Decode to errors seg fault from wrong input type

[oscarhiggott]

I accidentally gave decode_to_errors a bitstring of detector outcomes rather than indices of detection events, and got a segmentation fault. Example:

from tesseract_decoder import tesseract
import stim


dem = stim.DetectorErrorModel("""
    error(0.1) D0 D1 L0
    error(0.2) D1 D2 L1
    detector(0, 0, 0) D0
    detector(1, 0, 0) D1
    detector(2, 0, 0) D2
""")
config = tesseract.TesseractConfig(dem=dem, det_beam=50)
decoder = config.compile_decoder()
syndrome = [0, 1, 1]
decoder.decode_to_errors(syndrome)
# Segmentation fault (core dumped)

Either this should be handled by an exception correctly, or we should switch to have decode_to_errors accept a bitstring of detector outcomes, which would make it consistent with the decode method. WDYT?

[oscarhiggott]

@LalehB @noajshu would you agree it makes sense to have decode_to_errors accept a bitstring of detector outcomes to make it consistent with the decode method? (I'm referring to the python methods here, not the C++ ones)

[LalehB]

@oscarhiggott if it is more intuitive for the researchers I agree to change it (since it is in the python interface)

[noajshu]

could you see if https://github.com/quantumlib/tesseract-decoder/pull/111/files fixes this?