Batch category name/path resolution

CHANGELOG.md

@@ -1,3 +1,9 @@
+# Version 18.3.0
+
+## Features
+
+* Optimize category path resolution with batched entity ID processing and caching
+
 # Version 18.2.0
 
 ## Features

src/Assembler/CategoryAssembler.php

@@ -94,7 +94,7 @@ public function getCategoriesWithResolvedPath()
                 continue;
             }
 
-            // cut-off the root category
+            // cut off the root category
             array_shift($entityIds);
 
             // continue with the next category if no entity IDs are available
@@ -111,7 +111,7 @@ public function getCategoriesWithResolvedPath()
                 }
             }
 
-            // append the catogory with the string path as key
+            // append the category with the string path as key
             $categories[$this->serializer->implode($path)] = $category;
         }
 
@@ -131,40 +131,59 @@ public function getCategoriesWithResolvedPath()
     public function getCategoriesWithResolvedPathByStoreView($storeViewId)
     {
         // prepare the categories
-        $categories = array();
+        $categories = [];
 
         // load the categories from the database
         $availableCategories = $this->categoryRepository->findAllByStoreView($storeViewId);
 
-        // create the array with the resolved category path as keys
+        // step 1: collect the path entity IDs
+        $prepared = [];
+        $allEntityIds = [];
         foreach ($availableCategories as $category) {
-            // expload the entity IDs from the category path
+            // explode the entity IDs from the category path
             $entityIds = $this->serializer->explode($category[MemberNames::PATH]);
 
             // continue if nothing to explode
             if ($entityIds === null) {
                 continue;
             }
 
-            // cut-off the root category
+            // cut off the root category
             array_shift($entityIds);
 
             // continue with the next category if no entity IDs are available
             if (sizeof($entityIds) === 0 || !isset($category[MemberNames::NAME])) {
                 continue;
             }
 
-            // initialize the array for the path elements
-            $path = array();
+            $prepared[] = ['category' => $category, 'entityIds' => $entityIds];
             foreach ($entityIds as $entityId) {
-                $cat = $this->categoryVarcharRepository->findByEntityId($entityId);
-                if ($cat && isset($cat[MemberNames::VALUE])) {
-                    $path[] = $cat[MemberNames::VALUE];
+                $allEntityIds[(int)$entityId] = true;
+            }
+        }
+
+        // step 2: load all category names
+        $nameByEntityId = [];
+        if ($allEntityIds !== []) {
+            foreach ($this->categoryVarcharRepository->findAllByEntityIds(array_keys($allEntityIds)) as $row) {
+                if (isset($row[MemberNames::ENTITY_ID]) && isset($row[MemberNames::VALUE])) {
+                    $nameByEntityId[(int)$row[MemberNames::ENTITY_ID]] = $row[MemberNames::VALUE];
                 }
             }
+        }
 
-            // append the catogory with the string path as key
-            $categories[$this->serializer->implode($path)] = $category;
+        // step 3: resolve each category's path from index
+        foreach ($prepared as $item) {
+            // initialize the array for the path elements
+            $path = [];
+            foreach ($item['entityIds'] as $entityId) {
+                if (isset($nameByEntityId[(int)$entityId])) {
+                    $path[] = $nameByEntityId[(int)$entityId];
+                }
+            }
+
+            // append the category with the string path as key
+            $categories[$this->serializer->implode($path)] = $item['category'];
         }
 
         // return array with the categories

src/Repositories/CategoryVarcharRepository.php

@@ -28,6 +28,19 @@
  */
 class CategoryVarcharRepository extends AbstractRepository implements CategoryVarcharRepositoryInterface
 {
+    /**
+     * The default number of entity IDs to load in a single batch query.
+     *
+     * @var integer
+     */
+    private const int BATCH_SIZE = 1000;
+
+    /**
+     * Cache for the category varchar values loaded by entity ID.
+     *
+     * @var array
+     */
+    private array $cacheByEntityId = [];
 
     /**
      * Initializes the repository's prepared statements.
@@ -42,21 +55,38 @@ public function init()
      * Returns the category varchar values for the categories with
      * the passed with the passed entity IDs.
      *
-     * @param array $entityIds The array with the category IDs
+     * @param array   $entityIds The array with the category IDs
+     * @param integer $batchSize The maximum number of entity IDs per query
      *
-     * @return mixed The category varchar values
+     * @return array The category varchar values
      */
-    public function findAllByEntityIds(array $entityIds)
+    public function findAllByEntityIds(array $entityIds, int $batchSize = self::BATCH_SIZE): array
     {
+        $result = [];
+        $entityIds = array_values(array_unique(array_map('intval', $entityIds)));
 
-        // prepare the cache key
-        $vals = implode(',', $entityIds);
-        $sql = str_replace('?', $vals, $this->loadStatement(SqlStatementKeys::CATEGORY_VARCHARS_BY_ENTITY_IDS));
+        if ($entityIds === []) {
+            return $result;
+        }
 
-        // load the categories with the passed values and return them
-        if ($stmt = $this->getConnection()->query($sql)) {
-            return $stmt->fetchAll();
+        $batchSize = (int)$batchSize;
+        if ($batchSize < 1) {
+            $batchSize = self::BATCH_SIZE;
+        }
+
+        foreach (array_chunk($entityIds, $batchSize) as $chunk) {
+            $vals = implode(',', $chunk);
+            $sql = str_replace('?', $vals, $this->loadStatement(SqlStatementKeys::CATEGORY_VARCHARS_BY_ENTITY_IDS));
+
+            if ($stmt = $this->getConnection()->query($sql)) {
+                foreach ($stmt->fetchAll() as $row) {
+                    $result[] = $row;
+                }
+            }
         }
+
+        // return the collected result rows
+        return $result;
     }
 
     /**
@@ -69,14 +99,22 @@ public function findAllByEntityIds(array $entityIds)
      */
     public function findByEntityId($entityId)
     {
+        $entityId = (int)$entityId;
+
+        if (array_key_exists($entityId, $this->cacheByEntityId)) {
+            return $this->cacheByEntityId[$entityId];
+        }
 
-        // prepare the cache key
-        $sql = str_replace('?', $entityId, $this->loadStatement(SqlStatementKeys::CATEGORY_VARCHARS_BY_ENTITY_IDS));
+        $sql = str_replace(
+            '?',
+            (string)$entityId,
+            $this->loadStatement(SqlStatementKeys::CATEGORY_VARCHARS_BY_ENTITY_IDS)
+        );
 
-        // load the categories with the passed values and return them
         if ($stmt = $this->getConnection()->query($sql)) {
-            return $stmt->fetch();
+            return $this->cacheByEntityId[$entityId] = $stmt->fetch();
         }
-        return [];
+
+        return $this->cacheByEntityId[$entityId] = [];
     }
 }

src/Repositories/CategoryVarcharRepositoryInterface.php

@@ -25,16 +25,16 @@
  */
 interface CategoryVarcharRepositoryInterface extends RepositoryInterface
 {
-
     /**
      * Returns the category varchar values for the categories with
      * the passed with the passed entity IDs.
      *
      * @param array $entityIds The array with the category IDs
+     * @param integer $batchSize The maximum number of entity IDs per query
      *
-     * @return mixed The category varchar values
+     * @return array The category varchar values
      */
-    public function findAllByEntityIds(array $entityIds);
+    public function findAllByEntityIds(array $entityIds, int $batchSize = 1000): array;
 
     /**
      * Returns the category varchar values for the categories with