donnemartin/interactive-coding-challenges · error · TypeError

matrix and val cannot be None

Error message

matrix and val cannot be None

What it means

SortedMatrix.find_val raises TypeError('matrix and val cannot be None') when either the matrix or the searched value is None. The staircase search immediately indexes matrix[0], so a None matrix would crash with TypeError anyway; the guard also rejects None targets since equality comparison against None is meaningless here.

Source

Thrown at sorting_searching/search_sorted_matrix/search_sorted_matrix_solution.ipynb:117

  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "class SortedMatrix(object):\n",
    "\n",
    "    def find_val(self, matrix, val):\n",
    "        if matrix is None or val is None:\n",
    "            raise TypeError('matrix and val cannot be None')\n",
    "        row = 0\n",
    "        col = len(matrix[0]) - 1\n",
    "        while row < len(matrix) and col >= 0:\n",
    "            if matrix[row][col] == val:\n",
    "                return (row, col)\n",
    "            elif matrix[row][col] > val:\n",
    "                col -= 1\n",
    "            else:\n",
    "                row += 1\n",
    "        return None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Unit Test"
   ]

View on GitHub (pinned to 358f2cc604)

Solutions

1. Validate that both matrix and val are non-None before calling find_val
2. Trace where val is produced and supply a default or fail early if missing
3. Also verify the matrix is non-empty before calling, to avoid the separate IndexError on matrix[0]

Example fix

// before
pos = SortedMatrix().find_val(matrix, val)  # val may be None

// after
pos = SortedMatrix().find_val(matrix, val) if (matrix and val is not None) else None

if matrix is None or val is None or not matrix or not matrix[0]:
    return None
return SortedMatrix().find_val(matrix, val)

def is_searchable_matrix(m):
    return (isinstance(m, list) and len(m) > 0
            and isinstance(m[0], list) and m[0])

try:
    pos = sm.find_val(matrix, val)
except TypeError as e:
    if 'cannot be None' in str(e):
        pos = None
    else:
        raise

When it happens

Trigger: Calling find_val(None, x) or find_val(matrix, None); also when matrix is an empty list the code raises IndexError rather than this error, so this error specifically means one argument was None.

Common situations: Search targets sourced from optional request parameters or dict lookups that returned None; matrices loaded from files/APIs that failed and defaulted to None.

Related errors

• array cannot be None
• sentence cannot be None
• rows and cols cannot be None
• s or t cannot be None
• a or b cannot be None

AI-assisted analysis of donnemartin/interactive-coding-challenges@358f2cc604 (2026-08-28). Data as JSON: /api/errors/bb8ce746d20be899. Report an issue: GitHub.