A question about how SMILES is being parsed

[Shredderroy]

Assume I have constructed a molecule from the following SMILES:

using Graphs, MolecularGraph

sml = "CC(=O)OC1=CC=CC=C1C(=O)O";
mol = smilestomol(sml);

We see in Pluto that mol has one ring.

Now, I would like to extract all the bonds that belong to the ring:

ringbonds = [(s, t) for (s, (t, _)) in zip(is_edge_in_ring(mol), [(k, v) for (k, v) in mol.eprops]) if s == 1];

#=
6-element Vector{Tuple{Bool, MolecularGraph.EdgeKey{Int64}}}:
 (1, MolecularGraph.EdgeKey{Int64}(Edge 8 => 9))
 (1, MolecularGraph.EdgeKey{Int64}(Edge 6 => 7))
 (1, MolecularGraph.EdgeKey{Int64}(Edge 5 => 6))
 (1, MolecularGraph.EdgeKey{Int64}(Edge 4 => 5))
 (1, MolecularGraph.EdgeKey{Int64}(Edge 1 => 2))
 (1, MolecularGraph.EdgeKey{Int64}(Edge 9 => 10))
=#

However, as the image indicates, (1, 2) and (4, 5) are not ring bonds. Is this an error, or have I misunderstood the purpose and nature of is_edge_in_ring?

Thank you in advance for your help.