fix memory remove dup

Signed-off-by: ale <ale@manalejandro.com>

app/page.tsx

@@ -1,7 +1,7 @@
 'use client';
 
-import { useState, useEffect, useCallback } from 'react';
-import { useSearchParams, useRouter } from 'next/navigation';
+import { useState, useEffect, useCallback, Suspense } from 'react';
+import { useSearchParams } from 'next/navigation';
 import { Search, Copy, Check, Hash, Key, AlertCircle, Loader2, Database, Link } from 'lucide-react';
 
 interface SearchResult {
@@ -46,9 +46,8 @@ function formatNumber(num: number): string {
   return num.toLocaleString();
 }
 
-export default function Home() {
+function HasherContent() {
   const searchParams = useSearchParams();
-  const router = useRouter();
   const [query, setQuery] = useState('');
   const [result, setResult] = useState<SearchResult | null>(null);
   const [loading, setLoading] = useState(false);
@@ -56,8 +55,9 @@ export default function Home() {
   const [copiedField, setCopiedField] = useState<string | null>(null);
   const [stats, setStats] = useState<IndexStats | null>(null);
   const [copiedLink, setCopiedLink] = useState(false);
+  const [initialLoadDone, setInitialLoadDone] = useState(false);
 
-  const performSearch = useCallback(async (searchQuery: string) => {
+  const performSearch = useCallback(async (searchQuery: string, updateUrl: boolean = true) => {
     if (!searchQuery.trim()) return;
 
     setLoading(true);
@@ -78,25 +78,30 @@ export default function Home() {
       const data = await response.json();
       setResult(data);
       
-      // Update URL with search query
+      // Update URL with search query (using history API to avoid re-triggering effects)
+      if (updateUrl) {
         const newUrl = new URL(window.location.href);
         newUrl.searchParams.set('q', searchQuery.trim());
-      router.replace(newUrl.pathname + newUrl.search, { scroll: false });
+        window.history.replaceState(null, '', newUrl.pathname + newUrl.search);
+      }
     } catch (_err) {
       setError('Failed to perform search. Please check your connection.');
     } finally {
       setLoading(false);
     }
-  }, [router]);
+  }, []);
 
-  // Load query from URL on mount
+  // Load query from URL on mount (only once)
   useEffect(() => {
+    if (initialLoadDone) return;
+    
     const urlQuery = searchParams.get('q');
     if (urlQuery) {
       setQuery(urlQuery);
-      performSearch(urlQuery);
+      performSearch(urlQuery, false);
     }
-  }, [searchParams, performSearch]);
+    setInitialLoadDone(true);
+  }, [searchParams, performSearch, initialLoadDone]);
 
   useEffect(() => {
     const fetchStats = async () => {
@@ -361,3 +366,19 @@ export default function Home() {
   );
 }
 
+function LoadingFallback() {
+  return (
+    <div className="min-h-screen bg-gradient-to-br from-blue-50 via-white to-purple-50 flex items-center justify-center">
+      <Loader2 className="w-12 h-12 text-blue-600 animate-spin" />
+    </div>
+  );
+}
+
+export default function Home() {
+  return (
+    <Suspense fallback={<LoadingFallback />}>
+      <HasherContent />
+    </Suspense>
+  );
+}
+

scripts/remove-duplicates.ts

@@ -225,38 +225,41 @@ async function findDuplicatesForField(
   return duplicates;
 }
 
-async function removeDuplicates(parsedArgs: ParsedArgs) {
-  const client = new Client({ node: ELASTICSEARCH_NODE });
-  const fields = parsedArgs.field 
-    ? [parsedArgs.field] 
-    : ['plaintext', 'md5', 'sha1', 'sha256', 'sha512'];
-
-  console.log(`🔍 Hasher Duplicate Remover`);
+/**
+ * Phase 1: Initialize and connect to Elasticsearch
+ */
+async function phase1_InitAndConnect() {
+  console.log(`🔍 Hasher Duplicate Remover - Phase 1: Initialization`);
   console.log(`━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━`);
   console.log(`Elasticsearch: ${ELASTICSEARCH_NODE}`);
   console.log(`Index: ${INDEX_NAME}`);
-  console.log(`Mode: ${parsedArgs.dryRun ? '🔎 DRY RUN (no changes)' : '⚠️  EXECUTE (will delete)'}`);
-  console.log(`Batch size: ${parsedArgs.batchSize}`);
-  console.log(`Fields to check: ${fields.join(', ')}`);
   console.log('');
 
