"""Regular expressions for matching tokens of different types in kat script.
Any new token types added here should be added to `tokenizer.py` and `highlighter.py` as
well.
"""
# NOTE: do not run black on this file!
[docs]def group(*choices):
"""Match any of the specified choices."""
return f"({'|'.join(choices)})"
[docs]def any(*choices):
"""Specify that there should be zero or more of the specified choices."""
return f"{group(*choices)}*"
[docs]def maybe(*choices):
"""Specify that there should be zero or one of the specified choices."""
return f"{group(*choices)}?"
WHITESPACE = r"[ \f\t]+"
NEWLINE = r"\r?\n+"
COMMENT = r"#[^\r\n]*"
NONE = group(r"None", r"none")
FIXME = r"__FIXME__"
BOOLEAN = group(r"True", r"true", r"False", r"false")
NAME = r"&?[a-zA-Z_][a-zA-Z0-9_.]*" # Element name or property.
# Most of these are based on those in the Python `tokenize` module.
_INTEGER_NUMBER = r"(?:0(?:_?0)*|[1-9](?:_?[0-9])*)"
_EXPONENT = r"[eE][-+]?[0-9](?:_?[0-9])*"
_POINT_FLOAT = group(r"[0-9](?:_?[0-9])*\.(?:[0-9](?:_?[0-9])*)?", r"\.[0-9](?:_?[0-9])*") + maybe(_EXPONENT)
_EXPONENT_FLOAT = r"[0-9](?:_?[0-9])*" + _EXPONENT
_FLOAT_NUMBER = group(_POINT_FLOAT, _EXPONENT_FLOAT)
_SI_NUMBER = group(_INTEGER_NUMBER, _POINT_FLOAT) + r"[pnumkMGT]"
_INFINITY = r"inf"
_IMAGINARY_NUMBER = group(_INFINITY + r"[jJ]", r"[0-9](?:_?[0-9])*[jJ]", _FLOAT_NUMBER + r"[jJ]")
NUMBER = group(_IMAGINARY_NUMBER, _SI_NUMBER, _FLOAT_NUMBER, _INTEGER_NUMBER, _INFINITY)
_SINGLE_QUOTED_STRING = r"'[^\n'\\]*(?:\\.[^\n'\\]*)*'"
_DOUBLE_QUOTED_STRING = r'"[^\n"\\]*(?:\\.[^\n"\\]*)*"'
STRING = group(_SINGLE_QUOTED_STRING, _DOUBLE_QUOTED_STRING)
# Types that only take one form. The order here is important (highest priority first).
LITERALS = {
"EQUALS": "=",
"PLUS": "+",
"MINUS": "-",
"POWER": "**", # Has to be above TIMES.
"TIMES": "*",
"FLOORDIVIDE": "//", # Has to be above DIVIDE.
"DIVIDE": "/",
"COMMA": ",",
"LBRACKET": "[",
"RBRACKET": "]",
"LPAREN": "(",
"RPAREN": ")",
}