1 # The MIT License (MIT)
3 # Copyright (c) 2023 Philippe Proulx <eeppeliteloop@gmail.com>
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:
13 # The above copyright notice and this permission notice shall be
14 # included in all copies or substantial portions of the Software.
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.
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.
28 # Feel free to copy this module file to your own project to use Normand.
30 # Upstream repository: <https://github.com/efficios/normand>.
32 __author__
= "Philippe Proulx"
33 __version__
= "0.23.0"
61 from typing
import Any
, Set
, Dict
, List
, Union
, Pattern
, Callable
, NoReturn
, Optional
64 # Text location (line and column numbers).
67 def _create(cls
, line_no
: int, col_no
: int):
68 self
= cls
.__new
__(cls
)
69 self
._init
(line_no
, col_no
)
72 def __init__(*args
, **kwargs
): # type: ignore
73 raise NotImplementedError
75 def _init(self
, line_no
: int, col_no
: int):
76 self
._line
_no
= line_no
90 return "TextLocation({}, {})".format(self
._line
_no
, self
._col
_no
)
95 def __init__(self
, text_loc
: TextLocation
):
96 self
._text
_loc
= text_loc
98 # Source text location.
101 return self
._text
_loc
105 class _ScalarItem(_Item
):
106 # Returns the size, in bytes, of this item.
109 def size(self
) -> int:
119 class _Byte(_ScalarItem
, _RepableItem
):
120 def __init__(self
, val
: int, text_loc
: TextLocation
):
121 super().__init
__(text_loc
)
134 return "_Byte({}, {})".format(hex(self
._val
), repr(self
._text
_loc
))
138 class _LitStr(_ScalarItem
, _RepableItem
):
139 def __init__(self
, data
: bytes
, text_loc
: TextLocation
):
140 super().__init
__(text_loc
)
150 return len(self
._data
)
153 return "_LitStr({}, {})".format(repr(self
._data
), repr(self
._text
_loc
))
158 class ByteOrder(enum
.Enum
):
166 # Byte order setting.
168 def __init__(self
, bo
: ByteOrder
, text_loc
: TextLocation
):
169 super().__init
__(text_loc
)
177 return "_SetBo({}, {})".format(repr(self
._bo
), repr(self
._text
_loc
))
182 def __init__(self
, name
: str, text_loc
: TextLocation
):
183 super().__init
__(text_loc
)
192 return "_Label({}, {})".format(repr(self
._name
), repr(self
._text
_loc
))
196 class _SetOffset(_Item
):
197 def __init__(self
, val
: int, text_loc
: TextLocation
):
198 super().__init
__(text_loc
)
201 # Offset value (bytes).
207 return "_SetOffset({}, {})".format(repr(self
._val
), repr(self
._text
_loc
))
211 class _AlignOffset(_Item
):
212 def __init__(self
, val
: int, pad_val
: int, text_loc
: TextLocation
):
213 super().__init
__(text_loc
)
215 self
._pad
_val
= pad_val
217 # Alignment value (bits).
222 # Padding byte value.
228 return "_AlignOffset({}, {}, {})".format(
229 repr(self
._val
), repr(self
._pad
_val
), repr(self
._text
_loc
)
233 # Mixin of containing an AST expression and its string.
235 def __init__(self
, expr_str
: str, expr
: ast
.Expression
):
236 self
._expr
_str
= expr_str
242 return self
._expr
_str
244 # Expression node to evaluate.
250 # Fill until some offset.
251 class _FillUntil(_Item
, _ExprMixin
):
253 self
, expr_str
: str, expr
: ast
.Expression
, pad_val
: int, text_loc
: TextLocation
255 super().__init
__(text_loc
)
256 _ExprMixin
.__init
__(self
, expr_str
, expr
)
257 self
._pad
_val
= pad_val
259 # Padding byte value.
265 return "_FillUntil({}, {}, {}, {})".format(
266 repr(self
._expr
_str
),
269 repr(self
._text
_loc
),
273 # Variable assignment.
274 class _VarAssign(_Item
, _ExprMixin
):
276 self
, name
: str, expr_str
: str, expr
: ast
.Expression
, text_loc
: TextLocation
278 super().__init
__(text_loc
)
279 _ExprMixin
.__init
__(self
, expr_str
, expr
)
288 return "_VarAssign({}, {}, {}, {})".format(
290 repr(self
._expr
_str
),
292 repr(self
._text
_loc
),
296 # Fixed-length number, possibly needing more than one byte.
297 class _FlNum(_ScalarItem
, _RepableItem
, _ExprMixin
):
301 expr
: ast
.Expression
,
303 bo
: Optional
[ByteOrder
],
304 text_loc
: TextLocation
,
306 super().__init
__(text_loc
)
307 _ExprMixin
.__init
__(self
, expr_str
, expr
)
316 # Byte order override.
323 return self
._len
// 8
326 return "_FlNum({}, {}, {}, {}, {})".format(
327 repr(self
._expr
_str
),
331 repr(self
._text
_loc
),
336 class _Leb128Int(_Item
, _RepableItem
, _ExprMixin
):
337 def __init__(self
, expr_str
: str, expr
: ast
.Expression
, text_loc
: TextLocation
):
338 super().__init
__(text_loc
)
339 _ExprMixin
.__init
__(self
, expr_str
, expr
)
342 return "{}({}, {}, {})".format(
343 self
.__class
__.__name
__,
344 repr(self
._expr
_str
),
346 repr(self
._text
_loc
),
350 # Unsigned LEB128 integer.
351 class _ULeb128Int(_Leb128Int
, _RepableItem
, _ExprMixin
):
355 # Signed LEB128 integer.
356 class _SLeb128Int(_Leb128Int
, _RepableItem
, _ExprMixin
):
361 class _Str(_Item
, _RepableItem
, _ExprMixin
):
363 self
, expr_str
: str, expr
: ast
.Expression
, codec
: str, text_loc
: TextLocation
365 super().__init
__(text_loc
)
366 _ExprMixin
.__init
__(self
, expr_str
, expr
)
375 return "_Str({}, {}, {}, {})".format(
376 repr(self
._expr
_str
),
379 repr(self
._text
_loc
),
384 class _Group(_Item
, _RepableItem
):
385 def __init__(self
, items
: List
[_Item
], text_loc
: TextLocation
):
386 super().__init
__(text_loc
)
395 return "_Group({}, {})".format(repr(self
._items
), repr(self
._text
_loc
))
399 class _Rep(_Group
, _ExprMixin
):
404 expr
: ast
.Expression
,
405 text_loc
: TextLocation
,
407 super().__init
__(items
, text_loc
)
408 _ExprMixin
.__init
__(self
, expr_str
, expr
)
411 return "_Rep({}, {}, {}, {})".format(
413 repr(self
._expr
_str
),
415 repr(self
._text
_loc
),
420 class _Cond(_Item
, _ExprMixin
):
426 expr
: ast
.Expression
,
427 text_loc
: TextLocation
,
429 super().__init
__(text_loc
)
430 _ExprMixin
.__init
__(self
, expr_str
, expr
)
431 self
._true
_item
= true_item
432 self
._false
_item
= false_item
434 # Item when condition is true.
437 return self
._true
_item
439 # Item when condition is false.
441 def false_item(self
):
442 return self
._false
_item
445 return "_Cond({}, {}, {}, {}, {})".format(
446 repr(self
._true
_item
),
447 repr(self
._false
_item
),
448 repr(self
._expr
_str
),
450 repr(self
._text
_loc
),
455 class _Trans(_Group
, _RepableItem
):
460 func
: Callable
[[Union
[bytes
, bytearray
]], bytes
],
461 text_loc
: TextLocation
,
463 super().__init
__(items
, text_loc
)
471 # Transforms the data `data`.
472 def trans(self
, data
: Union
[bytes
, bytearray
]):
473 return self
._func
(data
)
476 return "_Trans({}, {}, {}, {})".format(
480 repr(self
._text
_loc
),
484 # Macro definition item.
485 class _MacroDef(_Group
):
489 param_names
: List
[str],
491 text_loc
: TextLocation
,
493 super().__init
__(items
, text_loc
)
495 self
._param
_names
= param_names
504 def param_names(self
):
505 return self
._param
_names
508 return "_MacroDef({}, {}, {}, {})".format(
510 repr(self
._param
_names
),
512 repr(self
._text
_loc
),
516 # Macro expansion parameter.
517 class _MacroExpParam
:
518 def __init__(self
, expr_str
: str, expr
: ast
.Expression
, text_loc
: TextLocation
):
519 self
._expr
_str
= expr_str
521 self
._text
_loc
= text_loc
526 return self
._expr
_str
533 # Source text location.
536 return self
._text
_loc
539 return "_MacroExpParam({}, {}, {})".format(
540 repr(self
._expr
_str
), repr(self
._expr
), repr(self
._text
_loc
)
544 # Macro expansion item.
545 class _MacroExp(_Item
, _RepableItem
):
549 params
: List
[_MacroExpParam
],
550 text_loc
: TextLocation
,
552 super().__init
__(text_loc
)
554 self
._params
= params
567 return "_MacroExp({}, {}, {})".format(
570 repr(self
._text
_loc
),
574 # A parsing error message: a string and a text location.
575 class ParseErrorMessage
:
577 def _create(cls
, text
: str, text_loc
: TextLocation
):
578 self
= cls
.__new
__(cls
)
579 self
._init
(text
, text_loc
)
582 def __init__(self
, *args
, **kwargs
): # type: ignore
583 raise NotImplementedError
585 def _init(self
, text
: str, text_loc
: TextLocation
):
587 self
._text
_loc
= text_loc
594 # Source text location.
596 def text_location(self
):
597 return self
._text
_loc
600 # A parsing error containing one or more messages (`ParseErrorMessage`).
601 class ParseError(RuntimeError):
603 def _create(cls
, msg
: str, text_loc
: TextLocation
):
604 self
= cls
.__new
__(cls
)
605 self
._init
(msg
, text_loc
)
608 def __init__(self
, *args
, **kwargs
): # type: ignore
609 raise NotImplementedError
611 def _init(self
, msg
: str, text_loc
: TextLocation
):
612 super().__init
__(msg
)
613 self
._msgs
= [] # type: List[ParseErrorMessage]
614 self
._add
_msg
(msg
, text_loc
)
616 def _add_msg(self
, msg
: str, text_loc
: TextLocation
):
618 ParseErrorMessage
._create
( # pyright: ignore[reportPrivateUsage]
623 # Parsing error messages.
