # Validate name | |
779 | name = m.group("name") | |
780 | ||
781 | if name == _icitte_name: | |
0e8e3169 PP |
782 | _raise_error( |
783 | "`{}` is a reserved variable name".format(_icitte_name), begin_text_loc | |
784 | ) | |
71aaa3f7 PP |
785 | |
786 | if name in self._label_names: | |
0e8e3169 | 787 | _raise_error("Existing label named `{}`".format(name), begin_text_loc) |
71aaa3f7 PP |
788 | |
789 | # Add to known variable names | |
790 | self._var_names.add(name) | |
791 | ||
792 | # Create an expression node from the expression string | |
793 | expr_str, expr = self._ast_expr_from_str(m.group("expr"), begin_text_loc) | |
794 | ||
795 | # Return item | |
2adf4336 | 796 | return _VarAssign( |
71aaa3f7 PP |
797 | name, |
798 | expr_str, | |
799 | expr, | |
0e8e3169 | 800 | begin_text_loc, |
71aaa3f7 PP |
801 | ) |
802 | ||
2adf4336 | 803 | # Pattern for _try_parse_set_bo() |
71aaa3f7 PP |
804 | _bo_pat = re.compile(r"[bl]e") |
805 | ||
2adf4336 PP |
806 | # Tries to parse a byte order name, returning a byte order setting |
807 | # item on success. | |
808 | def _try_parse_set_bo(self): | |
0e8e3169 PP |
809 | begin_text_loc = self._text_loc |
810 | ||
71aaa3f7 PP |
811 | # Match |
812 | m = self._try_parse_pat(self._bo_pat) | |
813 | ||
814 | if m is None: | |
815 | # No match | |
816 | return | |
817 | ||
818 | # Return corresponding item | |
819 | if m.group(0) == "be": | |
2adf4336 | 820 | return _SetBo(ByteOrder.BE, begin_text_loc) |
71aaa3f7 PP |
821 | else: |
822 | assert m.group(0) == "le" | |
2adf4336 | 823 | return _SetBo(ByteOrder.LE, begin_text_loc) |
71aaa3f7 PP |
824 | |
825 | # Patterns for _try_parse_val_or_bo() | |
2adf4336 PP |
826 | _val_var_assign_set_bo_prefix_pat = re.compile(r"\{\s*") |
827 | _val_var_assign_set_bo_suffix_pat = re.compile(r"\s*}") | |
71aaa3f7 | 828 | |
2adf4336 PP |
829 | # Tries to parse a value, a variable assignment, or a byte order |
830 | # setting, returning an item on success. | |
831 | def _try_parse_val_or_var_assign_or_set_bo(self): | |
71aaa3f7 | 832 | # Match prefix |
2adf4336 | 833 | if self._try_parse_pat(self._val_var_assign_set_bo_prefix_pat) is None: |
71aaa3f7 PP |
834 | # No match |
835 | return | |
836 | ||
2adf4336 PP |
837 | # Variable assignment item? |
838 | item = self._try_parse_var_assign() | |
71aaa3f7 PP |
839 | |
840 | if item is None: | |
269f6eb3 PP |
841 | # Number item? |
842 | item = self._try_parse_num_and_attr() | |
71aaa3f7 PP |
843 | |
844 | if item is None: | |
2adf4336 PP |
845 | # Byte order setting item? |
846 | item = self._try_parse_set_bo() | |
71aaa3f7 PP |
847 | |
848 | if item is None: | |
849 | # At this point it's invalid | |
2adf4336 | 850 | self._raise_error( |
269f6eb3 | 851 | "Expecting a fixed-length number, a variable assignment, or a byte order setting" |
2adf4336 | 852 | ) |
71aaa3f7 PP |
853 | |
854 | # Expect suffix | |
2adf4336 | 855 | self._expect_pat(self._val_var_assign_set_bo_suffix_pat, "Expecting `}`") |
71aaa3f7 PP |
856 | return item |
857 | ||
e57a18e1 | 858 | # Common positive constant integer pattern |
71aaa3f7 PP |
859 | _pos_const_int_pat = re.compile(r"0[Xx][A-Fa-f0-9]+|\d+") |
860 | ||
2adf4336 PP |
861 | # Tries to parse an offset setting value (after the initial `<`), |
862 | # returning an offset item on success. | |
863 | def _try_parse_set_offset_val(self): | |
0e8e3169 PP |
864 | begin_text_loc = self._text_loc |
865 | ||
71aaa3f7 PP |
866 | # Match |
867 | m = self._try_parse_pat(self._pos_const_int_pat) | |
868 | ||
869 | if m is None: | |
870 | # No match | |
871 | return | |
872 | ||
873 | # Return item | |
2adf4336 | 874 | return _SetOffset(int(m.group(0), 0), begin_text_loc) |
71aaa3f7 PP |
875 | |
876 | # Tries to parse a label name (after the initial `<`), returning a | |
877 | # label item on success. | |
878 | def _try_parse_label_name(self): | |
0e8e3169 PP |
879 | begin_text_loc = self._text_loc |
880 | ||
71aaa3f7 PP |
881 | # Match |
882 | m = self._try_parse_pat(_py_name_pat) | |
883 | ||
884 | if m is None: | |
885 | # No match | |
886 | return | |
887 | ||
888 | # Validate | |
889 | name = m.group(0) | |
890 | ||
891 | if name == _icitte_name: | |
0e8e3169 PP |
892 | _raise_error( |
893 | "`{}` is a reserved label name".format(_icitte_name), begin_text_loc | |
894 | ) | |
71aaa3f7 PP |
895 | |
896 | if name in self._label_names: | |
0e8e3169 | 897 | _raise_error("Duplicate label name `{}`".format(name), begin_text_loc) |
71aaa3f7 PP |
898 | |
899 | if name in self._var_names: | |
0e8e3169 | 900 | _raise_error("Existing variable named `{}`".format(name), begin_text_loc) |
71aaa3f7 PP |
901 | |
902 | # Add to known label names | |
903 | self._label_names.add(name) | |
904 | ||
905 | # Return item | |
0e8e3169 | 906 | return _Label(name, begin_text_loc) |
71aaa3f7 | 907 | |
2adf4336 PP |
908 | # Patterns for _try_parse_label_or_set_offset() |
909 | _label_set_offset_prefix_pat = re.compile(r"<\s*") | |
910 | _label_set_offset_suffix_pat = re.compile(r"\s*>") | |
71aaa3f7 | 911 | |
2adf4336 PP |
912 | # Tries to parse a label or an offset setting, returning an item on |
913 | # success. | |
914 | def _try_parse_label_or_set_offset(self): | |
71aaa3f7 | 915 | # Match prefix |
2adf4336 | 916 | if self._try_parse_pat(self._label_set_offset_prefix_pat) is None: |
71aaa3f7 PP |
917 | # No match |
918 | return | |
919 | ||
2adf4336 PP |
920 | # Offset setting item? |
921 | item = self._try_parse_set_offset_val() | |
71aaa3f7 PP |
922 | |
923 | if item is None: | |
924 | # Label item? | |
925 | item = self._try_parse_label_name() | |
926 | ||
927 | if item is None: | |
928 | # At this point it's invalid | |
2adf4336 | 929 | self._raise_error("Expecting a label name or an offset setting value") |
71aaa3f7 PP |
930 | |
931 | # Expect suffix | |
2adf4336 | 932 | self._expect_pat(self._label_set_offset_suffix_pat, "Expecting `>`") |
71aaa3f7 PP |
933 | return item |
934 | ||
676f6189 PP |
935 | # Patterns for _try_parse_align_offset() |
936 | _align_offset_prefix_pat = re.compile(r"@\s*") | |
937 | _align_offset_val_pat = re.compile(r"(\d+)\s*") | |
938 | _align_offset_pad_val_prefix_pat = re.compile(r"~\s*") | |
939 | ||
940 | # Tries to parse an offset alignment, returning an offset alignment | |
941 | # item on success. | |
942 | def _try_parse_align_offset(self): | |
943 | begin_text_loc = self._text_loc | |
944 | ||
945 | # Match prefix | |
946 | if self._try_parse_pat(self._align_offset_prefix_pat) is None: | |
947 | # No match | |
948 | return | |
949 | ||
950 | align_text_loc = self._text_loc | |
951 | m = self._expect_pat( | |
952 | self._align_offset_val_pat, | |
953 | "Expecting an alignment (positive multiple of eight bits)", | |
954 | ) | |
955 | ||
956 | # Validate alignment | |
957 | val = int(m.group(1)) | |
958 | ||
959 | if val <= 0 or (val % 8) != 0: | |
960 | _raise_error( | |
961 | "Invalid alignment value {} (not a positive multiple of eight)".format( | |
962 | val | |
963 | ), | |
964 | align_text_loc, | |
965 | ) | |
966 | ||
967 | # Padding value? | |
968 | pad_val = 0 | |
