fix: address Greptile review — deduplicate relatedTests, hoist prepared stmts, fix .raw() no-op

- queries.js: restore DISTINCT-by-file deduplication for relatedTests in
  explainFunctionImpl (lost when switching to findCallers)
- build-stmts.js: prepare SQL statements once in preparePurgeStmts() and
  loop with runPurge() instead of calling db.prepare() per file per table
- cfg.js: replace misleading .raw() with .get() in hasCfgTables

Impact: 7 functions changed, 14 affected

Author: carlos-alm

src/db/repository/build-stmts.js

@@ -1,82 +1,94 @@
+/**
+ * Prepare all purge statements once, returning an object of runnable stmts.
+ * Optional tables are wrapped in try/catch — if the table doesn't exist,
+ * that slot is set to null.
+ *
+ * @param {object} db - Open read-write database handle
+ * @returns {object} prepared statements (some may be null)
+ */
+function preparePurgeStmts(db) {
+  const tryPrepare = (sql) => {
+    try {
+      return db.prepare(sql);
+    } catch {
+      return null;
+    }
+  };
+
+  return {
+    embeddings: tryPrepare(
+      'DELETE FROM embeddings WHERE node_id IN (SELECT id FROM nodes WHERE file = ?)',
+    ),
+    cfgEdges: tryPrepare(
+      'DELETE FROM cfg_edges WHERE function_node_id IN (SELECT id FROM nodes WHERE file = ?)',
+    ),
+    cfgBlocks: tryPrepare(
+      'DELETE FROM cfg_blocks WHERE function_node_id IN (SELECT id FROM nodes WHERE file = ?)',
+    ),
+    dataflow: tryPrepare(
+      'DELETE FROM dataflow WHERE source_id IN (SELECT id FROM nodes WHERE file = ?) OR target_id IN (SELECT id FROM nodes WHERE file = ?)',
+    ),
+    complexity: tryPrepare(
+      'DELETE FROM function_complexity WHERE node_id IN (SELECT id FROM nodes WHERE file = ?)',
+    ),
+    nodeMetrics: tryPrepare(
+      'DELETE FROM node_metrics WHERE node_id IN (SELECT id FROM nodes WHERE file = ?)',
+    ),
+    astNodes: tryPrepare('DELETE FROM ast_nodes WHERE file = ?'),
+    // Core tables — always exist
+    edges: db.prepare(
+      'DELETE FROM edges WHERE source_id IN (SELECT id FROM nodes WHERE file = @f) OR target_id IN (SELECT id FROM nodes WHERE file = @f)',
+    ),
+    nodes: db.prepare('DELETE FROM nodes WHERE file = ?'),
+    fileHashes: tryPrepare('DELETE FROM file_hashes WHERE file = ?'),
+  };
+}
+
 /**
  * Cascade-delete all graph data for a single file across all tables.
  * Order: dependent tables first, then edges, then nodes, then hashes.
- * Tables that may not exist are wrapped in try/catch.
  *
  * @param {object} db - Open read-write database handle
  * @param {string} file - Relative file path to purge
  * @param {object} [opts]
  * @param {boolean} [opts.purgeHashes=true] - Also delete file_hashes entry
  */
 export function purgeFileData(db, file, opts = {}) {
+  const stmts = preparePurgeStmts(db);
+  runPurge(stmts, file, opts);
+}
+
+/**
+ * Run purge using pre-prepared statements for a single file.
+ * @param {object} stmts - Prepared statements from preparePurgeStmts
+ * @param {string} file - Relative file path to purge
+ * @param {object} [opts]
+ * @param {boolean} [opts.purgeHashes=true]
+ */
+function runPurge(stmts, file, opts = {}) {
   const { purgeHashes = true } = opts;
 