625 # The first message is the most specific one.
631 # Raises a parsing error, forwarding the parameters to the constructor.
632 def _raise_error(msg
: str, text_loc
: TextLocation
) -> NoReturn
:
633 raise ParseError
._create
(msg
, text_loc
) # pyright: ignore[reportPrivateUsage]
636 # Adds a message to the parsing error `exc`.
637 def _add_error_msg(exc
: ParseError
, msg
: str, text_loc
: TextLocation
):
638 exc
._add
_msg
(msg
, text_loc
) # pyright: ignore[reportPrivateUsage]
641 # Appends a message to the parsing error `exc` and reraises it.
642 def _augment_error(exc
: ParseError
, msg
: str, text_loc
: TextLocation
) -> NoReturn
:
643 _add_error_msg(exc
, msg
, text_loc
)
647 # Returns a normalized version (so as to be parseable by int()) of
648 # the constant integer string `s`, possibly negative, dealing with
650 def _norm_const_int(s
: str):
654 if s
.startswith("-"):
659 if pos
.startswith("0" + r
):
675 for suf
in asm_suf_base
:
677 s
= "{}0{}{}".format(neg
, asm_suf_base
[suf
], pos
.rstrip(suf
))
682 # Encodes the string `s` using the codec `codec`, raising `ParseError`
683 # with `text_loc` on encoding error.
684 def _encode_str(s
: str, codec
: str, text_loc
: TextLocation
):
686 return s
.encode(codec
)
687 except UnicodeEncodeError:
689 "Cannot encode `{}` with the `{}` encoding".format(s
, codec
), text_loc
693 # Variables dictionary type (for type hints).
694 VariablesT
= Dict
[str, Union
[int, float, str]]
697 # Labels dictionary type (for type hints).
698 LabelsT
= Dict
[str, int]
702 _py_name_pat
= re
.compile(r
"[a-zA-Z_][a-zA-Z0-9_]*")
703 _pos_const_int_pat
= re
.compile(
704 r
"(?:0[Xx][A-Fa-f0-9]+|0[Oo][0-7]+|0[Bb][01]+|[A-Fa-f0-9]+[hH]|[0-7]+[qQoO]|[01]+[bB]|\d+)\b"
706 _const_int_pat
= re
.compile(r
"(?P<neg>-)?(?:{})".format(_pos_const_int_pat
.pattern
))
707 _const_float_pat
= re
.compile(
708 r
"[-+]?(?:(?:\d*\.\d+)|(?:\d+\.))(?:[Ee][+-]?\d+)?(?=\W|)"
712 # Macro definition dictionary.
713 _MacroDefsT
= Dict
[str, _MacroDef
]
718 # The constructor accepts a Normand input. After building, use the `res`
719 # property to get the resulting main group.
721 # Builds a parser to parse the Normand input `normand`, parsing
723 def __init__(self
, normand
: str, variables
: VariablesT
, labels
: LabelsT
):
724 self
._normand
= normand
728 self
._label
_names
= set(labels
.keys())
729 self
._var
_names
= set(variables
.keys())
730 self
._macro
_defs
= {} # type: _MacroDefsT
731 self
._base
_item
_parse
_funcs
= [
732 self
._try
_parse
_byte
,
735 self
._try
_parse
_var
_assign
,
736 self
._try
_parse
_set
_bo
,
737 self
._try
_parse
_label
_or
_set
_offset
,
738 self
._try
_parse
_align
_offset
,
739 self
._try
_parse
_fill
_until
,
740 self
._try
_parse
_group
,
741 self
._try
_parse
_rep
_block
,
742 self
._try
_parse
_cond
_block
,
743 self
._try
_parse
_macro
_exp
,
744 self
._try
_parse
_trans
_block
,
748 # Result (main group).
755 def macro_defs(self
):
756 return self
._macro
_defs
758 # Current text location.
761 return TextLocation
._create
( # pyright: ignore[reportPrivateUsage]
762 self
._line
_no
, self
._col
_no
765 # Returns `True` if this parser is done parsing.
767 return self
._at
== len(self
._normand
)
769 # Returns `True` if this parser isn't done parsing.
770 def _isnt_done(self
):
771 return not self
._is
_done
()
773 # Raises a parse error, creating it using the message `msg` and the
774 # current text location.
775 def _raise_error(self
, msg
: str) -> NoReturn
:
776 _raise_error(msg
, self
._text
_loc
)
778 # Tries to make the pattern `pat` match the current substring,
779 # returning the match object and updating `self._at`,
780 # `self._line_no`, and `self._col_no` on success.
781 def _try_parse_pat(self
, pat
: Pattern
[str]):
782 m
= pat
.match(self
._normand
, self
._at
)
787 # Skip matched string
788 self
._at
+= len(m
.group(0))
791 self
._line
_no
+= m
.group(0).count("\n")
793 # Update column number
794 for i
in reversed(range(self
._at
)):
795 if self
._normand
[i
] == "\n" or i
== 0:
797 self
._col
_no
= self
._at
+ 1
799 self
._col
_no
= self
._at
- i
803 # Return match object
806 # Expects the pattern `pat` to match the current substring,
807 # returning the match object and updating `self._at`,
808 # `self._line_no`, and `self._col_no` on success, or raising a parse
809 # error with the message `error_msg` on error.
810 def _expect_pat(self
, pat
: Pattern
[str], error_msg
: str):
812 m
= self
._try
_parse
_pat
(pat
)
816 self
._raise
_error
(error_msg
)
818 # Return match object
821 # Patterns for _skip_*()
822 _comment_pat
= re
.compile(r
"#[^#]*?(?:$|#)", re
.M
)
823 _ws_or_comments_pat
= re
.compile(r
"(?:\s|{})*".format(_comment_pat
.pattern
), re
.M
)
824 _ws_or_syms_or_comments_pat
= re
.compile(
825 r
"(?:[\s/\\?&:;.,_=|-]|{})*".format(_comment_pat
.pattern
), re
.M
828 # Skips as many whitespaces and comments as possible, but not
829 # insignificant symbol characters.
830 def _skip_ws_and_comments(self
):
831 self
._try
_parse
_pat
(self
._ws
_or
_comments
_pat
)
833 # Skips as many whitespaces, insignificant symbol characters, and
834 # comments as possible.
835 def _skip_ws_and_comments_and_syms(self
):
836 self
._try
_parse
_pat
(self
._ws
_or
_syms
_or
_comments
_pat
)
838 # Pattern for _try_parse_hex_byte()
839 _nibble_pat
= re
.compile(r
"[A-Fa-f0-9]")
841 # Tries to parse a hexadecimal byte, returning a byte item on
843 def _try_parse_hex_byte(self
):
844 begin_text_loc
= self
._text
_loc
846 # Match initial nibble
847 m_high
= self
._try
_parse
_pat
(self
._nibble
_pat
)
853 # Expect another nibble
854 self
._skip
_ws
_and
_comments
_and
_syms
()
855 m_low
= self
._expect
_pat
(
856 self
._nibble
_pat
, "Expecting another hexadecimal nibble"
860 return _Byte(int(m_high
.group(0) + m_low
.group(0), 16), begin_text_loc
)
862 # Patterns for _try_parse_bin_byte()
863 _bin_byte_bit_pat
= re
.compile(r
"[01]")
864 _bin_byte_prefix_pat
= re
.compile(r
"%+")
866 # Tries to parse a binary byte, returning a byte item on success.
867 def _try_parse_bin_byte(self
):
868 begin_text_loc
= self
._text
_loc
871 m
= self
._try
_parse
_pat
(self
._bin
_byte
_prefix
_pat
)
877 # Expect as many bytes as there are `%` prefixes
878 items
= [] # type: List[_Item]
880 for _
in range(len(m
.group(0))):
881 self
._skip
_ws
_and
_comments
_and
_syms
()
882 byte_text_loc
= self
._text
_loc
883 bits
= [] # type: List[str]
887 self
._skip
_ws
_and
_comments
_and
_syms
()
888 m
= self
._expect
_pat
(
889 self
._bin
_byte
_bit
_pat
, "Expecting a bit (`0` or `1`)"
891 bits
.append(m
.group(0))
893 items
.append(_Byte(int("".join(bits
), 2), byte_text_loc
))
900 return _Group(items
, begin_text_loc
)
902 # Patterns for _try_parse_dec_byte()
903 _dec_byte_prefix_pat
= re
.compile(r
"\$")
904 _dec_byte_val_pat
= re
.compile(r
"(?P<neg>-?)(?P<val>\d+)")
906 # Tries to parse a decimal byte, returning a byte item on success.
907 def _try_parse_dec_byte(self
):
908 begin_text_loc
= self
._text
_loc
911 if self
._try
_parse
_pat
(self
._dec
_byte
_prefix
_pat
) is None:
916 self
._skip
_ws
_and
_comments
()
917 m
= self
._expect
_pat
(self
._dec
_byte
_val
_pat
, "Expecting a decimal constant")
920 val
= int(m
.group("val")) * (-1 if m
.group("neg") == "-" else 1)
923 if val
< -128 or val
> 255:
924 _raise_error("Invalid decimal byte value {}".format(val
), begin_text_loc
)
930 return _Byte(val
, begin_text_loc
)
932 # Tries to parse a byte, returning a byte item on success.
933 def _try_parse_byte(self
):
935 item
= self
._try
_parse
_hex
_byte
()
941 item
= self
._try
_parse
_bin
_byte
()
947 item
= self
._try
_parse
_dec
_byte
()
952 # Strings corresponding to escape sequence characters
953 _lit_str_escape_seq_strs
= {
967 # Patterns for _try_parse_lit_str()
968 _lit_str_prefix_suffix_pat
= re
.compile(r
'"')
969 _lit_str_contents_pat
= re
.compile(r
'(?:(?:\\.)|[^"])*')
971 # Parses a literal string between double quotes (without an encoding
972 # prefix) and returns the resulting string.
