একটি লেক্সার বানানো

KEYWORDS = {"if", "else", "while", "return"}

def match_identifier(s, i):
    if i >= len(s) or not (s[i].isalpha() or s[i] == '_'):
        return 0
    j = i
    while j < len(s) and (s[j].isalnum() or s[j] == '_'):
        j += 1
    return j - i

def match_number(s, i):
    if i >= len(s) or not s[i].isdigit():
        return 0
    j = i
    while j < len(s) and s[j].isdigit():
        j += 1
    return j - i

MULTI_OPS = [">=", "<=", "==", "!="]
SINGLE_OPS = set("+-*/()<>=,;")

def match_operator(s, i):
    for op in MULTI_OPS:
        if s.startswith(op, i):
            return len(op)
    if i < len(s) and s[i] in SINGLE_OPS:
        return 1
    return 0

def match_whitespace(s, i):
    j = i
    while j < len(s) and s[j] in " \t":
        j += 1
    return j - i

class LexError(Exception):
    pass

def tokenize(s):
    tokens = []
    i = 0
    while i < len(s):
        ws_len = match_whitespace(s, i)
        if ws_len > 0:
            i += ws_len
            continue

        candidates = []
        id_len = match_identifier(s, i)
        if id_len > 0:
            candidates.append(("IDENTIFIER", id_len))
        num_len = match_number(s, i)
        if num_len > 0:
            candidates.append(("NUMBER", num_len))
        op_len = match_operator(s, i)
        if op_len > 0:
            candidates.append(("OPERATOR", op_len))

        if not candidates:
            raise LexError(f"lexical error at position {i}: unrecognized character {s[i]!r}")

        # maximal munch: সবচেয়ে দীর্ঘ ম্যাচ বেছে নাও
        max_len = max(length for _, length in candidates)
        best = [kind for kind, length in candidates if length == max_len]
        kind = best[0]
        lexeme = s[i:i + max_len]

        # keyword priority: IDENTIFIER-এর সমান দৈর্ঘ্যে KEYWORD জিতবে
        if kind == "IDENTIFIER" and lexeme in KEYWORDS:
            kind = "KEYWORD"

        tokens.append((kind, lexeme))
        i += max_len
    return tokens

src = "if x12 >= 5"
print(f"tokenize({src!r}):")
for tok in tokenize(src):
    print(f"  {tok}")

bad_src = "if x12 @ 5"
print(f"\ntokenize({bad_src!r}) -- অচেনা ক্যারেক্টার '@' সহ:")
try:
    tokenize(bad_src)
    print("  ত্রুটি রিপোর্ট হয়নি -- এটি একটি বাগ হতো!")
except LexError as e:
    print(f"  সঠিকভাবে LexError রিপোর্ট হলো: {e}")