Normalize Chinese Part ordinals and support resumable complete-book translation

src/translation/glossary.py

@@ -2,7 +2,11 @@
 import hashlib
 import json
 import re
+import copy
 from pathlib import Path
+from src.common.utils import setup_logger
+
+logger = setup_logger('glossary')
 
 
 def source_digest(entries):
@@ -101,12 +105,16 @@ async def prepare_glossary(entries, llm, path, override=None, auto=True, chunk_c
     """Scan the whole book before translation; never race glossary updates."""
     path = Path(path)
     digest = source_digest(entries)
+    terms, conflicts, start = [], [], 0
     if override:
         return BookGlossary.load(override)
     if path.exists():
         data = json.loads(path.read_text())
         if data.get('source_digest') == digest:
-            return BookGlossary(data['terms'])
+            if data.get('scan_complete', True):
+                return BookGlossary(data['terms'])
+            if data.get('chunk_chars') == chunk_chars and data.get('model') == getattr(llm, 'model', ''):
+                terms, conflicts, start = data['terms'], data.get('conflicts', []), data.get('next_chunk', 0)
     if not auto:
         return BookGlossary()
     chunks, current, size = [], [], 0
@@ -116,8 +124,9 @@ async def prepare_glossary(entries, llm, path, override=None, auto=True, chunk_c
         current.append(entry.original_text); size += len(entry.original_text)
     if current:
         chunks.append('\n'.join(current))
-    terms, conflicts = [], []
-    for chunk in chunks:
+    path.parent.mkdir(parents=True, exist_ok=True)
+    for index in range(start, len(chunks)):
+        chunk = chunks[index]
         prompt = ('Extract recurring named people, organizations and specialist terminology. '
                   'Resolve transliterations across Japanese/English/Chinese. Return only JSON '
                   '{"terms":[{"source":"English name","target":"Chinese canonical name",'
@@ -127,21 +136,35 @@ async def prepare_glossary(entries, llm, path, override=None, auto=True, chunk_c
                   json.dumps(terms, ensure_ascii=False) + '\nExcerpt:\n' + chunk)
         response = await llm.raw_chat_completion('You are a book terminology editor.', prompt)
         response = re.sub(r'^```(?:json)?\s*|\s*```$', '', response.strip())
-        proposals = BookGlossary(json.loads(response)['terms'])
-        for term in proposals.terms:
+        # Validate proposals individually: a contradictory model suggestion must
+        # not discard the already frozen terms or stop the entire book scan.
+        for proposal in json.loads(response)['terms']:
+            try:
+                term = BookGlossary([proposal]).terms[0]
+            except (ValueError, AttributeError) as exc:
+                conflicts.append({'proposal': proposal, 'reason': str(exc)})
+                continue
             aliases = {a.casefold() for a in [term['source']] + term.get('aliases', [])}
             existing = next((t for t in terms if aliases & {a.casefold() for a in [t['source']] + t.get('aliases', [])}), None)
             if existing and existing['target'] != term['target']:
                 conflicts.append({'existing': existing, 'proposal': term})
                 continue
+            candidate = copy.deepcopy(terms)
             if existing:
-                existing['aliases'] = sorted(set(existing.get('aliases', []) + [term['source']] + term.get('aliases', [])) - {existing['source']})
+                candidate[terms.index(existing)]['aliases'] = sorted(set(existing.get('aliases', []) + [term['source']] + term.get('aliases', [])) - {existing['source']})
             else:
-                terms.append(term)
-        BookGlossary(terms)  # Fail on alias collisions before saving.
-    path.parent.mkdir(parents=True, exist_ok=True)
-    temp = path.with_suffix('.tmp')
-    temp.write_text(json.dumps({'source_digest': digest, 'terms': terms, 'conflicts': conflicts},
-                               ensure_ascii=False, indent=2), encoding='utf-8')
-    temp.replace(path)
+                candidate.append(term)
+            try:
+                BookGlossary(candidate)
+            except ValueError as exc:
+                conflicts.append({'proposal': term, 'reason': str(exc)})
+                continue
+            terms = candidate
+        temp = path.with_suffix('.tmp')
+        temp.write_text(json.dumps({'source_digest': digest, 'terms': terms, 'conflicts': conflicts,
+                                   'scan_complete': index + 1 == len(chunks), 'next_chunk': index + 1,
+                                   'chunk_chars': chunk_chars, 'model': getattr(llm, 'model', '')},
+                                  ensure_ascii=False, indent=2), encoding='utf-8')
+        temp.replace(path)
+        logger.info('Terminology scan %s/%s: %s terms, %s conflicts recorded', index + 1, len(chunks), len(terms), len(conflicts))
     return BookGlossary(terms)