973 def _try_parse_lit_str(self
, with_prefix
: bool):
974 # Match prefix if needed
976 if self
._try
_parse
_pat
(self
._lit
_str
_prefix
_suffix
_pat
) is None:
980 # Expect literal string
981 m
= self
._expect
_pat
(self
._lit
_str
_contents
_pat
, "Expecting a literal string")
983 # Expect end of string
985 self
._lit
_str
_prefix
_suffix
_pat
, 'Expecting `"` (end of literal string)'
988 # Replace escape sequences
991 for ec
in '0abefnrtv"\\':
992 val
= val
.replace(r
"\{}".format(ec
), self
._lit
_str
_escape
_seq
_strs
[ec
])
997 # Patterns for _try_parse_utf_str_encoding()
998 _str_encoding_utf_prefix_pat
= re
.compile(r
"u")
999 _str_encoding_utf_pat
= re
.compile(r
"(?:8|(?:(?:16|32)(?:[bl]e)))\b")
1001 # Tries to parse a UTF encoding specification, returning the Python
1002 # codec name on success.
1003 def _try_parse_utf_str_encoding(self
):
1005 if self
._try
_parse
_pat
(self
._str
_encoding
_utf
_prefix
_pat
) is None:
1009 # Expect UTF specification
1010 m
= self
._expect
_pat
(
1011 self
._str
_encoding
_utf
_pat
,
1012 "Expecting `8`, `16be`, `16le`, `32be` or `32le`",
1015 # Convert to codec name
1018 "16be": "utf_16_be",
1019 "16le": "utf_16_le",
1020 "32be": "utf_32_be",
1021 "32le": "utf_32_le",
1024 # Patterns for _try_parse_str_encoding()
1025 _str_encoding_gen_prefix_pat
= re
.compile(r
"s")
1026 _str_encoding_colon_pat
= re
.compile(r
":")
1027 _str_encoding_non_utf_pat
= re
.compile(r
"latin(?:[1-9]|10)\b")
1029 # Tries to parse a string encoding specification, returning the
1030 # Python codec name on success.
1032 # Requires the general prefix (`s:`) if `req_gen_prefix` is `True`.
1033 def _try_parse_str_encoding(self
, req_gen_prefix
: bool = False):
1035 if self
._try
_parse
_pat
(self
._str
_encoding
_gen
_prefix
_pat
) is not None:
1037 self
._skip
_ws
_and
_comments
()
1038 self
._expect
_pat
(self
._str
_encoding
_colon
_pat
, "Expecting `:`")
1040 # Expect encoding specification
1041 self
._skip
_ws
_and
_comments
()
1044 codec
= self
._try
_parse
_utf
_str
_encoding
()
1046 if codec
is not None:
1050 m
= self
._expect
_pat
(
1051 self
._str
_encoding
_non
_utf
_pat
,
1052 "Expecting `u8`, `u16be`, `u16le`, `u32be`, `u32le`, or `latin1` to `latin10`",
1057 if not req_gen_prefix
:
1058 return self
._try
_parse
_utf
_str
_encoding
()
1060 # Patterns for _try_parse_str()
1061 _lit_str_prefix_pat
= re
.compile(r
'"')
1062 _str_prefix_pat
= re
.compile(r
'"|\{')
1063 _str_expr_pat
= re
.compile(r
"[^}]+")
1064 _str_expr_suffix_pat
= re
.compile(r
"\}")
1066 # Tries to parse a string, returning a literal string or string item
1068 def _try_parse_str(self
):
1069 begin_text_loc
= self
._text
_loc
1072 codec
= self
._try
_parse
_str
_encoding
()
1074 # Match prefix (expect if there's an encoding specification)
1075 self
._skip
_ws
_and
_comments
()
1078 # No encoding: only a literal string (UTF-8) is legal
1079 m_prefix
= self
._try
_parse
_pat
(self
._lit
_str
_prefix
_pat
)
1081 if m_prefix
is None:
1084 # Encoding present: expect a string prefix
1085 m_prefix
= self
._expect
_pat
(self
._str
_prefix
_pat
, 'Expecting `"` or `{`')
1087 # Literal string or expression?
1088 prefix
= m_prefix
.group(0)
1091 # Expect literal string
1092 str_text_loc
= self
._text
_loc
1093 val
= self
._try
_parse
_lit
_str
(False)
1096 self
._raise
_error
("Expecting a literal string")
1099 data
= _encode_str(val
, "utf_8" if codec
is None else codec
, str_text_loc
)
1102 return _LitStr(data
, begin_text_loc
)
1105 self
._skip
_ws
_and
_comments
()
1106 expr_text_loc
= self
._text
_loc
1107 m
= self
._expect
_pat
(self
._str
_expr
_pat
, "Expecting an expression")
1110 self
._expect
_pat
(self
._str
_expr
_suffix
_pat
, "Expecting `}`")
1112 # Create an expression node from the expression string
1113 expr_str
, expr
= self
._ast
_expr
_from
_str
(m
.group(0), expr_text_loc
)
1116 assert codec
is not None
1117 return _Str(expr_str
, expr
, codec
, begin_text_loc
)
1119 # Common right parenthesis pattern
1120 _right_paren_pat
= re
.compile(r
"\)")
1122 # Patterns for _try_parse_group()
1123 _group_prefix_pat
= re
.compile(r
"\(|!g(?:roup)?\b")
1125 # Tries to parse a group, returning a group item on success.
1126 def _try_parse_group(self
):
1127 begin_text_loc
= self
._text
_loc
1130 m_open
= self
._try
_parse
_pat
(self
._group
_prefix
_pat
)
1137 items
= self
._parse
_items
()
1139 # Expect end of group
1140 self
._skip
_ws
_and
_comments
_and
_syms
()
1142 if m_open
.group(0) == "(":
1143 pat
= self
._right
_paren
_pat
1146 pat
= self
._block
_end
_pat
1149 self
._expect
_pat
(pat
, "Expecting an item or `{}` (end of group)".format(exp
))
1152 return _Group(items
, begin_text_loc
)
1154 # Returns a stripped expression string and an AST expression node
1155 # from the expression string `expr_str` at text location `text_loc`.
1156 def _ast_expr_from_str(self
, expr_str
: str, text_loc
: TextLocation
):
1157 # Create an expression node from the expression string
1158 expr_str
= expr_str
.strip().replace("\n", " ")
1161 expr
= ast
.parse(expr_str
, mode
="eval")
1164 "Invalid expression `{}`: invalid syntax".format(expr_str
),
1168 return expr_str
, expr
1170 # Returns a `ByteOrder` value from the _valid_ byte order string
1173 def _bo_from_str(bo_str
: str):
1179 # Patterns for _try_parse_val()
1180 _val_prefix_pat
= re
.compile(r
"\[")
1181 _val_expr_pat
= re
.compile(r
"([^\]:]+):")
1182 _fl_num_len_fmt_pat
= re
.compile(r
"(?P<len>8|16|24|32|40|48|56|64)(?P<bo>[bl]e)?")
1183 _leb128_int_fmt_pat
= re
.compile(r
"(u|s)leb128")
1184 _val_suffix_pat
= re
.compile(r
"]")
1186 # Tries to parse a value (number or string) and format (fixed length
1187 # in bits and optional byte order override, `uleb128`, `sleb128`, or
1188 # `s:` followed with an encoding name), returning an item on
1190 def _try_parse_val(self
):
1192 if self
._try
_parse
_pat
(self
._val
_prefix
_pat
) is None:
1196 # Expect expression and `:`
1197 self
._skip
_ws
_and
_comments
()
1198 expr_text_loc
= self
._text
_loc
1199 m
= self
._expect
_pat
(self
._val
_expr
_pat
, "Expecting an expression")
1201 # Create an expression node from the expression string
1202 expr_str
, expr
= self
._ast
_expr
_from
_str
(m
.group(1), expr_text_loc
)
1205 self
._skip
_ws
_and
_comments
()
1206 m_fmt
= self
._try
_parse
_pat
(self
._fl
_num
_len
_fmt
_pat
)
1208 if m_fmt
is not None:
1209 # Byte order override
1210 if m_fmt
.group("bo") is None:
1213 bo
= self
._bo
_from
_str
(m_fmt
.group("bo"))
1215 # Create fixed-length number item
1219 int(m_fmt
.group("len")),
1225 m_fmt
= self
._try
_parse
_pat
(self
._leb
128_int
_fmt
_pat
)
1227 if m_fmt
is not None:
1228 # Create LEB128 integer item
1229 cls
= _ULeb128Int
if m_fmt
.group(1) == "u" else _SLeb128Int
1230 item
= cls(expr_str
, expr
, expr_text_loc
)
1233 codec
= self
._try
_parse
_str
_encoding
(True)
1235 if codec
is not None:
1236 # Create string item
1237 item
= _Str(expr_str
, expr
, codec
, expr_text_loc
)
1239 # At this point it's invalid
1241 "Expecting a fixed length (multiple of eight bits and optional `be` or `le`), `uleb128`, `sleb128`, or `s:` followed with a valid encoding (`u8`, `u16be`, `u16le`, `u32be`, `u32le`, or `latin1` to `latin10`)"
1245 self
._skip
_ws
_and
_comments
()
1246 m
= self
._expect
_pat
(self
._val
_suffix
_pat
, "Expecting `]`")
1251 # Patterns for _try_parse_var_assign()
1252 _var_assign_prefix_pat
= re
.compile(r
"\{")
1253 _var_assign_equal_pat
= re
.compile(r
"=")
1254 _var_assign_expr_pat
= re
.compile(r
"[^}]+")
1255 _var_assign_suffix_pat
= re
.compile(r
"\}")
1257 # Tries to parse a variable assignment, returning a variable
1258 # assignment item on success.
1259 def _try_parse_var_assign(self
):
1261 if self
._try
_parse
_pat
(self
._var
_assign
_prefix
_pat
) is None:
1266 self
._skip
_ws
_and
_comments
()
1267 name_text_loc
= self
._text
_loc
1268 m
= self
._expect
_pat
(_py_name_pat
, "Expecting a valid Python name")
1272 self
._skip
_ws
_and
_comments
()
1273 self
._expect
_pat
(self
._var
_assign
_equal
_pat
, "Expecting `=`")
1276 self
._skip
_ws
_and
_comments
()
1277 expr_text_loc
= self
._text
_loc
1278 m_expr
= self
._expect
_pat
(self
._var
_assign
_expr
_pat
, "Expecting an expression")
1281 self
._skip
_ws
_and
_comments
()
1282 self
._expect
_pat
(self
._var
_assign
_suffix
_pat
, "Expecting `}`")
1285 if name
== _icitte_name
:
1287 "`{}` is a reserved variable name".format(_icitte_name
), name_text_loc
1290 if name
in self
._label
_names
:
1291 _raise_error("Existing label named `{}`".format(name
), name_text_loc
)
1293 # Create an expression node from the expression string
1294 expr_str
, expr
= self
._ast
_expr
_from
_str
(m_expr
.group(0), expr_text_loc
)
1296 # Add to known variable names
1297 self
._var
_names
.add(name
)
1307 # Pattern for _try_parse_set_bo()
1308 _set_bo_pat
= re
.compile(r
"!([bl]e)\b")
1310 # Tries to parse a byte order setting, returning a byte order
1311 # setting item on success.
1312 def _try_parse_set_bo(self
):
1313 begin_text_loc
= self
._text
_loc
1316 m
= self
._try
_parse
_pat
(self
._set
_bo
_pat
)
1322 # Return corresponding item
1323 if m
.group(1) == "be":
1326 assert m
.group(1) == "le"
1329 return _SetBo(bo
, begin_text_loc
)
1331 # Tries to parse an offset setting value (after the initial `<`),
1332 # returning an offset item on success.
1333 def _try_parse_set_offset_val(self
):
1334 begin_text_loc
= self
._text
_loc
1337 m
= self
._try
_parse
_pat
(_pos_const_int_pat
)
1344 return _SetOffset(int(_norm_const_int(m
.group(0)), 0), begin_text_loc
)
1346 # Tries to parse a label name (after the initial `<`), returning a
1347 # label item on success.
1348 def _try_parse_label_name(self
):
1349 begin_text_loc
= self
._text
_loc
1352 m
= self
._try
_parse
_pat
(_py_name_pat
)
1361 if name
== _icitte_name
:
1363 "`{}` is a reserved label name".format(_icitte_name
), begin_text_loc
1366 if name
in self
._label
_names
:
1367 _raise_error("Duplicate label name `{}`".format(name
), begin_text_loc
)
1369 if name
in self
._var
_names
:
1370 _raise_error("Existing variable named `{}`".format(name
), begin_text_loc
)
1372 # Add to known label names
1373 self
._label
_names
.add(name
)
1376 return _Label(name
, begin_text_loc
)
1378 # Patterns for _try_parse_label_or_set_offset()
1379 _label_set_offset_prefix_pat
= re
.compile(r
"<")
1380 _label_set_offset_suffix_pat
= re
.compile(r
">")
1382 # Tries to parse a label or an offset setting, returning an item on
1384 def _try_parse_label_or_set_offset(self
):
1386 if self
._try
_parse
_pat
(self
._label
_set
_offset
_prefix
_pat
) is None:
1390 # Offset setting item?
1391 self
._skip
_ws
_and
_comments
()
1392 item
= self
._try
_parse
_set
_offset
_val
()
1396 item
= self
._try
_parse
_label
_name
()
1399 # At this point it's invalid
1400 self
._raise
_error
("Expecting a label name or an offset setting value")
1403 self
._skip
_ws
_and
_comments
()
1404 self
._expect
_pat
(self
._label
_set
_offset
_suffix
_pat
, "Expecting `>`")
1407 # Pattern for _parse_pad_val()
1408 _pad_val_prefix_pat
= re
.compile(r
"~")
1410 # Tries to parse a padding value, returning the padding value, or 0
1412 def _parse_pad_val(self
):
1414 self
._skip
_ws
_and
_comments
()
1417 if self
._try
_parse
_pat
(self
._pad
_val
_prefix
_pat
) is not None:
1418 self
._skip
_ws
_and
_comments
()
1419 pad_val_text_loc
= self
._text
_loc
1420 m
= self
._expect
_pat
(
1422 "Expecting a positive constant integer (byte value)",
1426 pad_val
= int(_norm_const_int(m
.group(0)), 0)
1430 "Invalid padding byte value {}".format(pad_val
),
1436 # Patterns for _try_parse_align_offset()
1437 _align_offset_prefix_pat
= re
.compile(r
"@")
1438 _align_offset_val_pat
= re
.compile(r
"\d+")
1440 # Tries to parse an offset alignment, returning an offset alignment
1442 def _try_parse_align_offset(self
):
1443 begin_text_loc
= self
._text
_loc
1446 if self
._try
_parse
_pat
(self
._align
_offset
_prefix
_pat
) is None:
1450 # Expect an alignment
1451 self
._skip
_ws
_and
_comments
()
1452 align_text_loc
= self
._text
_loc
1453 m
= self
._expect
_pat
(
1454 self
._align
_offset
_val
_pat
,
1455 "Expecting an alignment (positive multiple of eight bits)",
1458 # Validate alignment
1459 val
= int(m
.group(0))
1461 if val
<= 0 or (val
% 8) != 0:
1463 "Invalid alignment value {} (not a positive multiple of eight)".format(
1470 pad_val
= self
._parse
_pad
_val
()
1473 return _AlignOffset(val
, pad_val
, begin_text_loc
)
1475 # Patterns for _expect_expr()
1476 _inner_expr_prefix_pat
= re
.compile(r
"\{")
1477 _inner_expr_pat
= re
.compile(r
"[^}]+")
1478 _inner_expr_suffix_pat
= re
.compile(r
"\}")
1480 # Parses an expression outside a `{`/`}` context.
1482 # This function accepts:
1484 # • A Python expression within `{` and `}`.
1486 # • A Python name.
1488 # • If `accept_const_int` is `True`: a constant integer, which may
1489 # be negative if `allow_neg_int` is `True`.
1491 # • If `accept_float` is `True`: a constant floating point number.
1493 # Returns the stripped expression string and AST expression.
1496 accept_const_int
: bool = False,
1497 allow_neg_int
: bool = False,
1498 accept_const_float
: bool = False,
1499 accept_lit_str
: bool = False,
1501 begin_text_loc
= self
._text
_loc
1503 # Constant floating point number?
1504 if accept_const_float
:
1505 m
= self
._try
_parse
_pat
(_const_float_pat
)
1508 return self
._ast
_expr
_from
_str
(m
.group(0), begin_text_loc
)
1511 if accept_const_int
:
1512 m
= self
._try
_parse
_pat
(_const_int_pat
)
1515 # Negative and allowed?
1516 if m
.group("neg") == "-" and not allow_neg_int
:
1518 "Expecting a positive constant integer", begin_text_loc
1521 expr_str
= _norm_const_int(m
.group(0))
1522 return self
._ast
_expr
_from
_str
(expr_str
, begin_text_loc
)
1525 m
= self
._try
_parse
_pat
(_py_name_pat
)
1528 return self
._ast
_expr
_from
_str
(m
.group(0), begin_text_loc
)
1532 val
= self
._try
_parse
_lit
_str
(True)
1535 return self
._ast
_expr
_from
_str
(repr(val
), begin_text_loc
)
1538 msg_accepted_parts
= ["a name", "or `{`"]
1541 msg_accepted_parts
.insert(0, "a literal string")
1543 if accept_const_float
:
1544 msg_accepted_parts
.insert(0, "a constant floating point number")
1546 if accept_const_int
:
1547 msg_pos
= "" if allow_neg_int
else "positive "
1548 msg_accepted_parts
.insert(0, "a {}constant integer".format(msg_pos
))
1550 if len(msg_accepted_parts
) == 2:
1551 msg_accepted
= " ".join(msg_accepted_parts
)
1553 msg_accepted
= ", ".join(msg_accepted_parts
)
1556 self
._inner
_expr
_prefix
_pat
,
1557 "Expecting {}".format(msg_accepted
),
1560 # Expect an expression
1561 self
._skip
_ws
_and
_comments
()
1562 expr_text_loc
= self
._text
_loc
1563 m
= self
._expect
_pat
(self
._inner
_expr
_pat
, "Expecting an expression")
1564 expr_str
= m
.group(0)
1567 self
._skip
_ws
_and
_comments
()
1568 self
._expect
_pat
(self
._inner
_expr
_suffix
_pat
, "Expecting `}`")
1570 return self
._ast
_expr
_from
_str
(expr_str
, expr_text_loc
)
1572 # Patterns for _try_parse_fill_until()
1573 _fill_until_prefix_pat
= re
.compile(r
"\+")
1574 _fill_until_pad_val_prefix_pat
= re
.compile(r
"~")
1576 # Tries to parse a filling, returning a filling item on success.
1577 def _try_parse_fill_until(self
):
1578 begin_text_loc
= self
._text
_loc
1581 if self
._try
_parse
_pat
(self
._fill
_until
_prefix
_pat
) is None:
1586 self
._skip
_ws
_and
_comments
()
1587 expr_str
, expr
= self
._expect
_expr
(accept_const_int
=True)
1590 pad_val
= self
._parse
_pad
_val
()
1593 return _FillUntil(expr_str
, expr
, pad_val
, begin_text_loc
)
1595 # Parses the multiplier expression of a repetition (block or
1596 # post-item) and returns the expression string and AST node.