969 | ||
970 | if self._try_parse_pat(self._align_offset_pad_val_prefix_pat) is not None: | |
971 | pad_val_text_loc = self._text_loc | |
972 | m = self._expect_pat(self._pos_const_int_pat, "Expecting a byte value") | |
973 | ||
974 | # Validate | |
975 | pad_val = int(m.group(0), 0) | |
976 | ||
977 | if pad_val > 255: | |
978 | _raise_error( | |
979 | "Invalid padding byte value {}".format(pad_val), | |
980 | pad_val_text_loc, | |
981 | ) | |
982 | ||
983 | # Return item | |
984 | return _AlignOffset(val, pad_val, begin_text_loc) | |
985 | ||
e57a18e1 | 986 | # Patterns for _expect_rep_mul_expr() |
27d52a19 PP |
987 | _rep_cond_expr_prefix_pat = re.compile(r"\{") |
988 | _rep_cond_expr_pat = re.compile(r"[^}]+") | |
989 | _rep_cond_expr_suffix_pat = re.compile(r"\}") | |
990 | ||
991 | # Parses the expression of a conditional block or of a repetition | |
992 | # (block or post-item) and returns the expression string and AST | |
993 | # node. | |
994 | def _expect_rep_cond_expr(self, accept_int: bool): | |
e57a18e1 PP |
995 | expr_text_loc = self._text_loc |
996 | ||
997 | # Constant integer? | |
27d52a19 PP |
998 | m = None |
999 | ||
1000 | if accept_int: | |
1001 | m = self._try_parse_pat(self._pos_const_int_pat) | |
e57a18e1 PP |
1002 | |
1003 | if m is None: | |
1004 | # Name? | |
1005 | m = self._try_parse_pat(_py_name_pat) | |
1006 | ||
1007 | if m is None: | |
1008 | # Expression? | |
27d52a19 PP |
1009 | if self._try_parse_pat(self._rep_cond_expr_prefix_pat) is None: |
1010 | if accept_int: | |
1011 | mid_msg = "a positive constant integer, a name, or `{`" | |
1012 | else: | |
1013 | mid_msg = "a name or `{`" | |
1014 | ||
e57a18e1 | 1015 | # At this point it's invalid |
27d52a19 | 1016 | self._raise_error("Expecting {}".format(mid_msg)) |
e57a18e1 PP |
1017 | |
1018 | # Expect an expression | |
1019 | expr_text_loc = self._text_loc | |
27d52a19 | 1020 | m = self._expect_pat(self._rep_cond_expr_pat, "Expecting an expression") |
e57a18e1 PP |
1021 | expr_str = m.group(0) |
1022 | ||
1023 | # Expect `}` | |
27d52a19 | 1024 | self._expect_pat(self._rep_cond_expr_suffix_pat, "Expecting `}`") |
e57a18e1 PP |
1025 | else: |
1026 | expr_str = m.group(0) | |
1027 | else: | |
1028 | expr_str = m.group(0) | |
1029 | ||
1030 | return self._ast_expr_from_str(expr_str, expr_text_loc) | |
1031 | ||
27d52a19 PP |
1032 | # Parses the multiplier expression of a repetition (block or |
1033 | # post-item) and returns the expression string and AST node. | |
1034 | def _expect_rep_mul_expr(self): | |
1035 | return self._expect_rep_cond_expr(True) | |
1036 | ||
1037 | # Common block end pattern | |
1038 | _block_end_pat = re.compile(r"!end\b\s*") | |
1039 | ||
e57a18e1 PP |
1040 | # Pattern for _try_parse_rep_block() |
1041 | _rep_block_prefix_pat = re.compile(r"!r(?:epeat)?\b\s*") | |
e57a18e1 PP |
1042 | |
1043 | # Tries to parse a repetition block, returning a repetition item on | |
1044 | # success. | |
1045 | def _try_parse_rep_block(self): | |
1046 | begin_text_loc = self._text_loc | |
1047 | ||
1048 | # Match prefix | |
1049 | if self._try_parse_pat(self._rep_block_prefix_pat) is None: | |
1050 | # No match | |
1051 | return | |
1052 | ||
1053 | # Expect expression | |
1054 | self._skip_ws_and_comments() | |
1055 | expr_str, expr = self._expect_rep_mul_expr() | |
1056 | ||
1057 | # Parse items | |
1058 | self._skip_ws_and_comments() | |
1059 | items_text_loc = self._text_loc | |
1060 | items = self._parse_items() | |
1061 | ||
1062 | # Expect end of block | |
1063 | self._skip_ws_and_comments() | |
1064 | self._expect_pat( | |
27d52a19 | 1065 | self._block_end_pat, "Expecting an item or `!end` (end of repetition block)" |
e57a18e1 PP |
1066 | ) |
1067 | ||
1068 | # Return item | |
1069 | return _Rep(_Group(items, items_text_loc), expr_str, expr, begin_text_loc) | |
1070 | ||
27d52a19 PP |
1071 | # Pattern for _try_parse_cond_block() |
1072 | _cond_block_prefix_pat = re.compile(r"!if\b\s*") | |
1073 | ||
1074 | # Tries to parse a conditional block, returning a conditional item | |
1075 | # on success. | |
1076 | def _try_parse_cond_block(self): | |
1077 | begin_text_loc = self._text_loc | |
1078 | ||
1079 | # Match prefix | |
1080 | if self._try_parse_pat(self._cond_block_prefix_pat) is None: | |
1081 | # No match | |
1082 | return | |
1083 | ||
1084 | # Expect expression | |
1085 | self._skip_ws_and_comments() | |
1086 | expr_str, expr = self._expect_rep_cond_expr(False) | |
1087 | ||
1088 | # Parse items | |
1089 | self._skip_ws_and_comments() | |
1090 | items_text_loc = self._text_loc | |
1091 | items = self._parse_items() | |
1092 | ||
1093 | # Expect end of block | |
1094 | self._skip_ws_and_comments() | |
1095 | self._expect_pat( | |
1096 | self._block_end_pat, | |
1097 | "Expecting an item or `!end` (end of conditional block)", | |
1098 | ) | |
1099 | ||
1100 | # Return item | |
1101 | return _Cond(_Group(items, items_text_loc), expr_str, expr, begin_text_loc) | |
1102 | ||
71aaa3f7 PP |
1103 | # Tries to parse a base item (anything except a repetition), |
1104 | # returning it on success. | |
1105 | def _try_parse_base_item(self): | |
1106 | # Byte item? | |
1107 | item = self._try_parse_byte() | |
1108 | ||
1109 | if item is not None: | |
1110 | return item | |
1111 | ||
1112 | # String item? | |
1113 | item = self._try_parse_str() | |
1114 | ||
1115 | if item is not None: | |
1116 | return item | |
1117 | ||
2adf4336 PP |
1118 | # Value, variable assignment, or byte order setting item? |
1119 | item = self._try_parse_val_or_var_assign_or_set_bo() | |
71aaa3f7 PP |
1120 | |
1121 | if item is not None: | |
1122 | return item | |
1123 | ||
2adf4336 PP |
1124 | # Label or offset setting item? |
1125 | item = self._try_parse_label_or_set_offset() | |
71aaa3f7 | 1126 | |
676f6189 PP |
1127 | if item is not None: |
1128 | return item | |
1129 | ||
1130 | # Offset alignment item? | |
1131 | item = self._try_parse_align_offset() | |
1132 | ||
71aaa3f7 PP |
1133 | if item is not None: |
1134 | return item | |
1135 | ||
1136 | # Group item? | |
1137 | item = self._try_parse_group() | |
1138 | ||
1139 | if item is not None: | |
1140 | return item | |
1141 | ||
e57a18e1 PP |
1142 | # Repetition (block) item? |
1143 | item = self._try_parse_rep_block() | |
71aaa3f7 | 1144 | |
e57a18e1 PP |
1145 | if item is not None: |
1146 | return item | |
1147 | ||
27d52a19 PP |
1148 | # Conditional block item? |
1149 | item = self._try_parse_cond_block() | |
1150 | ||
1151 | if item is not None: | |
1152 | return item | |
1153 | ||
e57a18e1 PP |
1154 | # Pattern for _try_parse_rep_post() |
1155 | _rep_post_prefix_pat = re.compile(r"\*") | |
1156 | ||
1157 | # Tries to parse a post-item repetition, returning the expression | |
1158 | # string and AST expression node on success. | |
1159 | def _try_parse_rep_post(self): | |
71aaa3f7 | 1160 | # Match prefix |
e57a18e1 | 1161 | if self._try_parse_pat(self._rep_post_prefix_pat) is None: |
71aaa3f7 | 1162 | # No match |
2adf4336 | 1163 | return |
71aaa3f7 | 1164 | |
