Skip to content

Fixed handling of the char 'K' during Append edits #2

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
lunolunoluno wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Fixed handling of the char 'K' during Append edits #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d7b38ad
Fixed handling of the char 'K' during Append edits
lunolunoluno Apr 1, 2026
48b2e2d
Fix unintended deletion of 'NIL' and 'PNX' substrings during word/cha…
lunolunoluno Apr 13, 2026
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
4 changes: 2 additions & 2 deletions edits/alignment/aligner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -442,8 +442,8 @@ def reduce_inserts_deletions(alignment):
assert len(example['src']) == len(example['tgt'])
src, tgt = example['src'], example['tgt']

src = [x.replace('PNX', '').replace('NIL','') for x in src]
tgt = [x.replace('PNX', '').replace('NIL','') for x in tgt]
src = [re.sub(r'\b(NIL|PNX)\b', '', x) for x in src]
tgt = [re.sub(r'\b(NIL|PNX)\b', '', x) for x in tgt]

i = 0
s_idx = 0
Expand Down
9 changes: 8 additions & 1 deletion edits/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -166,7 +166,14 @@ def insert_to_append(edits):

# Special case for appends at the beginning of the sequence
if processed_edits[0].startswith('A') and re.sub(r'A_\[.*?\]', '', processed_edits[0]) == 'K':
processed_edits[0] = processed_edits[0].replace('K', 'K' * len(subwords[0].replace('##', '')))
# Calculate length of kept section
keep_section = 'K' * len(subwords[0].replace('##', ''))
# Split the edit by bracketed segments
edit_sections = re.split(r'(\[[^\]]*\])', processed_edits[0])
# Replace 'K' only outside of the brackets
for i in range(0, len(edit_sections), 2):
edit_sections[i] = edit_sections[i].replace('K', keep_section)
processed_edits[0] = "".join(edit_sections)

processed_edits = [SubwordEdit(subword, raw_subword, edit)
for subword, raw_subword, edit in zip(subwords, raw_subwords, processed_edits)]
Expand Down