#Corpus

Manage conlang texts and Leipzig-standard interlinear glossing in a sidebar + editor panel layout.

Navigate: click Corpus in the sidebar.

#Text Management

• Create/edit/delete corpus texts
• Metadata: title, description, tags, timestamps
• Raw text (in your conlang) and free translation

#Interlinear Glossing

Leipzig glossing standard annotations:

• Auto-gloss: Raw text → split by sentences → split by spaces → auto-match from lexicon
• Manual editing: Each token can be edited independently
  • Surface form
  • Morpheme break (use - to separate morphemes)
  • Gloss labels
  • Linked entry ID
  • IPA

#Auto-gloss engine (advanced)

The auto-gloss pipeline supports:

• exact lexicon match
• concatenative reverse parsing (multi-prefix + stem + multi-suffix)
• non-concatenative candidates (infix/circumfix/reduplication/ablaut)
• confidence-based auto-apply + pending review queue

#Per-corpus report persistence

Auto-gloss report is stored per corpus item:

• total tokens
• auto-applied
• pending review
• unresolved

Switching to another corpus no longer reuses a previous report.

#Steps

1. Click "New" → enter title and description
2. Write a passage in your language in Raw Text
3. Write the natural language translation in Free Translation
4. Click "Auto-gloss" — the app tries to match each word from the lexicon
5. Manually edit unmatched tokens

#Step-by-step: Review pending suggestions

1. Run Auto-gloss.
2. Open the pending suggestions panel.
3. Check confidence and trace columns.
4. Apply one-by-one or click Apply All.
5. Re-check unresolved token count in report.

#Apply SCA to corpus with diff (advanced)

Use Preview & Apply SCA to Corpus to evolve corpus text directly.

#What diff includes

• scope: original text / gloss-line original / token surface
• before -> after values
• context snippet
• checkbox selection (default all selected)

#Step-by-step: Safe corpus evolution

1. Click Preview & Apply SCA to Corpus.
2. Review the diff table.
3. Deselect changes you do not want.
4. Click Apply Selected Changes.
5. The system automatically runs auto-gloss again on updated corpus.

#Close behavior

Use the red circular close icon at the top-right corner of the panel to dismiss preview anytime.

#Lexicon Integration

After linking a token to a lexicon entry, hovering shows entry details (POS, definition, IPA).

#Advanced example: diachronic corpus pass

Scenario: you maintain a proto text and evolve it for a daughter language.

1. Prepare baseline corpus in parent language.
2. Configure SCA in daughter language.
3. Run corpus SCA diff preview and apply selected changes.
4. Let auto-gloss rerun automatically.
5. Resolve remaining low-confidence items.