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.

web/src/pages/crawler/sourceStatus.ts

@@ -15,7 +15,41 @@ export function sourceStatus(source: SourceInfo, restarting = false, now = Date.
     return { type: "primary", text: "采集中" } as const
   if (source.consecutive_failures)
     return { type: "danger", text: "失败待重试" } as const
-  if (source.collection_stats?.document_gaps)
-    return { type: "warning", text: "存在文档缺口" } as const
+  if (source.collection_stats?.document_gaps || source.collection_stats?.attachment_gaps || source.collection_stats?.unresolved_bills)
+    return { type: "warning", text: "存在采集缺口" } as const
   return { type: "info", text: "等待调度" } as const
 }
+
+function originalUrl(url: string) {
+  const bill = /^https:\/\/api\.congress\.gov\/v3\/bill\/(\d+)\/([a-z]+)\/(\d+)\/?(?:\?[^#]*)?(?:#.*)?$/i.exec(url)
+  const billTypes: Record<string, string> = {
+    hr: "house-bill",
+    s: "senate-bill",
+    hjres: "house-joint-resolution",
+    sjres: "senate-joint-resolution",
+    hconres: "house-concurrent-resolution",
+    sconres: "senate-concurrent-resolution",
+    hres: "house-resolution",
+    sres: "senate-resolution"
+  }
+  if (bill && billTypes[bill[2].toLowerCase()])
+    return `https://www.congress.gov/bill/${bill[1]}th-congress/${billTypes[bill[2].toLowerCase()]}/${bill[3]}`
+  return url
+}
+
+export function collectionGaps(source: Pick<SourceInfo, "history_cursor">) {
+  const groups = [
+    ["document_retries", "正文"],
+    ["attachment_retries", "附件"],
+    ["bill_retries", "法案"]
+  ] as const
+  return groups.flatMap(([key, label]) => Object.entries(source.history_cursor?.[key] ?? {}).map(([url, retry]) => ({
+    key: `${key}:${url}`,
+    label,
+    url: originalUrl(retry.document.canonical ?? retry.document.url),
+    error: retry.error || "历史记录未保存具体原因，请核对原文",
+    attempts: retry.attempts,
+    exhausted: retry.attempts >= 3,
+    nextAt: new Date(retry.next_at * 1000).toISOString()
+  })))
+}