-  try {
-    // Test connection
+  const client = new Client({ node: ELASTICSEARCH_NODE });
+
   console.log('🔗 Connecting to Elasticsearch...');
   await client.cluster.health({});
   console.log('✅ Connected successfully\n');
 
-    // Get index stats
   const countResponse = await client.count({ index: INDEX_NAME });
   console.log(`📊 Total documents in index: ${countResponse.count}\n`);
 
-    const allDuplicates: DuplicateGroup[] = [];
-    const seenDeleteIds = new Set<string>();
+  return { client, totalDocuments: countResponse.count };
+}
 
-    // Find duplicates for each field
-    for (const field of fields) {
-      console.log(`🔍 Checking duplicates for field: ${field}...`);
-      const fieldDuplicates = await findDuplicatesForField(client, field, parsedArgs.batchSize);
+/**
+ * Phase 2: Find duplicates for a specific field
+ */
+async function phase2_FindDuplicatesForField(
+  client: Client,
+  field: string,
+  batchSize: number,
+  seenDeleteIds: Set<string>
+): Promise<{ duplicates: DuplicateGroup[], totalFound: number }> {
+  console.log(`\n🔍 Phase 2: Checking duplicates for field: ${field}...`);
+  
+  const fieldDuplicates = await findDuplicatesForField(client, field, batchSize);
+  const duplicates: DuplicateGroup[] = [];
   
   // Filter out already seen delete IDs to avoid counting the same document multiple times
   for (const dup of fieldDuplicates) {
@@ -264,11 +267,124 @@ async function removeDuplicates(parsedArgs: ParsedArgs) {
     if (newDeleteIds.length > 0) {
       dup.deleteIds = newDeleteIds;
       newDeleteIds.forEach(id => seenDeleteIds.add(id));
-          allDuplicates.push(dup);
+      duplicates.push(dup);
     }
   }
   
   console.log(`   Found ${fieldDuplicates.length} duplicate groups for ${field}`);
+  console.log(`   New unique documents to delete: ${duplicates.reduce((sum, dup) => sum + dup.deleteIds.length, 0)}`);
+  
+  // Force garbage collection if available
+  if (global.gc) {
+    global.gc();
+    console.log(`   ♻️  Memory freed after processing ${field}`);
+  }
+  
+  return { duplicates, totalFound: fieldDuplicates.length };
+}
+
+/**
+ * Phase 3: Process deletion for a batch of duplicates
+ */
+async function phase3_DeleteBatch(
+  client: Client,
+  deleteIds: string[],
+  batchSize: number,
+  startIndex: number
+): Promise<{ deleted: number, errors: number }> {
+  const batch = deleteIds.slice(startIndex, startIndex + batchSize);
+  let deleted = 0;
+  let errors = 0;
+
+  try {
+    const bulkOperations = batch.flatMap(id => [
+      { delete: { _index: INDEX_NAME, _id: id } }
+    ]);
+
+    const bulkResponse = await client.bulk({
+      operations: bulkOperations,
+      refresh: false
+    });
+
+    if (bulkResponse.errors) {
+      const errorCount = bulkResponse.items.filter((item: any) => item.delete?.error).length;
+      errors += errorCount;
+      deleted += batch.length - errorCount;
+    } else {
+      deleted += batch.length;
+    }
+  } catch (error) {
+    console.error(`\n❌ Error deleting batch:`, error);
+    errors += batch.length;
+  }
+
+  // Force garbage collection if available
+  if (global.gc) {
+    global.gc();
+  }
+
+  return { deleted, errors };
+}
+
+/**
+ * Phase 4: Finalize and report results
+ */
+async function phase4_Finalize(
+  client: Client,
+  totalDeleted: number,
+  totalErrors: number,
+  initialDocumentCount: number
+) {
+  console.log('\n\n🔄 Phase 4: Refreshing index...');
+  await client.indices.refresh({ index: INDEX_NAME });
+
+  const newCountResponse = await client.count({ index: INDEX_NAME });
+
+  console.log('\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━');
+  console.log('✅ Duplicate removal complete!');
+  console.log(`━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━`);
+  console.log(`Documents deleted: ${totalDeleted}`);
+  console.log(`Errors: ${totalErrors}`);
+  console.log(`Previous document count: ${initialDocumentCount}`);
+  console.log(`New document count: ${newCountResponse.count}`);
+  console.log('');
+}
+
+async function removeDuplicates(parsedArgs: ParsedArgs) {
+  const fields = parsedArgs.field 
+    ? [parsedArgs.field] 
+    : ['plaintext', 'md5', 'sha1', 'sha256', 'sha512'];
+
+  console.log(`Mode: ${parsedArgs.dryRun ? '🔎 DRY RUN (no changes)' : '⚠️  EXECUTE (will delete)'}`);
+  console.log(`Batch size: ${parsedArgs.batchSize}`);
+  console.log(`Fields to check: ${fields.join(', ')}`);
+  console.log('');
+
+  try {
+    // === PHASE 1: Initialize ===
+    const { client, totalDocuments } = await phase1_InitAndConnect();
+    
+    // Force garbage collection after phase 1
+    if (global.gc) {
+      global.gc();
+      console.log('♻️  Memory freed after initialization\n');
+    }
+
+    // === PHASE 2: Find duplicates field by field ===
+    const allDuplicates: DuplicateGroup[] = [];
+    const seenDeleteIds = new Set<string>();
+
+    for (const field of fields) {
+      const { duplicates } = await phase2_FindDuplicatesForField(
+        client,
+        field,
+        parsedArgs.batchSize,
+        seenDeleteIds
+      );
+      allDuplicates.push(...duplicates);
+      
+      // Clear field duplicates to free memory
+      duplicates.length = 0;
     }
 