e57a18e1 | 1165 | # Return expression string and AST expression |
71aaa3f7 | 1166 | self._skip_ws_and_comments() |
e57a18e1 | 1167 | return self._expect_rep_mul_expr() |
71aaa3f7 | 1168 | |
1ca7b5e1 PP |
1169 | # Tries to parse an item, possibly followed by a repetition, |
1170 | # returning `True` on success. | |
1171 | # | |
1172 | # Appends any parsed item to `items`. | |
1173 | def _try_append_item(self, items: List[_Item]): | |
71aaa3f7 PP |
1174 | self._skip_ws_and_comments() |
1175 | ||
1176 | # Parse a base item | |
1177 | item = self._try_parse_base_item() | |
1178 | ||
1179 | if item is None: | |
1180 | # No item | |
1ca7b5e1 | 1181 | return False |
71aaa3f7 PP |
1182 | |
1183 | # Parse repetition if the base item is repeatable | |
1184 | if isinstance(item, _RepableItem): | |
0e8e3169 PP |
1185 | self._skip_ws_and_comments() |
1186 | rep_text_loc = self._text_loc | |
e57a18e1 | 1187 | rep_ret = self._try_parse_rep_post() |
71aaa3f7 | 1188 | |
2adf4336 PP |
1189 | if rep_ret is not None: |
1190 | item = _Rep(item, rep_ret[0], rep_ret[1], rep_text_loc) | |
71aaa3f7 | 1191 | |
1ca7b5e1 PP |
1192 | items.append(item) |
1193 | return True | |
71aaa3f7 PP |
1194 | |
1195 | # Parses and returns items, skipping whitespaces, insignificant | |
1196 | # symbols, and comments when allowed, and stopping at the first | |
1197 | # unknown character. | |
1198 | def _parse_items(self) -> List[_Item]: | |
1199 | items = [] # type: List[_Item] | |
1200 | ||
1201 | while self._isnt_done(): | |
1ca7b5e1 PP |
1202 | # Try to append item |
1203 | if not self._try_append_item(items): | |
1204 | # Unknown at this point | |
1205 | break | |
71aaa3f7 PP |
1206 | |
1207 | return items | |
1208 | ||
1209 | # Parses the whole Normand input, setting `self._res` to the main | |
1210 | # group item on success. | |
1211 | def _parse(self): | |
1212 | if len(self._normand.strip()) == 0: | |
1213 | # Special case to make sure there's something to consume | |
1214 | self._res = _Group([], self._text_loc) | |
1215 | return | |
1216 | ||
1217 | # Parse first level items | |
1218 | items = self._parse_items() | |
1219 | ||
1220 | # Make sure there's nothing left | |
1221 | self._skip_ws_and_comments() | |
1222 | ||
1223 | if self._isnt_done(): | |
1224 | self._raise_error( | |
1225 | "Unexpected character `{}`".format(self._normand[self._at]) | |
1226 | ) | |
1227 | ||
1228 | # Set main group item | |
1229 | self._res = _Group(items, self._text_loc) | |
1230 | ||
1231 | ||
1232 | # The return type of parse(). | |
1233 | class ParseResult: | |
1234 | @classmethod | |
1235 | def _create( | |
1236 | cls, | |
1237 | data: bytearray, | |
e57a18e1 PP |
1238 | variables: VariablesT, |
1239 | labels: LabelsT, | |
71aaa3f7 PP |
1240 | offset: int, |
1241 | bo: Optional[ByteOrder], | |
1242 | ): | |
1243 | self = cls.__new__(cls) | |
1244 | self._init(data, variables, labels, offset, bo) | |
1245 | return self | |
1246 | ||
1247 | def __init__(self, *args, **kwargs): # type: ignore | |
1248 | raise NotImplementedError | |
1249 | ||
1250 | def _init( | |
1251 | self, | |
1252 | data: bytearray, | |
e57a18e1 PP |
1253 | variables: VariablesT, |
1254 | labels: LabelsT, | |
71aaa3f7 PP |
1255 | offset: int, |
1256 | bo: Optional[ByteOrder], | |
1257 | ): | |
1258 | self._data = data | |
1259 | self._vars = variables | |
1260 | self._labels = labels | |
1261 | self._offset = offset | |
1262 | self._bo = bo | |
1263 | ||
1264 | # Generated data. | |
1265 | @property | |
1266 | def data(self): | |
1267 | return self._data | |
1268 | ||
1269 | # Dictionary of updated variable names to their last computed value. | |
1270 | @property | |
1271 | def variables(self): | |
1272 | return self._vars | |
1273 | ||
1274 | # Dictionary of updated main group label names to their computed | |
1275 | # value. | |
1276 | @property | |
1277 | def labels(self): | |
1278 | return self._labels | |
1279 | ||
1280 | # Updated offset. | |
1281 | @property | |
1282 | def offset(self): | |
1283 | return self._offset | |
1284 | ||
1285 | # Updated byte order. | |
1286 | @property | |
1287 | def byte_order(self): | |
1288 | return self._bo | |
1289 | ||
1290 | ||
1291 | # Raises a parse error for the item `item`, creating it using the | |
1292 | # message `msg`. | |
1293 | def _raise_error_for_item(msg: str, item: _Item) -> NoReturn: | |
1294 | _raise_error(msg, item.text_loc) | |
1295 | ||
1296 | ||
1297 | # The `ICITTE` reserved name. | |
1298 | _icitte_name = "ICITTE" | |
1299 | ||
1300 | ||
2adf4336 PP |
1301 | # Base node visitor. |
1302 | # | |
1303 | # Calls the _visit_name() method for each name node which isn't the name | |
1304 | # of a call. | |
1305 | class _NodeVisitor(ast.NodeVisitor): | |
1306 | def __init__(self): | |
71aaa3f7 PP |
1307 | self._parent_is_call = False |
1308 | ||
1309 | def generic_visit(self, node: ast.AST): | |
1310 | if type(node) is ast.Call: | |
1311 | self._parent_is_call = True | |
1312 | elif type(node) is ast.Name and not self._parent_is_call: | |
2adf4336 | 1313 | self._visit_name(node.id) |
71aaa3f7 PP |
1314 | |
1315 | super().generic_visit(node) | |
1316 | self._parent_is_call = False | |
1317 | ||
2adf4336 PP |
1318 | @abc.abstractmethod |
1319 | def _visit_name(self, name: str): | |
1320 | ... | |
1321 | ||
71aaa3f7 | 1322 | |
2adf4336 PP |
1323 | # Expression validator: validates that all the names within the |
1324 | # expression are allowed. | |
1325 | class _ExprValidator(_NodeVisitor): | |
e57a18e1 | 1326 | def __init__(self, item: _ExprItemT, allowed_names: Set[str]): |
2adf4336 PP |
1327 | super().__init__() |
1328 | self._item = item | |
1329 | self._allowed_names = allowed_names | |
2adf4336 PP |
1330 | |
1331 | def _visit_name(self, name: str): | |
1332 | # Make sure the name refers to a known and reachable | |
1333 | # variable/label name. | |
e57a18e1 | 1334 | if name != _icitte_name and name not in self._allowed_names: |
2adf4336 PP |
1335 | msg = "Illegal (unknown or unreachable) variable/label name `{}` in expression `{}`".format( |
1336 | name, self._item.expr_str | |
1337 | ) | |
1338 | ||
05f81895 | 1339 | allowed_names = self._allowed_names.copy() |
e57a18e1 | 1340 | allowed_names.add(_icitte_name) |
2adf4336 | 1341 | |
05f81895 | 1342 | if len(allowed_names) > 0: |
2adf4336 PP |
1343 | allowed_names_str = ", ".join( |
1344 | sorted(["`{}`".format(name) for name in allowed_names]) | |
1345 | ) | |
1346 | msg += "; the legal names are {{{}}}".format(allowed_names_str) | |
1347 | ||
1348 | _raise_error( | |
1349 | msg, | |
1350 | self._item.text_loc, | |
1351 | ) | |
1352 | ||
1353 | ||
1354 | # Expression visitor getting all the contained names. | |
1355 | class _ExprNamesVisitor(_NodeVisitor): | |
71aaa3f7 | 1356 | def __init__(self): |
2adf4336 PP |
1357 | self._parent_is_call = False |
1358 | self._names = set() # type: Set[str] | |
1359 | ||
1360 | @property | |
1361 | def names(self): | |
1362 | return self._names | |
71aaa3f7 | 1363 | |
2adf4336 PP |
1364 | def _visit_name(self, name: str): |
1365 | self._names.add(name) | |
71aaa3f7 | 1366 | |
