Skip to content

[datasets] Reject negative indices in decode_sequence #2189

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
felixdittrich92 merged 1 commit into from
Oct 30, 2026
Merged

[datasets] Reject negative indices in decode_sequence #2189

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 5 additions & 4 deletions doctr/datasets/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -111,10 +111,11 @@ def decode_sequence(
"""
if not isinstance(input_seq, (Sequence, np.ndarray)):
raise TypeError("Invalid sequence type")
if isinstance(input_seq, np.ndarray) and (
not np.issubdtype(input_seq.dtype, np.integer) or (input_seq.size > 0 and input_seq.max() >= len(mapping))
):
raise AssertionError("Input must be an array of int, with max less than mapping size")
if isinstance(input_seq, np.ndarray) and not np.issubdtype(input_seq.dtype, np.integer):
raise AssertionError("Input must be an array of int, with values between 0 and the mapping size")
# A negative index would silently wrap around to the end of the mapping (e.g. an EOS of -1 read as a char)
if len(input_seq) > 0 and (min(input_seq) < 0 or max(input_seq) >= len(mapping)):
raise AssertionError("Input must be an array of int, with values between 0 and the mapping size")

return "".join(map(mapping.__getitem__, input_seq))

Expand Down
7 changes: 7 additions & 0 deletions tests/common/test_datasets_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,6 +61,13 @@ def test_decode_sequence():
utils.decode_sequence(np.array([2, 10]), mapping)
with pytest.raises(AssertionError):
utils.decode_sequence(np.array([2, 4.5]), mapping)
# negative indices (such as the default EOS of -1) are not wrapped around to the last chars of the mapping
with pytest.raises(AssertionError):
utils.decode_sequence(np.array([2, -1]), mapping)
with pytest.raises(AssertionError):
utils.decode_sequence([2, -1], mapping)
with pytest.raises(AssertionError):
utils.decode_sequence([2, 10], mapping)

assert utils.decode_sequence([3, 4, 3, 4], mapping) == "dede"
assert utils.decode_sequence(np.array([3, 4, 3, 4], dtype=np.int64), mapping) == "dede"
Expand Down
Loading