Expose persistent collection gaps and public Congress links

Retain bill and attachment failure reasons and derive retry counts from saved cursors. Show gap details and retry exhaustion in the collection console while preserving historical records and retry limits.

Convert Congress API bill links to public pages and cover diagnostics, mixed dated versions, persistent counts and legacy records with regression tests.

backend/src/alphapulse/api/crawler.py

@@ -194,6 +194,8 @@ async def list_sources(session: Annotated[AsyncSession, Depends(get_session)]) -
         pending = sum(value['attempts'] < 3 for value in retries.values())
         documents = (source.history_cursor or {}).get('document_retries', {})
         document_pending = sum(value['attempts'] < 3 for value in documents.values())
+        bills = (source.history_cursor or {}).get('bill_retries', {})
+        bill_pending = sum(value['attempts'] < 3 for value in bills.values())
         attachment_stats[source.code] = {
             'attachment_gaps': len(retries),
             'attachment_retry_pending': pending,
@@ -201,6 +203,9 @@ async def list_sources(session: Annotated[AsyncSession, Depends(get_session)]) -
             'document_gaps': len(documents),
             'document_retry_pending': document_pending,
             'document_retry_exhausted': len(documents) - document_pending,
+            'unresolved_bills': len(bills),
+            'bill_retry_pending': bill_pending,
+            'bill_retry_exhausted': len(bills) - bill_pending,
         }
     return success(SourcesOut(
         manager_alive=bool(statuses),

backend/src/alphapulse/crawler/spiders/us_official.py

@@ -493,11 +493,13 @@ class USOfficialSpider(NewsSpider):
                 for version in data['textVersions']:
                     if not version.get('date'):
                         self.cursor[lane]['bill']['missing_material'] = True
+                        self.cursor[lane]['bill'].setdefault('material_errors', []).append(f"{version.get('type', '正文版本')} 缺少官方日期")
                         continue
                     formats = version.get('formats', [])
                     chosen = next((f for f in formats if f['type'] == 'Formatted Text'), None) or next((f for f in formats if f['type'] == 'Formatted XML'), None) or next((f for f in formats if f['type'] == 'PDF'), None)
                     if not chosen:
                         self.cursor[lane]['bill']['missing_material'] = True
+                        self.cursor[lane]['bill'].setdefault('material_errors', []).append(f"{version.get('type', '正文版本')} 缺少支持的正文格式")
                     if chosen:
                         children.append({'url': chosen['url'], 'kind': 'text', 'title': doc['title'] + ' — ' + version['type'], 'at': version['date'], 'bill': doc['bill']})
             next_url = data.get('pagination', {}).get('next')
@@ -613,7 +615,10 @@ class USOfficialSpider(NewsSpider):
         retries = self.cursor.setdefault('bill_retries', {})
         if not context['bodies'] or context.get('missing_material'):
             previous = retries.get(context['url'], {})
-            retries[context['url']] = {'document': {'url': context['url'], 'kind': 'bill'}, 'version': context['version'], 'attempts': previous.get('attempts', 0) + 1, 'next_at': (datetime.now(UTC) + timedelta(hours=1)).timestamp()}
+            errors = list(context.get('material_errors', []))
+            if not context['bodies']:
+                errors.append('没有可用的官方正文')
+            retries[context['url']] = {'document': {'url': context['url'], 'kind': 'bill'}, 'version': context['version'], 'attempts': previous.get('attempts', 0) + 1, 'next_at': (datetime.now(UTC) + timedelta(hours=1)).timestamp(), 'error': '；'.join(dict.fromkeys(errors)) or '官方正文材料不完整'}
         else:
             retries.pop(context['url'], None)
             if not context['reasons']:
@@ -637,7 +642,8 @@ class USOfficialSpider(NewsSpider):
         retries = self.cursor.setdefault('attachment_retries', {})
         if incomplete and retry_doc:
             previous = retries.get(parsed['url'], {})
-            retries[parsed['url']] = {'document': retry_doc, 'attempts': previous.get('attempts', 0) + 1, 'next_at': (datetime.now(UTC) + timedelta(hours=1)).timestamp()}
+            error = '；'.join(f"{a['url']}：{a.get('error') or '附件尚未完成文字提取'}" for a in parsed['extra']['attachments'] if a['status'] != 'complete')
+            retries[parsed['url']] = {'document': retry_doc, 'attempts': previous.get('attempts', 0) + 1, 'next_at': (datetime.now(UTC) + timedelta(hours=1)).timestamp(), 'error': error}
         elif not incomplete:
             retries.pop(parsed['url'], None)
         if reasons: