Fix: ValueError when x_dtype != y_dtype with numpy 2.x

[yousuke08]

Problem

When parsing LTspice RAW files where x_dtype (time data type) differs from y_dtype (variable data type), the library raises a ValueError: ValueError: setting an array element with a sequence.
TypeError: only 0-dimensional arrays can be converted to Python scalars

Root Cause

In parse() method, when x_dtype != y_dtype, the code tries to assign an array (returned by np.frombuffer()) to a scalar position (self.x_raw[i]):

self.x_raw[i] = np.frombuffer(d, dtype=self._x_dtype)
This fails with numpy 2.x due to stricter type checking.

Solution

Changed the approach to concatenate byte buffers first, then convert to array in one operation:

time_buffer = b""
for i in range(self._point_num):
    start = i * (self._variable_num + diff) * variable_data_size
    end = start + time_data_size
    time_buffer += data[start:end]
self.x_raw = np.frombuffer(time_buffer, dtype=self._x_dtype)

Testing

• Tested with a RAW file (119,238 points, 14 signals) where x_dtype=float64 and y_dtype=float32
• Successfully parsed all signals.
• Tested with Python 3.14, numpy 2.4.2

Environment

• Python 3.14.2
• numpy 2.4.2
• ltspice 1.0.6

---

[mimeiners]

Thx for that fix @yousuke08, works for me with Python 3.13.13 and numpy 2.4.4
Please, merge this into main branch soon @donghoonpark.