71aaa3f7 | 1367 | |
2adf4336 PP |
1368 | # Generator state. |
1369 | class _GenState: | |
1370 | def __init__( | |
1b8aa84a | 1371 | self, |
e57a18e1 PP |
1372 | variables: VariablesT, |
1373 | labels: LabelsT, | |
1b8aa84a PP |
1374 | offset: int, |
1375 | bo: Optional[ByteOrder], | |
2adf4336 PP |
1376 | ): |
1377 | self.variables = variables.copy() | |
1378 | self.labels = labels.copy() | |
1379 | self.offset = offset | |
1380 | self.bo = bo | |
71aaa3f7 PP |
1381 | |
1382 | ||
2adf4336 | 1383 | # Generator of data and final state from a group item. |
71aaa3f7 PP |
1384 | # |
1385 | # Generation happens in memory at construction time. After building, use | |
1386 | # the `data`, `variables`, `labels`, `offset`, and `bo` properties to | |
1387 | # get the resulting context. | |
2adf4336 PP |
1388 | # |
1389 | # The steps of generation are: | |
1390 | # | |
27d52a19 PP |
1391 | # 1. Validate that each repetition, conditional, and LEB128 integer |
1392 | # expression uses only reachable names. | |
2adf4336 | 1393 | # |
27d52a19 PP |
1394 | # 2. Compute and keep the effective repetition count, conditional value, |
1395 | # and LEB128 integer value for each repetition and LEB128 integer | |
1396 | # instance. | |
2adf4336 PP |
1397 | # |
1398 | # 3. Generate bytes, updating the initial state as it goes which becomes | |
1399 | # the final state after the operation. | |
1400 | # | |
27d52a19 PP |
1401 | # During the generation, when handling a `_Rep`, `_Cond`, or |
1402 | # `_Leb128Int` item, we already have the effective repetition count, | |
1403 | # conditional value, or value of the instance. | |
2adf4336 PP |
1404 | # |
1405 | # When handling a `_Group` item, first update the current labels with | |
1406 | # all the immediate (not nested) labels, and then handle each | |
1407 | # contained item. This gives contained item access to "future" outer | |
1408 | # labels. Then remove the immediate labels from the state so that | |
1409 | # outer items don't have access to inner labels. | |
71aaa3f7 PP |
1410 | class _Gen: |
1411 | def __init__( | |
1412 | self, | |
1413 | group: _Group, | |
e57a18e1 PP |
1414 | variables: VariablesT, |
1415 | labels: LabelsT, | |
71aaa3f7 PP |
1416 | offset: int, |
1417 | bo: Optional[ByteOrder], | |
1418 | ): | |
05f81895 PP |
1419 | self._validate_vl_exprs(group, set(variables.keys()), set(labels.keys())) |
1420 | self._vl_instance_vals = self._compute_vl_instance_vals( | |
2adf4336 PP |
1421 | group, _GenState(variables, labels, offset, bo) |
1422 | ) | |
1423 | self._gen(group, _GenState(variables, labels, offset, bo)) | |
71aaa3f7 PP |
1424 | |
1425 | # Generated bytes. | |
1426 | @property | |
1427 | def data(self): | |
1428 | return self._data | |
1429 | ||
1430 | # Updated variables. | |
1431 | @property | |
1432 | def variables(self): | |
2adf4336 | 1433 | return self._final_state.variables |
71aaa3f7 PP |
1434 | |
1435 | # Updated main group labels. | |
1436 | @property | |
1437 | def labels(self): | |
2adf4336 | 1438 | return self._final_state.labels |
71aaa3f7 PP |
1439 | |
1440 | # Updated offset. | |
1441 | @property | |
1442 | def offset(self): | |
2adf4336 | 1443 | return self._final_state.offset |
71aaa3f7 PP |
1444 | |
1445 | # Updated byte order. | |
1446 | @property | |
1447 | def bo(self): | |
2adf4336 PP |
1448 | return self._final_state.bo |
1449 | ||
1450 | # Returns the set of used, non-called names within the AST | |
1451 | # expression `expr`. | |
1452 | @staticmethod | |
1453 | def _names_of_expr(expr: ast.Expression): | |
1454 | visitor = _ExprNamesVisitor() | |
1455 | visitor.visit(expr) | |
1456 | return visitor.names | |
1457 | ||
27d52a19 PP |
1458 | # Validates that all the repetition, conditional, and LEB128 integer |
1459 | # expressions within `group` don't refer, directly or indirectly, to | |
1460 | # subsequent labels. | |
71aaa3f7 | 1461 | # |
2adf4336 PP |
1462 | # The strategy here is to keep a set of allowed label names, per |
1463 | # group, initialized to `allowed_label_names`, and a set of allowed | |
1464 | # variable names initialized to `allowed_variable_names`. | |
1465 | # | |
1466 | # Then, depending on the type of `item`: | |
1467 | # | |
1468 | # `_Label`: | |
1469 | # Add its name to the local allowed label names: a label | |
1470 | # occurring before a repetition, and not within a nested group, | |
1471 | # is always reachable. | |
1472 | # | |
1473 | # `_VarAssign`: | |
1474 | # If all the names within its expression are allowed, then add | |
1475 | # its name to the allowed variable names. | |
1476 | # | |
1477 | # Otherwise, remove its name from the allowed variable names (if | |
1478 | # it's in there): a variable which refers to an unreachable name | |
1479 | # is unreachable itself. | |
1480 | # | |
27d52a19 | 1481 | # `_Rep`, `_Cond`, and `_Leb128`: |
2adf4336 PP |
1482 | # Make sure all the names within its expression are allowed. |
1483 | # | |
1484 | # `_Group`: | |
1485 | # Call this function for each contained item with a _copy_ of | |
1486 | # the current allowed label names and the same current allowed | |
1487 | # variable names. | |
1488 | @staticmethod | |
05f81895 | 1489 | def _validate_vl_exprs( |
2adf4336 PP |
1490 | item: _Item, allowed_variable_names: Set[str], allowed_label_names: Set[str] |
1491 | ): | |
1492 | if type(item) is _Label: | |
1493 | allowed_label_names.add(item.name) | |
1494 | elif type(item) is _VarAssign: | |
1495 | # Check if this variable name is allowed | |
1496 | allowed = True | |
1497 | ||
1498 | for name in _Gen._names_of_expr(item.expr): | |
1499 | if name not in ( | |
1500 | allowed_label_names | allowed_variable_names | {_icitte_name} | |
1501 | ): | |
1502 | # Not allowed | |
1503 | allowed = False | |
1504 | break | |
1505 | ||
1506 | if allowed: | |
1507 | allowed_variable_names.add(item.name) | |
1508 | elif item.name in allowed_variable_names: | |
1509 | allowed_variable_names.remove(item.name) | |
05f81895 | 1510 | elif isinstance(item, _Leb128Int): |
e57a18e1 PP |
1511 | # Validate the expression |
1512 | _ExprValidator(item, allowed_label_names | allowed_variable_names).visit( | |
1513 | item.expr | |
1514 | ) | |
27d52a19 | 1515 | elif type(item) is _Rep or type(item) is _Cond: |
e57a18e1 PP |
1516 | # Validate the expression first |
1517 | _ExprValidator(item, allowed_label_names | allowed_variable_names).visit( | |
1518 | item.expr | |
1519 | ) | |
2adf4336 PP |
1520 | |
1521 | # Validate inner item | |
05f81895 | 1522 | _Gen._validate_vl_exprs( |
2adf4336 PP |
1523 | item.item, allowed_variable_names, allowed_label_names |
1524 | ) | |
1525 | elif type(item) is _Group: | |
1526 | # Copy `allowed_label_names` so that this frame cannot | |
1527 | # access the nested label names. | |
1528 | group_allowed_label_names = allowed_label_names.copy() | |
71aaa3f7 | 1529 | |
2adf4336 | 1530 | for subitem in item.items: |
05f81895 | 1531 | _Gen._validate_vl_exprs( |
2adf4336 PP |
1532 | subitem, allowed_variable_names, group_allowed_label_names |
1533 | ) | |
71aaa3f7 | 1534 | |
2adf4336 PP |
1535 | # Evaluates the expression of `item` considering the current |
1536 | # generation state `state`. | |
1537 | # | |
269f6eb3 PP |
1538 | # If `allow_float` is `True`, then the type of the result may be |
1539 | # `float` too. | |
2adf4336 | 1540 | @staticmethod |
269f6eb3 PP |
1541 | def _eval_item_expr( |
1542 | item: _ExprItemT, | |
1543 | state: _GenState, | |
269f6eb3 PP |
1544 | allow_float: bool = False, |
1545 | ): | |
e57a18e1 PP |
1546 | syms = {} # type: VariablesT |
1547 | syms.update(state.labels) | |
71aaa3f7 | 1548 | |
e57a18e1 PP |
1549 | # Set the `ICITTE` name to the current offset |
1550 | syms[_icitte_name] = state.offset | |
71aaa3f7 PP |
1551 | |
1552 | # Add the current variables | |
2adf4336 | 1553 | syms.update(state.variables) |
71aaa3f7 PP |
1554 | |
1555 | # Validate the node and its children | |
e57a18e1 | 1556 | _ExprValidator(item, set(syms.keys())).visit(item.expr) |
71aaa3f7 PP |
1557 | |
1558 | # Compile and evaluate expression node | |
1559 | try: | |
1560 | val = eval(compile(item.expr, "", "eval"), None, syms) | |
1561 | except Exception as exc: | |
1562 | _raise_error_for_item( | |
1563 | "Failed to evaluate expression `{}`: {}".format(item.expr_str, exc), | |
1564 | item, | |
1565 | ) | |
1566 | ||
27d52a19 PP |
1567 | # Convert `bool` result type to `int` to normalize |
1568 | if type(val) is bool: | |
1569 | val = int(val) | |
1570 | ||
269f6eb3 PP |
1571 | # Validate result type |
1572 | expected_types = {int} # type: Set[type] | |
1573 | type_msg = "`int`" | |
1574 | ||
1575 | if allow_float: | |
1576 | expected_types.add(float) | |
1577 | type_msg += " or `float`" | |
1578 | ||
1579 | if type(val) not in expected_types: | |
71aaa3f7 | 1580 | _raise_error_for_item( |
269f6eb3 PP |
1581 | "Invalid expression `{}`: expecting result type {}, not `{}`".format( |
1582 | item.expr_str, type_msg, type(val).__name__ | |
71aaa3f7 PP |
1583 | ), |
1584 | item, | |
1585 | ) | |
1586 | ||
1587 | return val | |
1588 | ||
05f81895 PP |
1589 | # Returns the size, in bytes, required to encode the value `val` |
1590 | # with LEB128 (signed version if `is_signed` is `True`). | |
1591 | @staticmethod | |
1592 | def _leb128_size_for_val(val: int, is_signed: bool): | |
1593 | if val < 0: | |
1594 | # Equivalent upper bound. | |
1595 | # | |
1596 | # For example, if `val` is -128, then the full integer for | |
1597 | # this number of bits would be [-128, 127]. | |
1598 | val = -val - 1 | |
1599 | ||
1600 | # Number of bits (add one for the sign if needed) | |
1601 | bits = val.bit_length() + int(is_signed) | |
1602 | ||
1603 | if bits == 0: | |
1604 | bits = 1 | |
1605 | ||
1606 | # Seven bits per byte | |
1607 | return math.ceil(bits / 7) | |
1608 | ||
676f6189 PP |
1609 | # Returns the offset `offset` aligned according to `item`. |
1610 | @staticmethod | |
1611 | def _align_offset(offset: int, item: _AlignOffset): | |
1612 | align_bytes = item.val // 8 | |
1613 | return (offset + align_bytes - 1) // align_bytes * align_bytes | |
1614 | ||
27d52a19 PP |
1615 | # Computes the effective value for each repetition, conditional, and |
1616 | # LEB128 integer instance, filling `instance_vals` (if not `None`) | |
1617 | # and returning `instance_vals`. | |
2adf4336 | 1618 | # |
05f81895 PP |
1619 | # At this point it must be known that, for a given variable-length |
1620 | # item, its expression only contains reachable names. | |
2adf4336 | 1621 | # |
27d52a19 PP |
1622 | # When handling a `_Rep` or `_Cond` item, this function appends its |
1623 | # effective multiplier/value to `instance_vals` _before_ handling | |
1624 | # its repeated/conditional item. | |
2adf4336 | 1625 | # |
05f81895 PP |
1626 | # When handling a `_VarAssign` item, this function only evaluates it |
1627 | # if all its names are reachable. | |
2adf4336 | 1628 | @staticmethod |
05f81895 | 1629 | def _compute_vl_instance_vals( |
2adf4336 PP |
1630 | item: _Item, state: _GenState, instance_vals: Optional[List[int]] = None |
1631 | ): | |
1632 | if instance_vals is None: | |
1633 | instance_vals = [] | |
1634 | ||
1635 | if isinstance(item, _ScalarItem): | |
1636 | state.offset += item.size | |
1637 | elif type(item) is _Label: | |
1638 | state.labels[item.name] = state.offset | |
1639 | elif type(item) is _VarAssign: | |
1640 | # Check if all the names are reachable | |
1641 | do_eval = True | |
1642 | ||
1643 | for name in _Gen._names_of_expr(item.expr): | |
1644 | if ( | |
1645 | name != _icitte_name | |
1646 | and name not in state.variables | |
1647 | and name not in state.labels | |
1648 | ): | |
1649 | # A name is unknown: cannot evaluate | |
1650 | do_eval = False | |
1651 | break | |
1652 | ||
1653 | if do_eval: | |
1654 | # Evaluate the expression and keep the result | |
e57a18e1 | 1655 | state.variables[item.name] = _Gen._eval_item_expr(item, state, True) |
2adf4336 PP |
1656 | elif type(item) is _SetOffset: |
1657 | state.offset = item.val | |
676f6189 PP |
1658 | elif type(item) is _AlignOffset: |
1659 | state.offset = _Gen._align_offset(state.offset, item) | |
05f81895 PP |
1660 | elif isinstance(item, _Leb128Int): |
1661 | # Evaluate the expression | |
e57a18e1 | 1662 | val = _Gen._eval_item_expr(item, state) |
05f81895 PP |
1663 | |
1664 | # Validate result | |
1665 | if type(item) is _ULeb128Int and val < 0: | |
1666 | _raise_error_for_item( | |
1667 | "Invalid expression `{}`: unexpected negative result {:,} for a ULEB128 encoding".format( | |
1668 | item.expr_str, val | |
1669 | ), | |
1670 | item, | |
1671 | ) | |
1672 | ||
1673 | # Add the evaluation result to the to variable-length item | |
1674 | # instance values. | |
1675 | instance_vals.append(val) | |
1676 | ||
1677 | # Update offset | |
1678 | state.offset += _Gen._leb128_size_for_val(val, type(item) is _SLeb128Int) | |
2adf4336 PP |
1679 | elif type(item) is _Rep: |
1680 | # Evaluate the expression and keep the result | |
e57a18e1 | 1681 | val = _Gen._eval_item_expr(item, state) |
2adf4336 PP |
1682 | |
1683 | # Validate result | |
1684 | if val < 0: | |
1685 | _raise_error_for_item( | |
1686 | "Invalid expression `{}`: unexpected negative result {:,}".format( | |
1687 | item.expr_str, val | |
1688 | ), | |
1689 | item, | |
1690 | ) | |
1691 | ||
27d52a19 | 1692 | # Add to variable-length item instance values |
2adf4336 PP |
1693 | instance_vals.append(val) |
1694 | ||
1695 | # Process the repeated item `val` times | |
1696 | for _ in range(val): | |
05f81895 | 1697 | _Gen._compute_vl_instance_vals(item.item, state, instance_vals) |
27d52a19 PP |
1698 | elif type(item) is _Cond: |
1699 | # Evaluate the expression and keep the result | |
1700 | val = _Gen._eval_item_expr(item, state) | |
1701 | ||
1702 | # Add to variable-length item instance values | |
1703 | instance_vals.append(val) | |
1704 | ||
1705 | # Process the conditional item if needed | |
1706 | if val: | |
1707 | _Gen._compute_vl_instance_vals(item.item, state, instance_vals) | |
2adf4336 PP |
1708 | elif type(item) is _Group: |
1709 | prev_labels = state.labels.copy() | |