     const totalToDelete = allDuplicates.reduce((sum, dup) => sum + dup.deleteIds.length, 0);
@@ -310,57 +426,40 @@ async function removeDuplicates(parsedArgs: ParsedArgs) {
       return;
     }
 
-    // Execute deletion
-    console.log(`\n🗑️  Removing ${totalToDelete} duplicate documents...\n`);
+    // === PHASE 3: Execute deletion in batches ===
+    console.log(`\n🗑️  Phase 3: Removing ${totalToDelete} duplicate documents...\n`);
 
-    let deleted = 0;
-    let errors = 0;
+    let totalDeleted = 0;
+    let totalErrors = 0;
     const deleteIds = allDuplicates.flatMap(dup => dup.deleteIds);
 
-    // Delete in batches
+    // Clear allDuplicates to free memory
+    allDuplicates.length = 0;
+
+    // Delete in batches with memory management
     for (let i = 0; i < deleteIds.length; i += parsedArgs.batchSize) {
-      const batch = deleteIds.slice(i, i + parsedArgs.batchSize);
+      const { deleted, errors } = await phase3_DeleteBatch(
+        client,
+        deleteIds,
+        parsedArgs.batchSize,
+        i
+      );
       
-      try {
-        const bulkOperations = batch.flatMap(id => [
-          { delete: { _index: INDEX_NAME, _id: id } }
-        ]);
+      totalDeleted += deleted;
+      totalErrors += errors;
 
-        const bulkResponse = await client.bulk({
-          operations: bulkOperations,
-          refresh: false
-        });
-
-        if (bulkResponse.errors) {
-          const errorCount = bulkResponse.items.filter((item: any) => item.delete?.error).length;
-          errors += errorCount;
-          deleted += batch.length - errorCount;
-        } else {
-          deleted += batch.length;
+      process.stdout.write(
+        `\r⏳ Progress: ${Math.min(i + parsedArgs.batchSize, deleteIds.length)}/${deleteIds.length} - ` +
+        `Deleted: ${totalDeleted}, Errors: ${totalErrors}`
+      );
     }
 
-        process.stdout.write(`\r⏳ Progress: ${Math.min(i + parsedArgs.batchSize, deleteIds.length)}/${deleteIds.length} - Deleted: ${deleted}, Errors: ${errors}`);
-      } catch (error) {
-        console.error(`\n❌ Error deleting batch:`, error);
-        errors += batch.length;
-      }
-    }
+    // Clear deleteIds to free memory
+    deleteIds.length = 0;
+    seenDeleteIds.clear();
 
-    // Refresh index
-    console.log('\n\n🔄 Refreshing index...');
-    await client.indices.refresh({ index: INDEX_NAME });
-
-    // Get new count
-    const newCountResponse = await client.count({ index: INDEX_NAME });
-
-    console.log('\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━');
-    console.log('✅ Duplicate removal complete!');
-    console.log(`━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━`);
-    console.log(`Documents deleted: ${deleted}`);
-    console.log(`Errors: ${errors}`);
-    console.log(`Previous document count: ${countResponse.count}`);
-    console.log(`New document count: ${newCountResponse.count}`);
-    console.log('');
+    // === PHASE 4: Finalize ===
+    await phase4_Finalize(client, totalDeleted, totalErrors, totalDocuments);
 
   } catch (error) {
     console.error('\n❌ Error:', error instanceof Error ? error.message : error);