1597 def _expect_rep_mul_expr(self
):
1598 return self
._expect
_expr
(accept_const_int
=True)
1600 # Common block end pattern
1601 _block_end_pat
= re
.compile(r
"!end\b")
1603 # Pattern for _try_parse_rep_block()
1604 _rep_block_prefix_pat
= re
.compile(r
"!r(?:epeat)?\b")
1606 # Tries to parse a repetition block, returning a repetition item on
1608 def _try_parse_rep_block(self
):
1609 begin_text_loc
= self
._text
_loc
1612 if self
._try
_parse
_pat
(self
._rep
_block
_prefix
_pat
) is None:
1617 self
._skip
_ws
_and
_comments
()
1618 expr_str
, expr
= self
._expect
_rep
_mul
_expr
()
1621 self
._skip
_ws
_and
_comments
_and
_syms
()
1622 items
= self
._parse
_items
()
1624 # Expect end of block
1625 self
._skip
_ws
_and
_comments
_and
_syms
()
1627 self
._block
_end
_pat
, "Expecting an item or `!end` (end of repetition block)"
1631 return _Rep(items
, expr_str
, expr
, begin_text_loc
)
1633 # Pattern for _try_parse_cond_block()
1634 _cond_block_prefix_pat
= re
.compile(r
"!if\b")
1635 _cond_block_else_pat
= re
.compile(r
"!else\b")
1637 # Tries to parse a conditional block, returning a conditional item
1639 def _try_parse_cond_block(self
):
1640 begin_text_loc
= self
._text
_loc
1643 if self
._try
_parse
_pat
(self
._cond
_block
_prefix
_pat
) is None:
1648 self
._skip
_ws
_and
_comments
()
1649 expr_str
, expr
= self
._expect
_expr
()
1651 # Parse "true" items
1652 self
._skip
_ws
_and
_comments
_and
_syms
()
1653 true_items_text_loc
= self
._text
_loc
1654 true_items
= self
._parse
_items
()
1655 false_items
= [] # type: List[_Item]
1656 false_items_text_loc
= begin_text_loc
1659 self
._skip
_ws
_and
_comments
_and
_syms
()
1661 if self
._try
_parse
_pat
(self
._cond
_block
_else
_pat
) is not None:
1662 # Parse "false" items
1663 self
._skip
_ws
_and
_comments
_and
_syms
()
1664 false_items_text_loc
= self
._text
_loc
1665 false_items
= self
._parse
_items
()
1667 # Expect end of block
1669 self
._block
_end
_pat
,
1670 "Expecting an item, `!else`, or `!end` (end of conditional block)",
1675 _Group(true_items
, true_items_text_loc
),
1676 _Group(false_items
, false_items_text_loc
),
1682 # Pattern for _try_parse_trans_block()
1683 _trans_block_prefix_pat
= re
.compile(r
"!t(?:ransform)?\b")
1684 _trans_block_type_pat
= re
.compile(
1685 r
"(?:(?:base|b)64(?:u)?|(?:base|b)(?:16|32)|(?:ascii|a|base|b)85(?:p)?|(?:quopri|qp)(?:t)?|gzip|gz|bzip2|bz2)\b"
1688 # Tries to parse a transformation block, returning a transformation
1689 # block item on success.
1690 def _try_parse_trans_block(self
):
1691 begin_text_loc
= self
._text
_loc
1694 if self
._try
_parse
_pat
(self
._trans
_block
_prefix
_pat
) is None:
1699 self
._skip
_ws
_and
_comments
()
1700 m
= self
._expect
_pat
(
1701 self
._trans
_block
_type
_pat
, "Expecting a known transformation type"
1705 self
._skip
_ws
_and
_comments
_and
_syms
()
1706 items
= self
._parse
_items
()
1708 # Expect end of block
1710 self
._block
_end
_pat
,
1711 "Expecting an item or `!end` (end of transformation block)",
1714 # Choose encoding function
1717 if enc
in ("base64", "b64"):
1718 func
= base64
.standard_b64encode
1719 name
= "standard Base64"
1720 elif enc
in ("base64u", "b64u"):
1721 func
= base64
.urlsafe_b64encode
1722 name
= "URL-safe Base64"
1723 elif enc
in ("base32", "b32"):
1724 func
= base64
.b32encode
1726 elif enc
in ("base16", "b16"):
1727 func
= base64
.b16encode
1729 elif enc
in ("ascii85", "a85"):
1730 func
= base64
.a85encode
1732 elif enc
in ("ascii85p", "a85p"):
1733 func
= functools
.partial(base64
.a85encode
, pad
=True)
1734 name
= "padded Ascii85"
1735 elif enc
in ("base85", "b85"):
1736 func
= base64
.b85encode
1738 elif enc
in ("base85p", "b85p"):
1739 func
= functools
.partial(base64
.b85encode
, pad
=True)
1740 name
= "padded Base85"
1741 elif enc
in ("quopri", "qp"):
1742 func
= quopri
.encodestring
1743 name
= "MIME quoted-printable"
1744 elif enc
in ("quoprit", "qpt"):
1745 func
= functools
.partial(quopri
.encodestring
, quotetabs
=True)
1746 name
= "MIME quoted-printable (with quoted tabs)"
1747 elif enc
in ("gzip", "gz"):
1748 func
= gzip
.compress
1751 assert enc
in ("bzip2", "bz2")
1763 # Common left parenthesis pattern
1764 _left_paren_pat
= re
.compile(r
"\(")
1766 # Patterns for _try_parse_macro_def() and _try_parse_macro_exp()
1767 _macro_params_comma_pat
= re
.compile(",")
1769 # Patterns for _try_parse_macro_def()
1770 _macro_def_prefix_pat
= re
.compile(r
"!m(?:acro)?\b")
1772 # Tries to parse a macro definition, adding it to `self._macro_defs`
1773 # and returning `True` on success.
1774 def _try_parse_macro_def(self
):
1775 begin_text_loc
= self
._text
_loc
1778 if self
._try
_parse
_pat
(self
._macro
_def
_prefix
_pat
) is None:
1783 self
._skip
_ws
_and
_comments
()
1784 name_text_loc
= self
._text
_loc
1785 m
= self
._expect
_pat
(_py_name_pat
, "Expecting a valid macro name")
1790 if name
in self
._macro
_defs
:
1791 _raise_error("Duplicate macro named `{}`".format(name
), name_text_loc
)
1794 self
._skip
_ws
_and
_comments
()
1795 self
._expect
_pat
(self
._left
_paren
_pat
, "Expecting `(`")
1797 # Try to parse comma-separated parameter names
1798 param_names
= [] # type: List[str]
1799 expect_comma
= False
1802 self
._skip
_ws
_and
_comments
()
1805 if self
._try
_parse
_pat
(self
._right
_paren
_pat
) is not None:
1811 self
._expect
_pat
(self
._macro
_params
_comma
_pat
, "Expecting `,`")
1813 # Expect parameter name
1814 self
._skip
_ws
_and
_comments
()
1815 param_text_loc
= self
._text
_loc
1816 m
= self
._expect
_pat
(_py_name_pat
, "Expecting valid parameter name")
1818 if m
.group(0) in param_names
:
1820 "Duplicate macro parameter named `{}`".format(m
.group(0)),
1824 param_names
.append(m
.group(0))
1828 self
._skip
_ws
_and
_comments
_and
_syms
()
1829 old_var_names
= self
._var
_names
.copy()
1830 old_label_names
= self
._label
_names
.copy()
1831 self
._var
_names
= set() # type: Set[str]
1832 self
._label
_names
= set() # type: Set[str]
1833 items
= self
._parse
_items
()
1834 self
._var
_names
= old_var_names
1835 self
._label
_names
= old_label_names
1839 self
._block
_end
_pat
, "Expecting an item or `!end` (end of macro block)"
1843 self
._macro
_defs
[name
] = _MacroDef(name
, param_names
, items
, begin_text_loc
)
1847 # Patterns for _try_parse_macro_exp()
1848 _macro_exp_prefix_pat
= re
.compile(r
"m\b")
1849 _macro_exp_colon_pat
= re
.compile(r
":")
1851 # Tries to parse a macro expansion, returning a macro expansion item
1853 def _try_parse_macro_exp(self
):
1854 begin_text_loc
= self
._text
_loc
1857 if self
._try
_parse
_pat
(self
._macro
_exp
_prefix
_pat
) is None:
1862 self
._skip
_ws
_and
_comments
()
1863 self
._expect
_pat
(self
._macro
_exp
_colon
_pat
, "Expecting `:`")
1865 # Expect a macro name
1866 self
._skip
_ws
_and
_comments
()
1867 name_text_loc
= self
._text
_loc
1868 m
= self
._expect
_pat
(_py_name_pat
, "Expecting a valid macro name")
1872 macro_def
= self
._macro
_defs
.get(name
)
1874 if macro_def
is None:
1875 _raise_error("Unknown macro name `{}`".format(name
), name_text_loc
)
1878 self
._skip
_ws
_and
_comments
()
1879 self
._expect
_pat
(self
._left
_paren
_pat
, "Expecting `(`")
1881 # Try to parse comma-separated parameter values
1882 params_text_loc
= self
._text
_loc
1883 params
= [] # type: List[_MacroExpParam]
1884 expect_comma
= False
1887 self
._skip
_ws
_and
_comments
()
1890 if self
._try
_parse
_pat
(self
._right
_paren
_pat
) is not None:
1896 self
._expect
_pat
(self
._macro
_params
_comma
_pat
, "Expecting `,`")
1898 self
._skip
_ws
_and
_comments
()
1899 param_text_loc
= self
._text
_loc
1903 accept_const_int
=True,
1905 accept_const_float
=True,
1906 accept_lit_str
=True,
1908 text_loc
=param_text_loc
1913 # Validate parameter values
1914 if len(params
) != len(macro_def
.param_names
):
1915 sing_plur
= "" if len(params
) == 1 else "s"
1917 "Macro expansion passes {} parameter{} while the definition expects {}".format(
1918 len(params
), sing_plur
, len(macro_def
.param_names
)
1924 return _MacroExp(name
, params
, begin_text_loc
)
1926 # Tries to parse a base item (anything except a post-item
1927 # repetition), returning it on success.
1928 def _try_parse_base_item(self
):
1929 for func
in self
._base
_item
_parse
_funcs
:
1932 if item
is not None:
1935 # Pattern for _try_parse_rep_post()
1936 _rep_post_prefix_pat
= re
.compile(r
"\*")
1938 # Tries to parse a post-item repetition, returning the expression
1939 # string and AST expression node on success.