1710 | ||
1711 | # Process each item | |
1712 | for subitem in item.items: | |
05f81895 | 1713 | _Gen._compute_vl_instance_vals(subitem, state, instance_vals) |
2adf4336 PP |
1714 | |
1715 | state.labels = prev_labels | |
1716 | ||
1717 | return instance_vals | |
1718 | ||
676f6189 PP |
1719 | def _update_offset_noop(self, item: _Item, state: _GenState, next_vl_instance: int): |
1720 | return next_vl_instance | |
05f81895 | 1721 | |
676f6189 PP |
1722 | def _dry_handle_scalar_item( |
1723 | self, item: _ScalarItem, state: _GenState, next_vl_instance: int | |
1724 | ): | |
1725 | state.offset += item.size | |
1726 | return next_vl_instance | |
2adf4336 | 1727 | |
676f6189 PP |
1728 | def _dry_handle_leb128_int_item( |
1729 | self, item: _Leb128Int, state: _GenState, next_vl_instance: int | |
1730 | ): | |
05f81895 PP |
1731 | # Get the value from `self._vl_instance_vals` _before_ |
1732 | # incrementing `next_vl_instance` to honor the order of | |
1733 | # _compute_vl_instance_vals(). | |
676f6189 PP |
1734 | state.offset += self._leb128_size_for_val( |
1735 | self._vl_instance_vals[next_vl_instance], type(item) is _SLeb128Int | |
05f81895 | 1736 | ) |
2adf4336 | 1737 | |
676f6189 | 1738 | return next_vl_instance + 1 |
2adf4336 | 1739 | |
676f6189 PP |
1740 | def _dry_handle_group_item( |
1741 | self, item: _Group, state: _GenState, next_vl_instance: int | |
1742 | ): | |
2adf4336 | 1743 | for subitem in item.items: |
676f6189 | 1744 | next_vl_instance = self._dry_handle_item(subitem, state, next_vl_instance) |
2adf4336 | 1745 | |
676f6189 | 1746 | return next_vl_instance |
2adf4336 | 1747 | |
676f6189 | 1748 | def _dry_handle_rep_item(self, item: _Rep, state: _GenState, next_vl_instance: int): |
05f81895 PP |
1749 | # Get the value from `self._vl_instance_vals` _before_ |
1750 | # incrementing `next_vl_instance` to honor the order of | |
1751 | # _compute_vl_instance_vals(). | |
1752 | mul = self._vl_instance_vals[next_vl_instance] | |
1753 | next_vl_instance += 1 | |
2adf4336 PP |
1754 | |
1755 | for _ in range(mul): | |
676f6189 | 1756 | next_vl_instance = self._dry_handle_item(item.item, state, next_vl_instance) |
2adf4336 | 1757 | |
676f6189 | 1758 | return next_vl_instance |
2adf4336 | 1759 | |
27d52a19 PP |
1760 | def _dry_handle_cond_item( |
1761 | self, item: _Cond, state: _GenState, next_vl_instance: int | |
1762 | ): | |
1763 | # Get the value from `self._vl_instance_vals` _before_ | |
1764 | # incrementing `next_vl_instance` to honor the order of | |
1765 | # _compute_vl_instance_vals(). | |
1766 | val = self._vl_instance_vals[next_vl_instance] | |
1767 | next_vl_instance += 1 | |
1768 | ||
1769 | if val: | |
1770 | next_vl_instance = self._dry_handle_item(item.item, state, next_vl_instance) | |
1771 | ||
1772 | return next_vl_instance | |
1773 | ||
676f6189 PP |
1774 | def _dry_handle_align_offset_item( |
1775 | self, item: _AlignOffset, state: _GenState, next_vl_instance: int | |
1776 | ): | |
1777 | state.offset = self._align_offset(state.offset, item) | |
1778 | return next_vl_instance | |
1779 | ||
1780 | def _dry_handle_set_offset_item( | |
1781 | self, item: _SetOffset, state: _GenState, next_vl_instance: int | |
1782 | ): | |
1783 | state.offset = item.val | |
1784 | return next_vl_instance | |
1785 | ||
1786 | # Updates `state.offset` considering the generated data of `item`, | |
1787 | # without generating any, and returns the updated next | |
1788 | # variable-length item instance. | |
1789 | def _dry_handle_item(self, item: _Item, state: _GenState, next_vl_instance: int): | |
1790 | return self._dry_handle_item_funcs[type(item)](item, state, next_vl_instance) | |
2adf4336 PP |
1791 | |
1792 | # Handles the byte item `item`. | |
05f81895 | 1793 | def _handle_byte_item(self, item: _Byte, state: _GenState, next_vl_instance: int): |
2adf4336 PP |
1794 | self._data.append(item.val) |
1795 | state.offset += item.size | |
05f81895 | 1796 | return next_vl_instance |
2adf4336 PP |
1797 | |
1798 | # Handles the string item `item`. | |
05f81895 | 1799 | def _handle_str_item(self, item: _Str, state: _GenState, next_vl_instance: int): |
2adf4336 PP |
1800 | self._data += item.data |
1801 | state.offset += item.size | |
05f81895 | 1802 | return next_vl_instance |
2adf4336 PP |
1803 | |
1804 | # Handles the byte order setting item `item`. | |
1805 | def _handle_set_bo_item( | |
05f81895 | 1806 | self, item: _SetBo, state: _GenState, next_vl_instance: int |
2adf4336 PP |
1807 | ): |
1808 | # Update current byte order | |
1809 | state.bo = item.bo | |
05f81895 | 1810 | return next_vl_instance |
2adf4336 PP |
1811 | |
1812 | # Handles the variable assignment item `item`. | |
1813 | def _handle_var_assign_item( | |
05f81895 | 1814 | self, item: _VarAssign, state: _GenState, next_vl_instance: int |
2adf4336 | 1815 | ): |
71aaa3f7 | 1816 | # Update variable |
e57a18e1 | 1817 | state.variables[item.name] = self._eval_item_expr(item, state, True) |
05f81895 | 1818 | return next_vl_instance |
71aaa3f7 | 1819 | |
05f81895 | 1820 | # Handles the fixed-length integer item `item`. |
269f6eb3 | 1821 | def _handle_fl_int_item(self, val: int, item: _FlNum, state: _GenState): |
71aaa3f7 PP |
1822 | # Validate range |
1823 | if val < -(2 ** (item.len - 1)) or val > 2**item.len - 1: | |
1824 | _raise_error_for_item( | |
1825 | "Value {:,} is outside the {}-bit range when evaluating expression `{}` at byte offset {:,}".format( | |
2adf4336 | 1826 | val, item.len, item.expr_str, state.offset |
71aaa3f7 PP |
1827 | ), |
1828 | item, | |
1829 | ) | |
1830 | ||
1831 | # Encode result on 64 bits (to extend the sign bit whatever the | |
1832 | # value of `item.len`). | |
71aaa3f7 PP |
1833 | data = struct.pack( |
1834 | "{}{}".format( | |
2adf4336 | 1835 | ">" if state.bo in (None, ByteOrder.BE) else "<", |
71aaa3f7 PP |
1836 | "Q" if val >= 0 else "q", |
1837 | ), | |
1838 | val, | |
1839 | ) | |
1840 | ||
1841 | # Keep only the requested length | |
1842 | len_bytes = item.len // 8 | |
1843 | ||
2adf4336 | 1844 | if state.bo in (None, ByteOrder.BE): |
71aaa3f7 PP |
1845 | # Big endian: keep last bytes |
1846 | data = data[-len_bytes:] | |
1847 | else: | |
1848 | # Little endian: keep first bytes | |
2adf4336 | 1849 | assert state.bo == ByteOrder.LE |
71aaa3f7 PP |
1850 | data = data[:len_bytes] |
1851 | ||
1852 | # Append to current bytes and update offset | |
1853 | self._data += data | |
269f6eb3 PP |
1854 | |
1855 | # Handles the fixed-length integer item `item`. | |
1856 | def _handle_fl_float_item(self, val: float, item: _FlNum, state: _GenState): | |
1857 | # Validate length | |
1858 | if item.len not in (32, 64): | |
1859 | _raise_error_for_item( | |
1860 | "Invalid {}-bit length for a fixed-length floating point number (value {:,})".format( | |
1861 | item.len, val | |
1862 | ), | |
1863 | item, | |
1864 | ) | |
1865 | ||
1866 | # Encode result | |
1867 | self._data += struct.pack( | |
1868 | "{}{}".format( | |
1869 | ">" if state.bo in (None, ByteOrder.BE) else "<", | |
1870 | "f" if item.len == 32 else "d", | |
1871 | ), | |
