Merge pull request 'feat(ui): אינדיקטור התקדמות לחילוץ מטא-דאטה + מתג-מקטעים בספריית הפסיקה' (#70) from worktree-feat+metadata-extraction-progress into main

mcp-server/src/legal_mcp/services/db.py

@@ -610,6 +610,11 @@ ALTER TABLE case_law ADD COLUMN IF NOT EXISTS document_id UUID REFERENCES docume
 ALTER TABLE case_law ADD COLUMN IF NOT EXISTS extraction_status TEXT DEFAULT 'pending';
     -- 'pending' | 'processing' | 'completed' | 'failed'
 ALTER TABLE case_law ADD COLUMN IF NOT EXISTS halacha_extraction_status TEXT DEFAULT 'pending';
+ALTER TABLE case_law ADD COLUMN IF NOT EXISTS metadata_extraction_status TEXT DEFAULT 'pending';
+    -- 'pending' | 'processing' | 'completed' | 'failed'. Mirrors the
+    -- text/halacha status columns so the UI can show a live badge while the
+    -- local-MCP worker drains the metadata queue (previously only the
+    -- metadata_extraction_requested_at timestamp existed — no 'processing').
 ALTER TABLE case_law ADD COLUMN IF NOT EXISTS practice_area TEXT DEFAULT '';
 ALTER TABLE case_law ADD COLUMN IF NOT EXISTS appeal_subtype TEXT DEFAULT '';
 ALTER TABLE case_law ADD COLUMN IF NOT EXISTS headnote TEXT DEFAULT '';
@@ -3070,6 +3075,27 @@ async def set_case_law_halacha_status(case_law_id: UUID, status: str) -> None:
         )
 
 
+async def set_case_law_metadata_status(case_law_id: UUID, status: str) -> None:
+    """Set metadata-extraction status. Mirrors ``set_case_law_halacha_status``:
+    on terminal states ('completed'/'failed') we also clear
+    ``metadata_extraction_requested_at`` so the local-MCP queue
+    (`process_pending_extractions`, which scans ``WHERE *_requested_at IS NOT
+    NULL``) stops re-picking the row and the UI's ``isPrecedentActive`` check
+    settles."""
+    pool = await get_pool()
+    if status in ("completed", "failed"):
+        await pool.execute(
+            "UPDATE case_law SET metadata_extraction_status = $2, "
+            "metadata_extraction_requested_at = NULL WHERE id = $1",
+            case_law_id, status,
+        )
+    else:
+        await pool.execute(
+            "UPDATE case_law SET metadata_extraction_status = $2 WHERE id = $1",
+            case_law_id, status,
+        )
+
+
 async def list_external_case_law(
     practice_area: str = "",
     court: str = "",
@@ -3126,6 +3152,7 @@ async def list_external_case_law(
                summary, headnote, subject_tags, source_kind,
                chair_name, district, citation_formatted,
                extraction_status, halacha_extraction_status,
+               metadata_extraction_status,
                metadata_extraction_requested_at,
                halacha_extraction_requested_at,
                created_at,
@@ -4274,8 +4301,12 @@ async def request_metadata_extraction(case_law_id: UUID) -> bool:
     fills empty fields), so this is safe.
     """
     pool = await get_pool()
+    # Reset the status to 'pending' alongside the timestamp so a re-request
+    # after a prior 'completed'/'failed' run shows "בתור" again in the UI
+    # instead of a stale terminal badge.
     result = await pool.execute(
-        "UPDATE case_law SET metadata_extraction_requested_at = now() "
+        "UPDATE case_law SET metadata_extraction_requested_at = now(), "
+        "metadata_extraction_status = 'pending' "
         "WHERE id = $1",
         case_law_id,
     )