1940 def _try_parse_rep_post(self
):
1942 if self
._try
_parse
_pat
(self
._rep
_post
_prefix
_pat
) is None:
1946 # Return expression string and AST expression
1947 self
._skip
_ws
_and
_comments
()
1948 return self
._expect
_rep
_mul
_expr
()
1950 # Tries to parse an item, possibly followed by a repetition,
1951 # returning `True` on success.
1953 # Appends any parsed item to `items`.
1954 def _try_append_item(self
, items
: List
[_Item
]):
1955 self
._skip
_ws
_and
_comments
_and
_syms
()
1958 item
= self
._try
_parse
_base
_item
()
1963 # Parse repetition if the base item is repeatable
1964 if isinstance(item
, _RepableItem
):
1965 self
._skip
_ws
_and
_comments
()
1966 rep_text_loc
= self
._text
_loc
1967 rep_ret
= self
._try
_parse
_rep
_post
()
1969 if rep_ret
is not None:
1970 item
= _Rep([item
], *rep_ret
, text_loc
=rep_text_loc
)
1975 # Parses and returns items, skipping whitespaces, insignificant
1976 # symbols, and comments when allowed, and stopping at the first
1977 # unknown character.
1979 # Accepts and registers macro definitions if `accept_macro_defs`
1981 def _parse_items(self
, accept_macro_defs
: bool = False) -> List
[_Item
]:
1982 items
= [] # type: List[_Item]
1984 while self
._isnt
_done
():
1985 # Try to append item
1986 if not self
._try
_append
_item
(items
):
1987 if accept_macro_defs
and self
._try
_parse
_macro
_def
():
1990 # Unknown at this point
1995 # Parses the whole Normand input, setting `self._res` to the main
1996 # group item on success.
1998 if len(self
._normand
.strip()) == 0:
1999 # Special case to make sure there's something to consume
2000 self
._res
= _Group([], self
._text
_loc
)
2003 # Parse first level items
2004 items
= self
._parse
_items
(True)
2006 # Make sure there's nothing left
2007 self
._skip
_ws
_and
_comments
_and
_syms
()
2009 if self
._isnt
_done
():
2011 "Unexpected character `{}`".format(self
._normand
[self
._at
])
2014 # Set main group item
2015 self
._res
= _Group(items
, self
._text
_loc
)
2018 # The return type of parse().
2024 variables
: VariablesT
,
2027 bo
: Optional
[ByteOrder
],
2029 self
= cls
.__new
__(cls
)
2030 self
._init
(data
, variables
, labels
, offset
, bo
)
2033 def __init__(self
, *args
, **kwargs
): # type: ignore
2034 raise NotImplementedError
2039 variables
: VariablesT
,
2042 bo
: Optional
[ByteOrder
],
2045 self
._vars
= variables
2046 self
._labels
= labels
2047 self
._offset
= offset
2055 # Dictionary of updated variable names to their last computed value.
2057 def variables(self
):
2060 # Dictionary of updated main group label names to their computed
2071 # Updated byte order.
2073 def byte_order(self
):
2077 # Raises a parse error for the item `item`, creating it using the
2079 def _raise_error_for_item(msg
: str, item
: _Item
) -> NoReturn
:
2080 _raise_error(msg
, item
.text_loc
)
2083 # The `ICITTE` reserved name.
2084 _icitte_name
= "ICITTE"
2087 # Base node visitor.
2089 # Calls the _visit_name() method for each name node which isn't the name
2091 class _NodeVisitor(ast
.NodeVisitor
):
2093 self
._parent
_is
_call
= False
2095 def generic_visit(self
, node
: ast
.AST
):
2096 if type(node
) is ast
.Call
:
2097 self
._parent
_is
_call
= True
2098 elif type(node
) is ast
.Name
and not self
._parent
_is
_call
:
2099 self
._visit
_name
(node
.id)
2101 super().generic_visit(node
)
2102 self
._parent
_is
_call
= False
2105 def _visit_name(self
, name
: str):
2109 # Expression validator: validates that all the names within the
2110 # expression are allowed.
2111 class _ExprValidator(_NodeVisitor
):
2112 def __init__(self
, expr_str
: str, text_loc
: TextLocation
, allowed_names
: Set
[str]):
2114 self
._expr
_str
= expr_str
2115 self
._text
_loc
= text_loc
2116 self
._allowed
_names
= allowed_names
2118 def _visit_name(self
, name
: str):
2119 # Make sure the name refers to a known and reachable
2120 # variable/label name.
2121 if name
!= _icitte_name
and name
not in self
._allowed
_names
:
2122 msg
= "Illegal (unknown or unreachable) variable/label name `{}` in expression `{}`".format(
2123 name
, self
._expr
_str
2126 allowed_names
= self
._allowed
_names
.copy()
2127 allowed_names
.add(_icitte_name
)
2129 if len(allowed_names
) > 0:
2130 allowed_names_str
= ", ".join(
2131 sorted(["`{}`".format(name
) for name
in allowed_names
])
2133 msg
+= "; the legal names are {{{}}}".format(allowed_names_str
)
2145 variables
: VariablesT
,
2148 bo
: Optional
[ByteOrder
],
2150 self
.variables
= variables
.copy()
2151 self
.labels
= labels
.copy()
2152 self
.offset
= offset
2156 return "_GenState({}, {}, {}, {})".format(
2157 repr(self
.variables
), repr(self
.labels
), repr(self
.offset
), repr(self
.bo
)
2161 # Fixed-length number item instance.
2162 class _FlNumItemInst
:
2166 offset_in_data
: int,
2168 parse_error_msgs
: List
[ParseErrorMessage
],
2171 self
._offset
_in
_data
= offset_in_data
2173 self
._parse
_error
_msgs
= parse_error_msgs
2180 def offset_in_data(self
):
2181 return self
._offset
_in
_data
2188 def parse_error_msgs(self
):
2189 return self
._parse
_error
_msgs
2192 # Generator of data and final state from a group item.
2194 # Generation happens in memory at construction time. After building, use
2195 # the `data`, `variables`, `labels`, `offset`, and `bo` properties to
2196 # get the resulting context.
2198 # The steps of generation are:
2200 # 1. Handle each item in prefix order.
2202 # The handlers append bytes to `self._data` and update some current
2203 # state object (`_GenState` instance).
2205 # When handling a fixed-length number item, try to evaluate its
2206 # expression using the current state. If this fails, then it might be
2207 # because the expression refers to a "future" label: save the current
2208 # offset in `self._data` (generated data) and a snapshot of the
2209 # current state within `self._fl_num_item_insts` (`_FlNumItemInst`
2210 # object). _gen_fl_num_item_insts() will deal with this later. A
2211 # `_FlNumItemInst` instance also contains a snapshot of the current
2212 # parsing error messages (`self._parse_error_msgs`) which need to be
2213 # taken into account when handling the instance later.
2215 # When handling the items of a group, keep a map of immediate label
2216 # names to their offset. Then, after having processed all the items,
2217 # update the relevant saved state snapshots in
2218 # `self._fl_num_item_insts` with those immediate label values.
2219 # _gen_fl_num_item_insts() will deal with this later.
2221 # 2. Handle all the fixed-length number item instances of which the
2222 # expression evaluation failed before.
2224 # At this point, `self._fl_num_item_insts` contains everything that's
2225 # needed to evaluate the expressions, including the values of
2226 # "future" labels from the point of view of some fixed-length number
2229 # If an evaluation fails at this point, then it's a user error. Add
2230 # to the parsing error all the saved parsing error messages of the
2231 # instance. Those additional messages add precious context to the
2237 macro_defs
: _MacroDefsT
,
2238 variables
: VariablesT
,
2241 bo
: Optional
[ByteOrder
],
2243 self
._macro
_defs
= macro_defs
2244 self
._fl
_num
_item
_insts
= [] # type: List[_FlNumItemInst]
2245 self
._parse
_error
_msgs
= [] # type: List[ParseErrorMessage]
2246 self
._in
_trans
= False
2247 self
._gen
(group
, _GenState(variables
, labels
, offset
, bo
))
2254 # Updated variables.
2256 def variables(self
):
2257 return self
._final
_state
.variables
2259 # Updated main group labels.
2262 return self
._final
_state
.labels
2267 return self
._final
_state
.offset
2269 # Updated byte order.
2272 return self
._final
_state
.bo
2274 # Evaluates the expression `expr` of which the original string is
2275 # `expr_str` at the location `text_loc` considering the current
2276 # generation state `state`.
2278 # If `accept_float` is `True`, then the type of the result may be
2281 # If `accept_str` is `True`, then the type of the result may be
2286 expr
: ast
.Expression
,
2287 text_loc
: TextLocation
,
2289 accept_float
: bool = False,
2290 accept_str
: bool = False,
2292 syms
= {} # type: VariablesT
2293 syms
.update(state
.labels
)
2295 # Set the `ICITTE` name to the current offset
2296 syms
[_icitte_name
] = state
.offset
2298 # Add the current variables
2299 syms
.update(state
.variables
)
2301 # Validate the node and its children
2302 _ExprValidator(expr_str
, text_loc
, set(syms
.keys())).visit(expr
)
2304 # Compile and evaluate expression node
2306 val
= eval(compile(expr
, "", "eval"), None, syms
)
2307 except Exception as exc
:
2309 "Failed to evaluate expression `{}`: {}".format(expr_str
, exc
),
2313 # Convert `bool` result type to `int` to normalize
2314 if type(val
) is bool:
2317 # Validate result type
2318 expected_types
= {int}
# type: Set[type]
2321 expected_types
.add(float)
2324 expected_types
.add(str)
2326 if type(val
) not in expected_types
:
2327 expected_types_str
= sorted(
2328 ["`{}`".format(t
.__name
__) for t
in expected_types
]
2331 if len(expected_types_str
) == 1:
2332 msg_expected
= expected_types_str
[0]
2333 elif len(expected_types_str
) == 2:
2334 msg_expected
= " or ".join(expected_types_str
)
2336 expected_types_str
[-1] = "or {}".format(expected_types_str
[-1])
2337 msg_expected
= ", ".join(expected_types_str
)
2340 "Invalid expression `{}`: expecting result type {}, not `{}`".format(
2341 expr_str
, msg_expected
, type(val
).__name
__
2348 # Forwards to _eval_expr() with the expression and text location of
2351 def _eval_item_expr(
2352 item
: Union
[_Cond
, _FillUntil
, _FlNum
, _Leb128Int
, _Rep
, _Str
, _VarAssign
],
2354 accept_float
: bool = False,
2355 accept_str
: bool = False,
2357 return _Gen
._eval
_expr
(
2358 item
.expr_str
, item
.expr
, item
.text_loc
, state
, accept_float
, accept_str
2361 # Handles the byte item `item`.
