species atom order still isn't perfect

Initializing O=CCCCCO with xyz=

O       2.21824695    1.64569701    1.58182966
C       1.85164754    1.92822298    0.44498030
C       1.80785898    0.95170228   -0.70690251
C       2.15084544   -0.48672860   -0.32326943
C       3.64378419   -0.83127781   -0.41106260
C       4.56131644   -0.07170499    0.53965886
O       4.15329388   -0.24543989    1.88652479
H       1.53434969    2.95621672    0.19807184
H       0.77868559    0.98100019   -1.08376000
H       2.46376982    1.32165670   -1.50137346
H       1.76426719   -0.73229937    0.67268406
H       1.62960788   -1.15190687   -1.02405174
H       3.74029879   -1.90314968   -0.19552773
H       3.99338137   -0.68319416   -1.43970147
H       5.58112748   -0.45943095    0.44908527
H       4.59834243    0.99685921    0.30586603
H       3.41559543    0.38462569    2.04301052

results in a wrong atom order in the molecule: the two O atoms are mixed.

The atom order according to the xyz is:

\But looking at the molecule, the connectivity is:
atom 1: {<Atom 'C'>: <Bond "1.0">, <Atom 'H'>: <Bond "1.0">}
atom 6: {<Atom 'C'>: <Bond "2.0">}

Start by reproducing the reported initialization of O=CCCCCO from the supplied xyz coordinates and inspect the atom-ordering and connectivity path; no source file or test is named in the issue. Done means the two oxygen atoms retain the correct xyz order while the molecule connectivity remains correct for this input.