mcp-server/src/legal_mcp/services/precedent_library.py

@@ -235,6 +235,12 @@ async def process_pending_extractions(kind: str = "metadata", limit: int = 20) -
         attempts = 0
         result: dict = {}
         try:
+            # Flip to 'processing' so the UI badge shows live progress while
+            # this row is being worked (metadata has no per-chunk status of
+            # its own — this is the only signal). Halacha already sets its own
+            # 'processing' inside the extractor.
+            if kind == "metadata":
+                await db.set_case_law_metadata_status(cid, "processing")
             result = await _run_once(cid)
             # Retry only on systematic extraction failure (rate-limit storm).
             # Don't retry on 'no_halachot' — that means Claude looked and
@@ -259,9 +265,15 @@ async def process_pending_extractions(kind: str = "metadata", limit: int = 20) -
             # Finalise: success or terminal failure both clear the request
             # so the queue moves on. (Use 'failed' DB state for terminal
             # extraction_failed so the UI shows the warning chip.)
-            if kind == "halacha" and result.get("status") == "extraction_failed":
+            if kind == "halacha":
-                await db.set_case_law_halacha_status(cid, "failed")
+                if result.get("status") == "extraction_failed":
-            await db.clear_extraction_request(cid, kind=kind)
+                    await db.set_case_law_halacha_status(cid, "failed")
+                await db.clear_extraction_request(cid, kind=kind)
+            else:
+                # metadata — set terminal 'completed' status (also clears the
+                # request timestamp) so the UI badge settles instead of
+                # lingering on 'processing'.
+                await db.set_case_law_metadata_status(cid, "completed")
             processed += 1
             results.append({
                 "case_law_id": str(cid),
@@ -273,6 +285,15 @@ async def process_pending_extractions(kind: str = "metadata", limit: int = 20) -
             })
         except Exception as e:
             logger.exception("process_pending_extractions failed for %s: %s", cid, e)
+            # Don't clear the request — it stays for the next run. But for
+            # metadata, revert the badge from 'processing' back to 'pending'
+            # (the timestamp is preserved) so the row shows "בתור" rather than
+            # a stuck "מחלץ" until the retry picks it up.
+            if kind == "metadata":
+                try:
+                    await db.set_case_law_metadata_status(cid, "pending")
+                except Exception:
+                    logger.exception("failed to revert metadata status for %s", cid)
             results.append({
                 "case_law_id": str(cid),
                 "case_number": row.get("case_number", ""),
@@ -280,7 +301,6 @@ async def process_pending_extractions(kind: str = "metadata", limit: int = 20) -
                 "error": str(e),
                 "retry_attempts": attempts,
             })
-            # Don't clear the request — it stays for the next run.
 
     return {
         "status": "completed",
@@ -314,12 +334,18 @@ async def reextract_metadata(
         raise ValueError("precedent not found")
     # See note in db.request_metadata_extraction — opened to all source kinds.
 
+    # Mark 'processing' so a concurrent UI poll shows the live badge.
+    await db.set_case_law_metadata_status(case_law_id, "processing")
     await progress("extracting_metadata", 40, "מחלץ מטא-דאטה (תקציר, תגיות)")
     result = await precedent_metadata_extractor.extract_and_apply(case_law_id)
-    # Clear the queue timestamp so the UI / worker stop showing this row.
+    # Settle to terminal 'completed' (also NULLs the queue timestamp) so the
-    # See note in reextract_halachot.
+    # UI / worker stop showing this row. See note in reextract_halachot.
     if result.get("status") in ("completed", "no_changes"):
-        await db.clear_extraction_request(case_law_id, kind="metadata")
+        await db.set_case_law_metadata_status(case_law_id, "completed")
+    else:
+        # e.g. 'no_metadata' (no full_text) — don't leave the badge stuck on
+        # 'processing'; revert to 'pending' (preserves any queue timestamp).
+        await db.set_case_law_metadata_status(case_law_id, "pending")
     fields = result.get("fields") or []
     msg = (
         f"מולאו {len(fields)} שדות: {', '.join(fields)}"