-  // Optional tables — may not exist in older DBs
-  try {
-    db.prepare('DELETE FROM embeddings WHERE node_id IN (SELECT id FROM nodes WHERE file = ?)').run(
-      file,
-    );
-  } catch {
-    /* table may not exist */
-  }
-  try {
-    db.prepare(
-      'DELETE FROM cfg_edges WHERE function_node_id IN (SELECT id FROM nodes WHERE file = ?)',
-    ).run(file);
-  } catch {
-    /* table may not exist */
-  }
-  try {
-    db.prepare(
-      'DELETE FROM cfg_blocks WHERE function_node_id IN (SELECT id FROM nodes WHERE file = ?)',
-    ).run(file);
-  } catch {
-    /* table may not exist */
-  }
-  try {
-    db.prepare(
-      'DELETE FROM dataflow WHERE source_id IN (SELECT id FROM nodes WHERE file = ?) OR target_id IN (SELECT id FROM nodes WHERE file = ?)',
-    ).run(file, file);
-  } catch {
-    /* table may not exist */
-  }
-  try {
-    db.prepare(
-      'DELETE FROM function_complexity WHERE node_id IN (SELECT id FROM nodes WHERE file = ?)',
-    ).run(file);
-  } catch {
-    /* table may not exist */
-  }
-  try {
-    db.prepare(
-      'DELETE FROM node_metrics WHERE node_id IN (SELECT id FROM nodes WHERE file = ?)',
-    ).run(file);
-  } catch {
-    /* table may not exist */
-  }
-  try {
-    db.prepare('DELETE FROM ast_nodes WHERE file = ?').run(file);
-  } catch {
-    /* table may not exist */
-  }
+  // Optional tables
+  stmts.embeddings?.run(file);
+  stmts.cfgEdges?.run(file);
+  stmts.cfgBlocks?.run(file);
+  stmts.dataflow?.run(file, file);
+  stmts.complexity?.run(file);
+  stmts.nodeMetrics?.run(file);
+  stmts.astNodes?.run(file);
 
-  // Core tables — always exist
-  db.prepare(
-    'DELETE FROM edges WHERE source_id IN (SELECT id FROM nodes WHERE file = @f) OR target_id IN (SELECT id FROM nodes WHERE file = @f)',
-  ).run({ f: file });
-  db.prepare('DELETE FROM nodes WHERE file = ?').run(file);
+  // Core tables
+  stmts.edges.run({ f: file });
+  stmts.nodes.run(file);
 
   if (purgeHashes) {
-    try {
-      db.prepare('DELETE FROM file_hashes WHERE file = ?').run(file);
-    } catch {
-      /* table may not exist */
-    }
+    stmts.fileHashes?.run(file);
   }
 }
 
 /**
- * Purge all graph data for multiple files (transactional).
+ * Purge all graph data for multiple files.
+ * Prepares statements once and loops over files for efficiency.
  *
  * @param {object} db - Open read-write database handle
  * @param {string[]} files - Relative file paths to purge
@@ -85,7 +97,8 @@ export function purgeFileData(db, file, opts = {}) {
  */
 export function purgeFilesData(db, files, opts = {}) {
   if (!files || files.length === 0) return;
+  const stmts = preparePurgeStmts(db);
   for (const file of files) {
-    purgeFileData(db, file, opts);
+    runPurge(stmts, file, opts);
   }
 }

src/db/repository/cfg.js

@@ -5,7 +5,7 @@
  */
 export function hasCfgTables(db) {
   try {
-    db.prepare('SELECT 1 FROM cfg_blocks LIMIT 0').raw();
+    db.prepare('SELECT 1 FROM cfg_blocks LIMIT 0').get();
     return true;
   } catch {
     return false;

src/queries.js

@@ -1898,8 +1898,9 @@ function explainFunctionImpl(db, target, noTests, getFileLines) {
     if (noTests) callers = callers.filter((c) => !isTestFile(c.file));
 
     const testCallerRows = findCallers(db, node.id);
+    const seenFiles = new Set();
     const relatedTests = testCallerRows
-      .filter((r) => isTestFile(r.file))
+      .filter((r) => isTestFile(r.file) && !seenFiles.has(r.file) && seenFiles.add(r.file))
       .map((r) => ({ file: r.file }));
 
     // Complexity metrics