import json
import os
import string

DATASET_SIZE = 120

MIN_INPUT_LENGTH = 2
MAX_INPUT_LENGTH = 20

MIN_SUBMISSION_SIZE = 1
MAX_SUBMISSION_SIZE = 1024 * 1024 * 120  # 120 MB.
MAX_SINGLE_SUBMISSION_SIZE = 1024 * 1024  # 1MB.
MAX_SUBMISSION_LINES = DATASET_SIZE + 1  # Allow empty line.


def is_valid(
    s: str,
    min_length: int = MIN_INPUT_LENGTH,
    max_length: int = MAX_INPUT_LENGTH,
) -> bool:
    """
    @brief Checks whether the given string is valid.
    @param s The string to validate.
    @return True iff all characters are in [a-zA-Z0-9], spaces, or '.' and '-', and the length if between
            min length and max length.
    """

    characters = [c for c in s]  # Not using the length from len(.) as that includes unicode characters.
    if len(characters) < min_length or len(characters) > max_length:
        return False

    # Very important: We delimit using underscores. So these _CANNOT_ be allowed in sanitised strings.
    ALLOWED = (
        [c for c in string.ascii_lowercase]
        + [c for c in string.ascii_uppercase]
        + [c for c in string.digits]
        + [" ", ".", "-"]
    )
    for c in s:
        if c not in ALLOWED:
            return False
    return True


def is_submission_file_valid(submission_path: str) -> bool:
    """
    @brief Checks whether the given submission file is valid.
    @param submission_path The path to the submission file.
    @return True iff the file is within the size constraints, a JSONL, and every line is no longer than
            the fixed maximum bound.
    """

    if not os.path.exists(submission_path):
        return False

    submission_size = os.stat(submission_path).st_size
    if submission_size < MIN_SUBMISSION_SIZE or submission_size > MAX_SUBMISSION_SIZE:
        return False

    with open(submission_path, "r") as f:

        # Not using readlines() to avoid consuming a large buffer at once.
        n_lines = 0
        seen_ids = set()
        while len(line := f.readline(MAX_SINGLE_SUBMISSION_SIZE)) > 0:
            n_lines += 1
            if n_lines > MAX_SUBMISSION_LINES:
                return False

            if not line.startswith("{") or not line.endswith("}"):
                return False

            d = json.loads(line)
            if set(d.keys()) != set(["problem_id", "solution"]):
                return False

            if not ((type(d["problem_id"]) is str or type(d["problem_id"]) is int) and type(d["solution"] is str)):
                return False
            if not d["problem_id"].isdigit():
                return False
            problem_id = int(d["problem_id"])
            if problem_id < 0 or problem_id >= DATASET_SIZE:
                return False

            if problem_id in seen_ids:
                return False  # Duplicate submission.
            seen_ids.add(problem_id)

    return True