web-ui/src/components/precedents/library-list-panel.tsx

@@ -2,7 +2,7 @@
 
 import { useState } from "react";
 import Link from "next/link";
-import { Trash2, Plus, Pencil, Wand2 } from "lucide-react";
+import { Trash2, Plus, Pencil, Wand2, Loader2 } from "lucide-react";
 import { toast } from "sonner";
 import {
   Table, TableBody, TableCell, TableHead, TableHeader, TableRow,
@@ -10,7 +10,9 @@ import {
 import { Button } from "@/components/ui/button";
 import { Badge } from "@/components/ui/badge";
 import { Input } from "@/components/ui/input";
+import { Progress } from "@/components/ui/progress";
 import { Skeleton } from "@/components/ui/skeleton";
+import { cn } from "@/lib/utils";
 import {
   Select, SelectContent, SelectItem, SelectTrigger, SelectValue,
 } from "@/components/ui/select";
@@ -87,6 +89,28 @@ function StatusPill({ p }: { p: Precedent }) {
       </Badge>
     );
   }
+  // Metadata extraction (orthogonal one-off re-extract). Surface it while
+  // active so the chair sees the queued button actually ran on something.
+  if (p.metadata_extraction_status === "processing") {
+    return <ActivePill label="מחלץ מטא-דאטה" />;
+  }
+  if (p.metadata_extraction_status === "failed") {
+    return (
+      <Badge variant="outline" className="bg-danger-bg text-danger border-danger/40">
+        מטא-דאטה נכשל
+      </Badge>
+    );
+  }
+  if (
+    p.metadata_extraction_requested_at &&
+    p.metadata_extraction_status !== "completed"
+  ) {
+    return (
+      <Badge variant="outline" className="bg-rule-soft text-ink-muted">
+        ממתין למטא-דאטה
+      </Badge>
+    );
+  }
   if (p.halacha_extraction_status === "processing") {
     return <ActivePill label="מחלץ הלכות" />;
   }
@@ -325,6 +349,92 @@ function CommitteeRow({ p, onEdit }: { p: Precedent; onEdit: (id: string) => voi
   );
 }
 
+/**
+ * Live banner for the metadata-extraction queue. The local-MCP worker
+ * (`precedent_process_pending kind=metadata`) drains stamped rows serially;
+ * while it runs, rows carry metadata_extraction_status='processing' (current
+ * item) or a request timestamp (still queued). We surface a real, advancing
+ * percentage by tracking the session high-water-mark of the queue depth:
+ * percent = (peak - remaining) / peak. Hidden when the queue is empty.
+ */
+function MetadataQueueBanner({ items }: { items: Precedent[] }) {
+  const processing = items.filter(
+    (p) => p.metadata_extraction_status === "processing",
+  ).length;
+  const pending = items.filter(
+    (p) =>
+      p.metadata_extraction_status !== "processing" &&
+      p.metadata_extraction_requested_at !== null &&
+      p.metadata_extraction_status !== "completed" &&
+      p.metadata_extraction_status !== "failed",
+  ).length;
+  const remaining = processing + pending;
+
+  // Session high-water-mark of the queue depth, using React's documented
+  // "adjust state during render" pattern (no effect, no ref-in-render). React
+  // re-renders immediately without committing the intermediate UI. Resets to 0
+  // when the queue drains, so the next batch starts a fresh 0→100% sweep.
+  const [peak, setPeak] = useState(0);
+  if (remaining === 0) {
+    if (peak !== 0) setPeak(0);
+  } else if (remaining > peak) {
+    setPeak(remaining);
+  }
+
+  if (remaining === 0) return null;
+  const done = Math.max(0, peak - remaining);
+  const percent = peak > 0 ? Math.round((done / peak) * 100) : 0;
+
+  return (
+    <div className="rounded-lg border border-gold/40 bg-gold-wash/50 p-4 space-y-2">
+      <div className="flex items-center gap-2 text-sm text-gold-deep">
+        <Loader2 className="w-4 h-4 animate-spin shrink-0" />
+        <span>
+          חילוץ מטא-דאטה פעיל — מעבד {processing}, נותרו {pending} בתור
+          {peak > 0 ? ` · ${done}/${peak}` : ""}
+        </span>
+      </div>
+      <Progress value={percent} className="h-1.5" />
+    </div>
+  );
+}
+
+function SegButton({
+  active, onClick, label, count,
+}: {
+  active: boolean;
+  onClick: () => void;
+  label: string;
+  count?: number;
+}) {
+  return (
+    <button
+      type="button"
+      onClick={onClick}
+      aria-pressed={active}
+      className={cn(
+        "flex items-center gap-2 rounded-md px-4 py-1.5 text-sm font-semibold transition-colors",
+        active
+          ? "bg-surface text-navy shadow-sm"
+          : "text-ink-muted hover:text-navy",
+      )}
+    >
+      {label}
+      {count !== undefined && (
+        <Badge
+          variant="outline"
+          className={cn(
+            "text-[0.72rem]",
+            active ? "border-navy/30 text-navy" : "text-ink-muted",
+          )}
+        >
+          {count}
+        </Badge>
+      )}
+    </button>
+  );
+}
+
 function TableSkeleton({ cols }: { cols: number }) {
   return (
     <>
@@ -344,6 +454,9 @@ export function LibraryListPanel() {
   const [search, setSearch] = useState("");
   const [uploadOpen, setUploadOpen] = useState(false);
   const [editingId, setEditingId] = useState<string | null>(null);
+  // Which of the two corpora to show. Both queries stay mounted (counts +
+  // cross-table polling), but only the active table renders — no long scroll.
+  const [view, setView] = useState<"court" | "committee">("court");
 
