Every now and then Overflow errors pop up in the issue tracker, that I can't reproduce locally with the spectrometers available to me.
My current guess is that the issue is that the transport class should check the endpoint size and read data in the provided packet size. Under some situations (firmware? hardware version?) the hardcoded values seem to mismatch with the actual endpoint size. Packets are zero padded by libusb. So we can just read whatever size is reported there.
Hopefully this should fix those issues.
Every now and then Overflow errors pop up in the issue tracker, that I can't reproduce locally with the spectrometers available to me.
My current guess is that the issue is that the transport class should check the endpoint size and read data in the provided packet size. Under some situations (firmware? hardware version?) the hardcoded values seem to mismatch with the actual endpoint size. Packets are zero padded by libusb. So we can just read whatever size is reported there.
Hopefully this should fix those issues.