1872 | val, | |
1873 | ) | |
1874 | ||
1875 | # Handles the fixed-length number item `item`. | |
1876 | def _handle_fl_num_item( | |
1877 | self, item: _FlNum, state: _GenState, next_vl_instance: int | |
1878 | ): | |
1879 | # Compute value | |
e57a18e1 | 1880 | val = self._eval_item_expr(item, state, True) |
269f6eb3 PP |
1881 | |
1882 | # Validate current byte order | |
1883 | if state.bo is None and item.len > 8: | |
1884 | _raise_error_for_item( | |
1885 | "Current byte order isn't defined at first fixed-length number (`{}`) to encode on more than 8 bits".format( | |
1886 | item.expr_str | |
1887 | ), | |
1888 | item, | |
1889 | ) | |
1890 | ||
1891 | # Handle depending on type | |
1892 | if type(val) is int: | |
1893 | self._handle_fl_int_item(val, item, state) | |
1894 | else: | |
1895 | assert type(val) is float | |
1896 | self._handle_fl_float_item(val, item, state) | |
1897 | ||
1898 | # Update offset | |
1899 | state.offset += item.size | |
1900 | ||
05f81895 PP |
1901 | return next_vl_instance |
1902 | ||
1903 | # Handles the LEB128 integer item `item`. | |
1904 | def _handle_leb128_int_item( | |
1905 | self, item: _Leb128Int, state: _GenState, next_vl_instance: int | |
1906 | ): | |
1907 | # Get the precomputed value | |
1908 | val = self._vl_instance_vals[next_vl_instance] | |
1909 | ||
1910 | # Size in bytes | |
1911 | size = self._leb128_size_for_val(val, type(item) is _SLeb128Int) | |
1912 | ||
1913 | # For each byte | |
1914 | for _ in range(size): | |
1915 | # Seven LSBs, MSB of the byte set (continue) | |
1916 | self._data.append((val & 0x7F) | 0x80) | |
1917 | val >>= 7 | |
1918 | ||
1919 | # Clear MSB of last byte (stop) | |
1920 | self._data[-1] &= ~0x80 | |
1921 | ||
1922 | # Consumed this instance | |
1923 | return next_vl_instance + 1 | |
71aaa3f7 | 1924 | |
2adf4336 PP |
1925 | # Handles the group item `item`, only removing the immediate labels |
1926 | # from `state.labels` if `remove_immediate_labels` is `True`. | |
1927 | def _handle_group_item( | |
1928 | self, | |
1929 | item: _Group, | |
1930 | state: _GenState, | |
05f81895 | 1931 | next_vl_instance: int, |
2adf4336 PP |
1932 | remove_immediate_labels: bool = True, |
1933 | ): | |
1934 | # Compute the values of the immediate (not nested) labels. Those | |
1935 | # labels are reachable by any expression within the group. | |
676f6189 | 1936 | tmp_state = _GenState({}, {}, state.offset, None) |
2adf4336 | 1937 | immediate_label_names = set() # type: Set[str] |
05f81895 | 1938 | tmp_next_vl_instance = next_vl_instance |
71aaa3f7 | 1939 | |
2adf4336 | 1940 | for subitem in item.items: |
676f6189 | 1941 | if type(subitem) is _Label: |
2adf4336 | 1942 | # New immediate label |
676f6189 | 1943 | state.labels[subitem.name] = tmp_state.offset |
2adf4336 PP |
1944 | immediate_label_names.add(subitem.name) |
1945 | ||
676f6189 PP |
1946 | tmp_next_vl_instance = self._dry_handle_item( |
1947 | subitem, tmp_state, tmp_next_vl_instance | |
2adf4336 | 1948 | ) |
71aaa3f7 | 1949 | |
2adf4336 | 1950 | # Handle each item now with the actual state |
71aaa3f7 | 1951 | for subitem in item.items: |
05f81895 | 1952 | next_vl_instance = self._handle_item(subitem, state, next_vl_instance) |
2adf4336 PP |
1953 | |
1954 | # Remove immediate labels if required so that outer items won't | |
1955 | # reach inner labels. | |
1956 | if remove_immediate_labels: | |
1957 | for name in immediate_label_names: | |
1958 | del state.labels[name] | |
71aaa3f7 | 1959 | |
05f81895 | 1960 | return next_vl_instance |
71aaa3f7 | 1961 | |
2adf4336 | 1962 | # Handles the repetition item `item`. |
05f81895 PP |
1963 | def _handle_rep_item(self, item: _Rep, state: _GenState, next_vl_instance: int): |
1964 | # Get the precomputed repetition count | |
1965 | mul = self._vl_instance_vals[next_vl_instance] | |
1966 | ||
1967 | # Consumed this instance | |
1968 | next_vl_instance += 1 | |
71aaa3f7 | 1969 | |
2adf4336 | 1970 | for _ in range(mul): |
05f81895 | 1971 | next_vl_instance = self._handle_item(item.item, state, next_vl_instance) |
71aaa3f7 | 1972 | |
05f81895 | 1973 | return next_vl_instance |
71aaa3f7 | 1974 | |
27d52a19 PP |
1975 | # Handles the conditional item `item`. |
1976 | def _handle_cond_item(self, item: _Rep, state: _GenState, next_vl_instance: int): | |
1977 | # Get the precomputed conditional value | |
1978 | val = self._vl_instance_vals[next_vl_instance] | |
1979 | ||
1980 | # Consumed this instance | |
1981 | next_vl_instance += 1 | |
1982 | ||
1983 | if val: | |
1984 | next_vl_instance = self._handle_item(item.item, state, next_vl_instance) | |
1985 | ||
1986 | return next_vl_instance | |
1987 | ||
2adf4336 PP |
1988 | # Handles the offset setting item `item`. |
1989 | def _handle_set_offset_item( | |
05f81895 | 1990 | self, item: _SetOffset, state: _GenState, next_vl_instance: int |
2adf4336 PP |
1991 | ): |
1992 | state.offset = item.val | |
05f81895 | 1993 | return next_vl_instance |
2adf4336 | 1994 | |
676f6189 PP |
1995 | # Handles offset alignment item `item` (adds padding). |
1996 | def _handle_align_offset_item( | |
1997 | self, item: _AlignOffset, state: _GenState, next_vl_instance: int | |
1998 | ): | |
1999 | init_offset = state.offset | |
2000 | state.offset = self._align_offset(state.offset, item) | |
2001 | self._data += bytes([item.pad_val] * (state.offset - init_offset)) | |
2002 | return next_vl_instance | |
2003 | ||
2adf4336 | 2004 | # Handles the label item `item`. |
05f81895 PP |
2005 | def _handle_label_item(self, item: _Label, state: _GenState, next_vl_instance: int): |
2006 | return next_vl_instance | |
2adf4336 PP |
2007 | |
2008 | # Handles the item `item`, returning the updated next repetition | |
2009 | # instance. | |
05f81895 PP |
2010 | def _handle_item(self, item: _Item, state: _GenState, next_vl_instance: int): |
2011 | return self._item_handlers[type(item)](item, state, next_vl_instance) | |
2adf4336 PP |
2012 | |
2013 | # Generates the data (`self._data`) and final state | |
2014 | # (`self._final_state`) from `group` and the initial state `state`. | |
2015 | def _gen(self, group: _Group, state: _GenState): | |
71aaa3f7 PP |
2016 | # Initial state |
2017 | self._data = bytearray() | |
71aaa3f7 PP |
2018 | |
2019 | # Item handlers | |
2020 | self._item_handlers = { | |
676f6189 | 2021 | _AlignOffset: self._handle_align_offset_item, |
71aaa3f7 | 2022 | _Byte: self._handle_byte_item, |
27d52a19 | 2023 | _Cond: self._handle_cond_item, |
269f6eb3 | 2024 | _FlNum: self._handle_fl_num_item, |
71aaa3f7 | 2025 | _Group: self._handle_group_item, |
2adf4336 | 2026 | _Label: self._handle_label_item, |
71aaa3f7 | 2027 | _Rep: self._handle_rep_item, |
2adf4336 PP |
2028 | _SetBo: self._handle_set_bo_item, |
2029 | _SetOffset: self._handle_set_offset_item, | |
05f81895 | 2030 | _SLeb128Int: self._handle_leb128_int_item, |
2adf4336 | 2031 | _Str: self._handle_str_item, |
05f81895 | 2032 | _ULeb128Int: self._handle_leb128_int_item, |
2adf4336 PP |
2033 | _VarAssign: self._handle_var_assign_item, |
2034 | } # type: Dict[type, Callable[[Any, _GenState, int], int]] | |