   const sharedFilters = {
     practiceArea: practiceArea || undefined,
@@ -354,6 +467,12 @@ export function LibraryListPanel() {
   const courts = usePrecedents({ ...sharedFilters, sourceKind: "external_upload", sourceType: "court_ruling" });
   const committee = usePrecedents({ ...sharedFilters, sourceKind: "all_committees" });
 
+  // Metadata-queue banner draws on both corpora (a stamped row can be either).
+  const allItems = [
+    ...(courts.data?.items ?? []),
+    ...(committee.data?.items ?? []),
+  ];
+
   return (
     <div className="space-y-8">
       {/* Shared filters */}
@@ -385,16 +504,28 @@ export function LibraryListPanel() {
         </Button>
       </div>
 
+      {/* Metadata-extraction queue progress (hidden when idle) */}
+      <MetadataQueueBanner items={allItems} />
+
+      {/* Segmented control — show one corpus at a time */}
+      <div className="inline-flex rounded-lg border border-rule bg-rule-soft/40 p-1 gap-1">
+        <SegButton
+          active={view === "court"}
+          onClick={() => setView("court")}
+          label="פסיקת בתי משפט"
+          count={courts.data?.count}
+        />
+        <SegButton
+          active={view === "committee"}
+          onClick={() => setView("committee")}
+          label="החלטות ועדות ערר"
+          count={committee.data?.count}
+        />
+      </div>
+
       {/* Table 1 — Court rulings */}
+      {view === "court" && (
       <section>
-        <div className="flex items-center gap-2 mb-3">
-          <h3 className="text-base font-semibold text-navy">פסיקת בתי משפט</h3>
-          {courts.data && (
-            <Badge variant="outline" className="text-ink-muted text-[0.72rem]">
-              {courts.data.count}
-            </Badge>
-          )}
-        </div>
         {courts.error ? (
           <div className="rounded bg-danger-bg border border-danger/40 px-6 py-4 text-danger text-center text-sm">
             {courts.error.message}
@@ -433,17 +564,11 @@ export function LibraryListPanel() {
           </div>
         )}
       </section>
+      )}
 
       {/* Table 2 — Appeals committee decisions */}
+      {view === "committee" && (
       <section>
-        <div className="flex items-center gap-2 mb-3">
-          <h3 className="text-base font-semibold text-navy">החלטות ועדות ערר</h3>
-          {committee.data && (
-            <Badge variant="outline" className="text-ink-muted text-[0.72rem]">
-              {committee.data.count}
-            </Badge>
-          )}
-        </div>
         {committee.error ? (
           <div className="rounded bg-danger-bg border border-danger/40 px-6 py-4 text-danger text-center text-sm">
             {committee.error.message}
@@ -483,6 +608,7 @@ export function LibraryListPanel() {
           </div>
         )}
       </section>
+      )}
 
       <PrecedentUploadSheet open={uploadOpen} onOpenChange={setUploadOpen} />
       <PrecedentEditSheet

web-ui/src/lib/api/precedent-library.ts

@@ -51,6 +51,7 @@ export type Precedent = {
   citation_formatted: string;
   extraction_status: string;
   halacha_extraction_status: string;
+  metadata_extraction_status: string;
   metadata_extraction_requested_at: string | null;
   halacha_extraction_requested_at: string | null;
   created_at: string;
@@ -232,12 +233,14 @@ export function isPrecedentActive(p: Precedent): boolean {
     return true;
   }
 
-  // Metadata extraction has no status column — only the timestamp.
+  // Metadata extraction now has its own status column. Active while the
-  // Treat as active only when extraction hasn't yet fully completed
+  // worker is processing the row, or while it's queued (timestamp set) and
-  // (otherwise stale timestamps linger after success).
+  // hasn't reached a terminal state yet.
+  if (p.metadata_extraction_status === "processing") return true;
   if (
     p.metadata_extraction_requested_at !== null &&
-    p.extraction_status !== "completed"
+    p.metadata_extraction_status !== "completed" &&
+    p.metadata_extraction_status !== "failed"
   ) {
     return true;
   }