2362 def _handle_byte_item(self
, item
: _Byte
, state
: _GenState
):
2363 self
._data
.append(item
.val
)
2364 state
.offset
+= item
.size
2366 # Handles the literal string item `item`.
2367 def _handle_lit_str_item(self
, item
: _LitStr
, state
: _GenState
):
2368 self
._data
+= item
.data
2369 state
.offset
+= item
.size
2371 # Handles the byte order setting item `item`.
2372 def _handle_set_bo_item(self
, item
: _SetBo
, state
: _GenState
):
2373 # Update current byte order
2376 # Handles the variable assignment item `item`.
2377 def _handle_var_assign_item(self
, item
: _VarAssign
, state
: _GenState
):
2379 state
.variables
[item
.name
] = self
._eval
_item
_expr
(
2380 item
, state
, accept_float
=True, accept_str
=True
2383 # Returns the effective byte order to use to encode the fixed-length
2384 # number `item` considering the current state `state`.
2386 def _fl_num_item_effective_bo(item
: _FlNum
, state
: _GenState
):
2387 return state
.bo
if item
.bo
is None else item
.bo
2389 # Handles the fixed-length number item `item`.
2390 def _handle_fl_num_item(self
, item
: _FlNum
, state
: _GenState
):
2391 # Effective byte order
2392 bo
= self
._fl
_num
_item
_effective
_bo
(item
, state
)
2394 # Validate current byte order
2395 if bo
is None and item
.len > 8:
2396 _raise_error_for_item(
2397 "Current byte order isn't defined at first fixed-length number (`{}`) to encode on more than 8 bits".format(
2403 # Try an immediate evaluation. If it fails, then keep everything
2404 # needed to (try to) generate the bytes of this item later.
2406 data
= self
._gen
_fl
_num
_item
_inst
_data
(item
, state
)
2409 _raise_error_for_item(
2410 "Invalid expression `{}`: failed to evaluate within a transformation block".format(
2416 self
._fl
_num
_item
_insts
.append(
2420 copy
.deepcopy(state
),
2421 copy
.deepcopy(self
._parse
_error
_msgs
),
2425 # Reserve space in `self._data` for this instance
2426 data
= bytes([0] * (item
.len // 8))
2432 state
.offset
+= len(data
)
2434 # Returns the size, in bytes, required to encode the value `val`
2435 # with LEB128 (signed version if `is_signed` is `True`).
2437 def _leb128_size_for_val(val
: int, is_signed
: bool):
2439 # Equivalent upper bound.
2441 # For example, if `val` is -128, then the full integer for
2442 # this number of bits would be [-128, 127].
2445 # Number of bits (add one for the sign if needed)
2446 bits
= val
.bit_length() + int(is_signed
)
2451 # Seven bits per byte
2452 return math
.ceil(bits
/ 7)
2454 # Handles the LEB128 integer item `item`.
2455 def _handle_leb128_int_item(self
, item
: _Leb128Int
, state
: _GenState
):
2457 val
= self
._eval
_item
_expr
(item
, state
)
2460 size
= self
._leb
128_size
_for
_val
(val
, type(item
) is _SLeb128Int
)
2463 for _
in range(size
):
2464 # Seven LSBs, MSB of the byte set (continue)
2465 self
._data
.append((val
& 0x7F) |
0x80)
2468 # Clear MSB of last byte (stop)
2469 self
._data
[-1] &= ~
0x80
2472 state
.offset
+= size
2474 # Handles the string item `item`.
2475 def _handle_str_item(self
, item
: _Str
, state
: _GenState
):
2477 val
= str(self
._eval
_item
_expr
(item
, state
, accept_float
=True, accept_str
=True))
2480 data
= _encode_str(val
, item
.codec
, item
.text_loc
)
2486 state
.offset
+= len(data
)
2488 # Handles the group item `item`, removing the immediate labels from
2489 # `state` at the end if `remove_immediate_labels` is `True`.
2490 def _handle_group_item(
2491 self
, item
: _Group
, state
: _GenState
, remove_immediate_labels
: bool = True
2493 first_fl_num_item_inst_index
= len(self
._fl
_num
_item
_insts
)
2494 immediate_labels
= {} # type: LabelsT
2497 for subitem
in item
.items
:
2498 if type(subitem
) is _Label
:
2499 # Add to local immediate labels
2500 immediate_labels
[subitem
.name
] = state
.offset
2502 self
._handle
_item
(subitem
, state
)
2504 # Remove immediate labels from current state if needed
2505 if remove_immediate_labels
:
2506 for name
in immediate_labels
:
2507 del state
.labels
[name
]
2509 # Add all immediate labels to all state snapshots since
2510 # `first_fl_num_item_inst_index`.
2511 for inst
in self
._fl
_num
_item
_insts
[first_fl_num_item_inst_index
:]:
2512 inst
.state
.labels
.update(immediate_labels
)
2514 # Handles the repetition item `item`.
2515 def _handle_rep_item(self
, item
: _Rep
, state
: _GenState
):
2516 # Compute the repetition count
2517 mul
= _Gen
._eval
_item
_expr
(item
, state
)
2521 _raise_error_for_item(
2522 "Invalid expression `{}`: unexpected negative result {:,}".format(
2528 # Generate group data `mul` times
2529 for _
in range(mul
):
2530 self
._handle
_group
_item
(item
, state
)
2532 # Handles the conditional item `item`.
2533 def _handle_cond_item(self
, item
: _Cond
, state
: _GenState
):
2534 # Compute the conditional value
2535 val
= _Gen
._eval
_item
_expr
(item
, state
)
2537 # Generate selected group data
2539 self
._handle
_group
_item
(item
.true_item
, state
)
2541 self
._handle
_group
_item
(item
.false_item
, state
)
2543 # Handles the transformation item `item`.
2544 def _handle_trans_item(self
, item
: _Trans
, state
: _GenState
):
2545 init_in_trans
= self
._in
_trans
2546 self
._in
_trans
= True
2547 init_data_len
= len(self
._data
)
2548 init_offset
= state
.offset
2550 # Generate group data
2551 self
._handle
_group
_item
(item
, state
)
2553 # Remove and keep group data
2554 to_trans
= self
._data
[init_data_len
:]
2555 del self
._data
[init_data_len
:]
2557 # Encode group data and append to current data
2559 transformed
= item
.trans(to_trans
)
2560 except Exception as exc
:
2561 _raise_error_for_item(
2562 "Cannot apply the {} transformation to this data: {}".format(
2568 self
._data
+= transformed
2570 # Update offset and restore
2571 state
.offset
= init_offset
+ len(transformed
)
2572 self
._in
_trans
= init_in_trans
2574 # Evaluates the parameters of the macro expansion item `item`
2575 # considering the initial state `init_state` and returns a new state
2576 # to handle the items of the macro.
2577 def _eval_macro_exp_params(self
, item
: _MacroExp
, init_state
: _GenState
):
2579 exp_state
= _GenState({}, {}, init_state
.offset
, init_state
.bo
)
2581 # Evaluate the parameter expressions
2582 macro_def
= self
._macro
_defs
[item
.name
]
2584 for param_name
, param
in zip(macro_def
.param_names
, item
.params
):
2585 exp_state
.variables
[param_name
] = _Gen
._eval
_expr
(
2596 # Handles the macro expansion item `item`.
2597 def _handle_macro_exp_item(self
, item
: _MacroExp
, state
: _GenState
):
2598 parse_error_msg_text
= "While expanding the macro `{}`:".format(item
.name
)
2602 exp_state
= self
._eval
_macro
_exp
_params
(item
, state
)
2604 # Process the contained group
2605 init_data_size
= len(self
._data
)
2607 ParseErrorMessage
._create
( # pyright: ignore[reportPrivateUsage]
2608 parse_error_msg_text
, item
.text_loc
2611 self
._parse
_error
_msgs
.append(parse_error_msg
)
2612 self
._handle
_group
_item
(self
._macro
_defs
[item
.name
], exp_state
)
2613 self
._parse
_error
_msgs
.pop()
2614 except ParseError
as exc
:
2615 _augment_error(exc
, parse_error_msg_text
, item
.text_loc
)
2617 # Update state offset and return
2618 state
.offset
+= len(self
._data
) - init_data_size
2620 # Handles the offset setting item `item`.
2621 def _handle_set_offset_item(self
, item
: _SetOffset
, state
: _GenState
):
2622 state
.offset
= item
.val
2624 # Handles the offset alignment item `item` (adds padding).
2625 def _handle_align_offset_item(self
, item
: _AlignOffset
, state
: _GenState
):
2626 init_offset
= state
.offset
2627 align_bytes
= item
.val
// 8
2628 state
.offset
= (state
.offset
+ align_bytes
- 1) // align_bytes
* align_bytes
2629 self
._data
+= bytes([item
.pad_val
] * (state
.offset
- init_offset
))
2631 # Handles the filling item `item` (adds padding).
2632 def _handle_fill_until_item(self
, item
: _FillUntil
, state
: _GenState
):
2633 # Compute the new offset
2634 new_offset
= _Gen
._eval
_item
_expr
(item
, state
)
2636 # Validate the new offset
2637 if new_offset
< state
.offset
:
2638 _raise_error_for_item(
2639 "Invalid expression `{}`: new offset {:,} is less than current offset {:,}".format(
2640 item
.expr_str
, new_offset
, state
.offset
2646 self
._data
+= bytes([item
.pad_val
] * (new_offset
- state
.offset
))
2649 state
.offset
= new_offset
2651 # Handles the label item `item`.
