disabled
 ?÷     """Implements a Jinja / Python combination lexer. The ``Lexer`` class
is used to do some preprocessing. It filters out invalid operators like
the bitshift operators we don't allow in templates. It separates
template code and python code in expressions.
"""
import re
import typing as t
from ast import literal_eval
try:
    from collections.abc import deque
except ImportError:
    from collections import deque
from sys import intern

from ._identifier import pattern as name_re
from .exceptions import TemplateSyntaxError
from .utils import LRUCache

if t.TYPE_CHECKING:
    import typing_extensions as te
    from .environment import Environment

# cache for the lexers. Exists in order to be able to have multiple
# environments with the same lexer
_lexer_cache: t.MutableMapping[t.Tuple, "Lexer"] = LRUCache(50)  # type: ignore

# static regular expressions
whitespace_re = re.compile(r"\s+")
newline_re = re.compile(r"(\r\n|\r|\n)")
string_re = re.compile(
    r"('([^'\\]*(?:\\.[^'\\]*)*)'" r'|"([^"\\]*(?:\\.[^"\\]*)*)")', re.S
)
integer_re = re.compile(
    r"""
    (
        0b(_?[0-1])+ # binary
    |
        0o(_?[0-7])+ # octal
    |
        0x(_?[\da-f])+ # hex
    |
        [1-9](_?\d)* # decimal
    |
        0(_?0)* # decimal zero
    )
    """,
    re.IGNORECASE | re.VERBOSE,
)
float_re = re.compile(
    r"""
    (?<!\.)  # doesn't start with a .
    (\d+_)*\d+  # digits, possibly _ separated
    (
        (\.(\d+_)*\d+)?  # optional fractional part
        e[+\-]?(\d+_)*\d+  # exponent part
    |
        \.(\d+_)*\d+  # required fractional part
    )
    """,
    re.IGNORECASE | re.VERBOSE,
)

# internal the tokens and keep references to them
TOKEN_ADD = intern("add")
TOKEN_ASSIGN = intern("assign")
TOKEN_COLON = intern("colon")
TOKEN_COMMA = intern("comma")
TOKEN_DIV = intern("div")
TOKEN_DOT = intern("dot")
TOKEN_EQ = intern("eq")
TOKEN_FLOORDIV = intern("floordiv")
TOKEN_GT = intern("gt")
TOKEN_GTEQ = intern("gteq")
TOKEN_LBRACE = intern("lbrace")
TOKEN_LBRACKET = intern("lbracket")
TOKEN_LPAREN = intern("lparen")
TOKEN_LT = intern("lt")
TOKEN_LTEQ = intern("lteq")
TOKEN_MOD = intern("mod")
TOKEN_MUL = intern("mul")
TOKEN_NE = intern("ne")
TOKEN_PIPE = intern("pipe")
TOKEN_POW = intern("pow")
TOKEN_RBRACE = intern("rbrace")
TOKEN_RBRACKET = intern("rbracket")
TOKEN_RPAREN = intern("rparen")
TOKEN_SEMICOLON = intern("semicolon")
TOKEN_SUB = intern("sub")
TOKEN_TILDE = intern("tilde")
TOKEN_WHITESPACE = intern("whitespace")
TOKEN_FLOAT = intern("float")
TOKEN_INTEGER = intern("integer")
TOKEN_NAME = intern("name")
TOKEN_STRING = intern("string")
TOKEN_OPERATOR = intern("operator")
TOKEN_BLOCK_BEGIN = intern("block_begin")
TOKEN_BLOCK_END = intern("block_end")
TOKEN_VARIABLE_BEGIN = intern("variable_begin")
TOKEN_VARIABLE_END = intern("variable_end")
TOKEN_RAW_BEGIN = intern("raw_begin")
TOKEN_RAW_END = intern("raw_end")
TOKEN_COMMENT_BEGIN = intern("comment_begin")
TOKEN_COMMENT_END = intern("comment_end")
TOKEN_COMMENT = intern("comment")
TOKEN_LINESTATEMENT_BEGIN = intern("linestatement_begin")
TOKEN_LINESTATEMENT_END = intern("linestatement_end")
TOKEN_LINECOMMENT_BEGIN = intern("linecomment_begin")
TOKEN_LINECOMMENT_END = intern("linecomment_end")
TOKEN_LINECOMMENT = intern("linecomment")
TOKEN_DATA = intern("data")
TOKEN_INITIAL = intern("initial")
TOKEN_EOF = intern("eof")

# bind operators to token types
operators = {
    "+": TOKEN_ADD,
    "-": TOKEN_SUB,
    "/": TOKEN_DIV,
    "//": TOKEN_FLOORDIV,
    "*": TOKEN_MUL,
    "%": TOKEN_MOD,
    "**": TOKEN_POW,
    "~": TOKEN_TILDE,
    "[": TOKEN_LBRACKET,
    "]": TOKEN_RBRACKET,
    "(": TOKEN_LPAREN,
    ")": TOKEN_RPAREN,
    "{": TOKEN_LBRACE,
    "}": TOKEN_RBRACE,
    "==": TOKEN_EQ,
    "!=": TOKEN_NE,
    ">": TOKEN_GT,
    ">=": TOKEN_GTEQ,
    "<": TOKEN_LT,
    "<=": TOKEN_LTEQ,
    "=": TOKEN_ASSIGN,
    ".": TOKEN_DOT,
    ":": TOKEN_COLON,
    "|": TOKEN_PIPE,
    ",": TOKEN_COMMA,
    ";": TOKEN_SEMICOLON,
}

reverse_operators = {v: k for k, v in operators.items()}
assert len(operators) == len(reverse_operators), "operators 