2035 | ||
676f6189 PP |
2036 | # Dry item handlers (only updates the state offset) |
2037 | self._dry_handle_item_funcs = { | |
2038 | _AlignOffset: self._dry_handle_align_offset_item, | |
2039 | _Byte: self._dry_handle_scalar_item, | |
27d52a19 | 2040 | _Cond: self._dry_handle_cond_item, |
676f6189 PP |
2041 | _FlNum: self._dry_handle_scalar_item, |
2042 | _Group: self._dry_handle_group_item, | |
2043 | _Label: self._update_offset_noop, | |
2044 | _Rep: self._dry_handle_rep_item, | |
2045 | _SetBo: self._update_offset_noop, | |
2046 | _SetOffset: self._dry_handle_set_offset_item, | |
2047 | _SLeb128Int: self._dry_handle_leb128_int_item, | |
2048 | _Str: self._dry_handle_scalar_item, | |
2049 | _ULeb128Int: self._dry_handle_leb128_int_item, | |
2050 | _VarAssign: self._update_offset_noop, | |
2051 | } # type: Dict[type, Callable[[Any, _GenState, int], int]] | |
2adf4336 PP |
2052 | |
2053 | # Handle the group item, _not_ removing the immediate labels | |
2054 | # because the `labels` property offers them. | |
2055 | self._handle_group_item(group, state, 0, False) | |
2056 | ||
2057 | # This is actually the final state | |
2058 | self._final_state = state | |
71aaa3f7 PP |
2059 | |
2060 | ||
2061 | # Returns a `ParseResult` instance containing the bytes encoded by the | |
2062 | # input string `normand`. | |
2063 | # | |
2064 | # `init_variables` is a dictionary of initial variable names (valid | |
2065 | # Python names) to integral values. A variable name must not be the | |
2066 | # reserved name `ICITTE`. | |
2067 | # | |
2068 | # `init_labels` is a dictionary of initial label names (valid Python | |
2069 | # names) to integral values. A label name must not be the reserved name | |
2070 | # `ICITTE`. | |
2071 | # | |
2072 | # `init_offset` is the initial offset. | |
2073 | # | |
2074 | # `init_byte_order` is the initial byte order. | |
2075 | # | |
2076 | # Raises `ParseError` on any parsing error. | |
2077 | def parse( | |
2078 | normand: str, | |
e57a18e1 PP |
2079 | init_variables: Optional[VariablesT] = None, |
2080 | init_labels: Optional[LabelsT] = None, | |
71aaa3f7 PP |
2081 | init_offset: int = 0, |
2082 | init_byte_order: Optional[ByteOrder] = None, | |
2083 | ): | |
2084 | if init_variables is None: | |
2085 | init_variables = {} | |
2086 | ||
2087 | if init_labels is None: | |
2088 | init_labels = {} | |
2089 | ||
2090 | gen = _Gen( | |
2091 | _Parser(normand, init_variables, init_labels).res, | |
2092 | init_variables, | |
2093 | init_labels, | |
2094 | init_offset, | |
2095 | init_byte_order, | |
2096 | ) | |
2097 | return ParseResult._create( # pyright: ignore[reportPrivateUsage] | |
2098 | gen.data, gen.variables, gen.labels, gen.offset, gen.bo | |
2099 | ) | |
2100 | ||
2101 | ||
2102 | # Parses the command-line arguments. | |
2103 | def _parse_cli_args(): | |
2104 | import argparse | |
2105 | ||
2106 | # Build parser | |
2107 | ap = argparse.ArgumentParser() | |
2108 | ap.add_argument( | |
2109 | "--offset", | |
2110 | metavar="OFFSET", | |
2111 | action="store", | |
2112 | type=int, | |
2113 | default=0, | |
2114 | help="initial offset (positive)", | |
2115 | ) | |
2116 | ap.add_argument( | |
2117 | "-b", | |
2118 | "--byte-order", | |
2119 | metavar="BO", | |
2120 | choices=["be", "le"], | |
2121 | type=str, | |
2122 | help="initial byte order (`be` or `le`)", | |
2123 | ) | |
2124 | ap.add_argument( | |
2125 | "--var", | |
2126 | metavar="NAME=VAL", | |
2127 | action="append", | |
2128 | help="add an initial variable (may be repeated)", | |
2129 | ) | |
2130 | ap.add_argument( | |
2131 | "-l", | |
2132 | "--label", | |
2133 | metavar="NAME=VAL", | |
2134 | action="append", | |
2135 | help="add an initial label (may be repeated)", | |
2136 | ) | |
2137 | ap.add_argument( | |
2138 | "--version", action="version", version="Normand {}".format(__version__) | |
2139 | ) | |
2140 | ap.add_argument( | |
2141 | "path", | |
2142 | metavar="PATH", | |
2143 | action="store", | |
2144 | nargs="?", | |
2145 | help="input path (none means standard input)", | |
2146 | ) | |
2147 | ||
2148 | # Parse | |
2149 | return ap.parse_args() | |
2150 | ||
2151 | ||
2152 | # Raises a command-line error with the message `msg`. | |
2153 | def _raise_cli_error(msg: str) -> NoReturn: | |
2154 | raise RuntimeError("Command-line error: {}".format(msg)) | |
2155 | ||
2156 | ||
2157 | # Returns a dictionary of string to integers from the list of strings | |
2158 | # `args` containing `NAME=VAL` entries. | |
2159 | def _dict_from_arg(args: Optional[List[str]]): | |
e57a18e1 | 2160 | d = {} # type: LabelsT |
71aaa3f7 PP |
2161 | |
2162 | if args is None: | |
2163 | return d | |
2164 | ||
2165 | for arg in args: | |
2166 | m = re.match(r"({})=(\d+)$".format(_py_name_pat.pattern), arg) | |
2167 | ||
2168 | if m is None: | |
2169 | _raise_cli_error("Invalid assignment {}".format(arg)) | |
2170 | ||
2e1c1acd PP |
2171 | d[m.group(1)] = int(m.group(2)) |
2172 | ||
71aaa3f7 PP |
2173 | return d |
2174 | ||
2175 | ||
2176 | # CLI entry point without exception handling. | |
2177 | def _try_run_cli(): | |
2178 | import os.path | |
2179 | ||
2180 | # Parse arguments | |
2181 | args = _parse_cli_args() | |
2182 | ||
2183 | # Read input | |
2184 | if args.path is None: | |
2185 | normand = sys.stdin.read() | |
2186 | else: | |
2187 | with open(args.path) as f: | |
2188 | normand = f.read() | |
2189 | ||
2190 | # Variables and labels | |
e57a18e1 | 2191 | variables = typing.cast(VariablesT, _dict_from_arg(args.var)) |
71aaa3f7 PP |
2192 | labels = _dict_from_arg(args.label) |
2193 | ||
2194 | # Validate offset | |
2195 | if args.offset < 0: | |
2196 | _raise_cli_error("Invalid negative offset {}") | |
2197 | ||
2198 | # Validate and set byte order | |
2199 | bo = None # type: Optional[ByteOrder] | |
2200 | ||
2201 | if args.byte_order is not None: | |
2202 | if args.byte_order == "be": | |
2203 | bo = ByteOrder.BE | |
2204 | else: | |
2205 | assert args.byte_order == "le" | |
2206 | bo = ByteOrder.LE | |
2207 | ||
2208 | # Parse | |
2209 | try: | |
2210 | res = parse(normand, variables, labels, args.offset, bo) | |
2211 | except ParseError as exc: | |
2212 | prefix = "" | |
2213 | ||
2214 | if args.path is not None: | |
2215 | prefix = "{}:".format(os.path.abspath(args.path)) | |
2216 | ||
2217 | _fail( | |
2218 | "{}{}:{} - {}".format( | |
2219 | prefix, exc.text_loc.line_no, exc.text_loc.col_no, str(exc) | |
2220 | ) | |
2221 | ) | |
2222 | ||
2223 | ||
2224 | sys.stdout.buffer.write(res.data) | |
2225 | ||
2226 | ||
2227 | # Prints the exception message `msg` and exits with status 1. | |
2228 | def _fail(msg: str) -> NoReturn: | |
2229 | if not msg.endswith("."): | |
2230 | msg += "." | |
2231 | ||
2232 | print(msg, file=sys.stderr) | |
2233 | sys.exit(1) | |
2234 | ||
2235 | ||
2236 | # CLI entry point. | |
2237 | def _run_cli(): | |
2238 | try: | |
2239 | _try_run_cli() | |
2240 | except Exception as exc: | |
2241 | _fail(str(exc)) | |
2242 | ||
2243 | ||
2244 | if __name__ == "__main__": | |
2245 | _run_cli() |