2652 def _handle_label_item(self
, item
: _Label
, state
: _GenState
):
2653 state
.labels
[item
.name
] = state
.offset
2655 # Handles the item `item`, returning the updated next repetition
2657 def _handle_item(self
, item
: _Item
, state
: _GenState
):
2658 return self
._item
_handlers
[type(item
)](item
, state
)
2660 # Generates the data for a fixed-length integer item instance having
2661 # the value `val` and the effective byte order `bo` and returns it.
2662 def _gen_fl_int_item_inst_data(
2663 self
, val
: int, bo
: Optional
[ByteOrder
], item
: _FlNum
2666 if val
< -(2 ** (item
.len - 1)) or val
> 2**item
.len - 1:
2667 _raise_error_for_item(
2668 "Value {:,} is outside the {}-bit range when evaluating expression `{}`".format(
2669 val
, item
.len, item
.expr_str
2674 # Encode result on 64 bits (to extend the sign bit whatever the
2675 # value of `item.len`).
2678 ">" if bo
in (None, ByteOrder
.BE
) else "<",
2679 "Q" if val
>= 0 else "q",
2684 # Keep only the requested length
2685 len_bytes
= item
.len // 8
2687 if bo
in (None, ByteOrder
.BE
):
2688 # Big endian: keep last bytes
2689 data
= data
[-len_bytes
:]
2691 # Little endian: keep first bytes
2692 assert bo
== ByteOrder
.LE
2693 data
= data
[:len_bytes
]
2698 # Generates the data for a fixed-length floating point number item
2699 # instance having the value `val` and the effective byte order `bo`
2701 def _gen_fl_float_item_inst_data(
2702 self
, val
: float, bo
: Optional
[ByteOrder
], item
: _FlNum
2705 if item
.len not in (32, 64):
2706 _raise_error_for_item(
2707 "Invalid {}-bit length for a fixed-length floating point number (value {:,})".format(
2713 # Encode and return result
2716 ">" if bo
in (None, ByteOrder
.BE
) else "<",
2717 "f" if item
.len == 32 else "d",
2722 # Generates the data for a fixed-length number item instance and
2724 def _gen_fl_num_item_inst_data(self
, item
: _FlNum
, state
: _GenState
):
2725 # Effective byte order
2726 bo
= self
._fl
_num
_item
_effective
_bo
(item
, state
)
2729 val
= self
._eval
_item
_expr
(item
, state
, True)
2731 # Handle depending on type
2732 if type(val
) is int:
2733 return self
._gen
_fl
_int
_item
_inst
_data
(val
, bo
, item
)
2735 assert type(val
) is float
2736 return self
._gen
_fl
_float
_item
_inst
_data
(val
, bo
, item
)
2738 # Generates the data for all the fixed-length number item instances
2739 # and writes it at the correct offset within `self._data`.
2740 def _gen_fl_num_item_insts(self
):
2741 for inst
in self
._fl
_num
_item
_insts
:
2744 data
= self
._gen
_fl
_num
_item
_inst
_data
(inst
.item
, inst
.state
)
2745 except ParseError
as exc
:
2746 # Add all the saved parse error messages for this
2748 for msg
in reversed(inst
.parse_error_msgs
):
2749 _add_error_msg(exc
, msg
.text
, msg
.text_location
)
2753 # Insert bytes into `self._data`
2754 self
._data
[inst
.offset_in_data
: inst
.offset_in_data
+ len(data
)] = data
2756 # Generates the data (`self._data`) and final state
2757 # (`self._final_state`) from `group` and the initial state `state`.
2758 def _gen(self
, group
: _Group
, state
: _GenState
):
2760 self
._data
= bytearray()
2763 self
._item
_handlers
= {
2764 _AlignOffset
: self
._handle
_align
_offset
_item
,
2765 _Byte
: self
._handle
_byte
_item
,
2766 _Cond
: self
._handle
_cond
_item
,
2767 _FillUntil
: self
._handle
_fill
_until
_item
,
2768 _FlNum
: self
._handle
_fl
_num
_item
,
2769 _Group
: self
._handle
_group
_item
,
2770 _Label
: self
._handle
_label
_item
,
2771 _LitStr
: self
._handle
_lit
_str
_item
,
2772 _MacroExp
: self
._handle
_macro
_exp
_item
,
2773 _Rep
: self
._handle
_rep
_item
,
2774 _SetBo
: self
._handle
_set
_bo
_item
,
2775 _SetOffset
: self
._handle
_set
_offset
_item
,
2776 _SLeb128Int
: self
._handle
_leb
128_int
_item
,
2777 _Str
: self
._handle
_str
_item
,
2778 _Trans
: self
._handle
_trans
_item
,
2779 _ULeb128Int
: self
._handle
_leb
128_int
_item
,
2780 _VarAssign
: self
._handle
_var
_assign
_item
,
2781 } # type: Dict[type, Callable[[Any, _GenState], None]]
2783 # Handle the group item, _not_ removing the immediate labels
2784 # because the `labels` property offers them.
2785 self
._handle
_group
_item
(group
, state
, False)
2787 # This is actually the final state
2788 self
._final
_state
= state
2790 # Generate all the fixed-length number bytes now that we know
2792 self
._gen
_fl
_num
_item
_insts
()
2795 # Returns a `ParseResult` instance containing the bytes encoded by the
2796 # input string `normand`.
2798 # `init_variables` is a dictionary of initial variable names (valid
2799 # Python names) to integral values. A variable name must not be the
2800 # reserved name `ICITTE`.
2802 # `init_labels` is a dictionary of initial label names (valid Python
2803 # names) to integral values. A label name must not be the reserved name
2806 # `init_offset` is the initial offset.
2808 # `init_byte_order` is the initial byte order.
2810 # Raises `ParseError` on any parsing error.
2813 init_variables
: Optional
[VariablesT
] = None,
2814 init_labels
: Optional
[LabelsT
] = None,
2815 init_offset
: int = 0,
2816 init_byte_order
: Optional
[ByteOrder
] = None,
2818 if init_variables
is None:
2821 if init_labels
is None:
2824 parser
= _Parser(normand
, init_variables
, init_labels
)
2833 return ParseResult
._create
( # pyright: ignore[reportPrivateUsage]
2834 gen
.data
, gen
.variables
, gen
.labels
, gen
.offset
, gen
.bo
2838 # Raises a command-line error with the message `msg`.
2839 def _raise_cli_error(msg
: str) -> NoReturn
:
2840 raise RuntimeError("Command-line error: {}".format(msg
))
2843 # Returns the `int` or `float` value out of a CLI assignment value.
2844 def _val_from_assign_val_str(s
: str, is_label
: bool):
2847 # Floating point number?
2849 m
= _const_float_pat
.fullmatch(s
)
2852 return float(m
.group(0))
2855 m
= _const_int_pat
.fullmatch(s
)
2858 return int(_norm_const_int(m
.group(0)), 0)
2860 exp
= "an integer" if is_label
else "a number"
2861 _raise_cli_error("Invalid assignment value `{}`: expecting {}".format(s
, exp
))
2864 # Returns a dictionary of string to numbers from the list of strings
2865 # `args` containing `NAME=VAL` entries.
2866 def _dict_from_arg(args
: Optional
[List
[str]], is_label
: bool, is_str_only
: bool):
2867 d
= {} # type: VariablesT
2873 m
= re
.match(r
"({})\s*=\s*(.*)$".format(_py_name_pat
.pattern
), arg
)
2876 _raise_cli_error("Invalid assignment `{}`".format(arg
))
2881 val
= _val_from_assign_val_str(m
.group(2), is_label
)
2888 # Parses the command-line arguments and returns, in this order:
2890 # 1. The input file path, or `None` if none.
2891 # 2. The Normand input text.
2892 # 3. The initial offset.
2893 # 4. The initial byte order.
2894 # 5. The initial variables.
2895 # 6. The initial labels.
2896 def _parse_cli_args():
2900 ap
= argparse
.ArgumentParser()
2907 help="initial offset (positive)",
2913 choices
=["be", "le"],
2915 help="initial byte order (`be` or `le`)",
2922 help="add an initial numeric variable (may be repeated)",
2929 help="add an initial string variable (may be repeated)",
2936 help="add an initial label (may be repeated)",
2939 "--version", action
="version", version
="Normand {}".format(__version__
)
2946 help="input path (none means standard input)",
2950 args
= ap
.parse_args()
2953 if args
.path
is None:
2954 normand
= sys
.stdin
.read()
2956 with
open(args
.path
) as f
:
2959 # Variables and labels
2960 variables
= _dict_from_arg(args
.var
, False, False)
2961 variables
.update(_dict_from_arg(args
.var_str
, False, True))
2962 labels
= _dict_from_arg(args
.label
, True, False)
2966 _raise_cli_error("Invalid negative offset {}")
2968 # Validate and set byte order
2969 bo
= None # type: Optional[ByteOrder]
2971 if args
.byte_order
is not None:
2972 if args
.byte_order
== "be":
2975 assert args
.byte_order
== "le"
2978 # Return input and initial state
2979 return args
.path
, normand
, args
.offset
, bo
, variables
, typing
.cast(LabelsT
, labels
)
2982 # CLI entry point without exception handling.
2983 def _run_cli_with_args(
2986 bo
: Optional
[ByteOrder
],
2987 variables
: VariablesT
,
2990 sys
.stdout
.buffer.write(parse(normand
, variables
, labels
, offset
, bo
).data
)
2993 # Prints the exception message `msg` and exits with status 1.
2994 def _fail(msg
: str) -> NoReturn
:
2995 if not msg
.endswith("."):
2998 print(msg
.strip(), file=sys
.stderr
)
3005 args
= _parse_cli_args()
3006 except Exception as exc
:
3010 _run_cli_with_args(*args
[1:])
3011 except ParseError
as exc
:
3014 prefix
= "" if args
[0] is None else "{}:".format(os
.path
.abspath(args
[0]))
3017 for msg
in reversed(exc
.messages
):
3018 fail_msg
+= "{}{}:{} - {}".format(
3020 msg
.text_location
.line_no
,
3021 msg
.text_location
.col_no
,
3025 if fail_msg
[-1] not in ".:;":
3030 _fail(fail_msg
.strip())
3031 except Exception as exc
:
3035 if __name__
== "__main__":