Skip to content

Verse segmentation fix #883

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

Open
laura-burdick-sil wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Verse segmentation fix #883

Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
be5f4f3
Fixing the case where there's a passage that's a single verse (previo…
laura-burdick-sil Dec 2, 2025
e95592b
Includes an option to output the verses in vref format (so that they …
laura-burdick-sil Dec 6, 2025
43af2c7
Changing to original versification before mapping verses to vref format
laura-burdick-sil Dec 8, 2025
7e64b09
Using to_versification rather than change_versification
laura-burdick-sil Dec 9, 2025
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
25 changes: 24 additions & 1 deletion silnlp/alignment/segment_verses.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -181,6 +181,12 @@ def _create_target_verses_from_offsets(
) -> SegmentedPassage:
target_verses = []

# Special case where passage is a single verse
if len(target_verse_offsets) == 0:
verse_ref = references[0]
target_verses.append(Verse(verse_ref, target_text))
return self._adjust_verse_boundaries(target_verses)

current_verse_starting_char_index = 0
current_verse_ending_char_index = 0
current_verse_offset_index = 0
Expand Down Expand Up @@ -284,7 +290,7 @@ def predict_target_verse_token_offsets(
if len(best_split_indices) > 1:
best_split_index = best_split_indices[0]
for split_index in best_split_indices:
if not contains_letter(target_tokens[split_index-1]):
if not contains_letter(target_tokens[split_index - 1]):
best_split_index = split_index
break
target_verse_offsets.append(best_split_index)
Expand Down Expand Up @@ -696,6 +702,7 @@ def main() -> None:
parser.add_argument(
"--use-saved-alignments", help="Use pre-computed alignments from a previous run", default=None, action="store_true"
)
parser.add_argument("--vref", help="Output vref file for target verses", default=None, action="store_true")
args = parser.parse_args()

parallel_passages = ParallelPassageCollectionFactory(args.save_alignments, args.use_saved_alignments).create(
Expand All @@ -713,6 +720,22 @@ def main() -> None:
src_passage.write_to_file(src_output)
trg_passage.write_to_file(trg_output)

if args.vref is not None:
vref_path = Path(args.target_passages).with_suffix(".vref.txt")
template_vref_path = SIL_NLP_ENV.assets_dir / "vref.txt"

verse_map: Dict[str, str] = {
str(verse.reference): verse.text for trg_passage in trg_segmented_passages for verse in trg_passage.verses
}

with (
open(template_vref_path, "r", encoding="utf-8") as template_file,
open(vref_path, "w", encoding="utf-8") as vref_output,
):
for line in template_file:
template_ref = line.rstrip("\n")
vref_output.write(f"{verse_map.get(template_ref, '')}\n")

if args.compare_against is not None:
reference_segmentations = ReferenceVerseSegmentationReader().read_passages(
args.compare_against, Path(args.target_passages)
Expand Down