[FEATURE] Allow recursive indexing of additional content

Refactor the method to recursively fetch and combine content and files from additional tables. Modify `GenericRepository` to support queries filtering by parent table for recursive indexing.
tpwd · Oct 30, 2024 · 5c294ac · 5c294ac
1 parent 0a0b277
commit 5c294ac
Show file tree

Hide file tree

Showing 4 changed files with 133 additions and 29 deletions.
diff --git a/ChangeLog b/ChangeLog
@@ -3,6 +3,7 @@ ChangeLog
 Upcoming version
 [BUGFIX]  parse_ini_string(): Passing null to parameter #1 ($ini_string) of type string is deprecated. Thanks to medarob. https://github.com/tpwd/ke_search/issues/247
 [FEATURE] Make it possible to index multiple additional tables for a content element. https://github.com/tpwd/ke_search/issues/230
+[FEATURE] Allow recursive indexing of additional content
 
 Version 6.0.1, 25 October 2024
 [BUGFIX]  Add missing storage page and recursive option. Thanks to Torben Hansen. https://github.com/tpwd/ke_search/issues/251

diff --git a/Classes/Domain/Repository/GenericRepository.php b/Classes/Domain/Repository/GenericRepository.php
@@ -161,6 +161,31 @@ public function findByReferenceField(
             ->fetchAllAssociative();
     }
 
+    public function findByReferenceFieldAndParentTable(
+        string $table,
+        string $parentTable,
+        string $fieldName,
+        int $value,
+        bool $includeHiddenAndTimeRestricted = false
+    ) {
+        $queryBuilder = $this->getQueryBuilder($table, $includeHiddenAndTimeRestricted);
+        return $queryBuilder
+            ->select('*')
+            ->from($table)
+            ->where(
+                $queryBuilder->expr()->eq(
+                    'parenttable',
+                    $queryBuilder->createNamedParameter($parentTable)
+                ),
+                $queryBuilder->expr()->eq(
+                    $fieldName,
+                    $queryBuilder->createNamedParameter($value, Connection::PARAM_INT)
+                )
+            )
+            ->executeQuery()
+            ->fetchAllAssociative();
+    }
+
     public function tableExists(string $table): bool
     {
         $table = strip_tags(htmlentities($table));

diff --git a/Classes/Service/AdditionalContentService.php b/Classes/Service/AdditionalContentService.php
@@ -48,25 +48,68 @@ public function init(array $indexerConfig)
      * @return array
      */
     public function getContentAndFilesFromAdditionalTables(array $ttContentRow): array
+    {
+        $combinedAdditionalContentAndFiles = [
+            'content' => '',
+            'files' => [],
+        ];
+        foreach ($this->getConfigs($ttContentRow['CType']) as $config) {
+            $additionalContentAndFiles = $this->getContentAndFilesForSingleConfig(
+                $ttContentRow,
+                $ttContentRow['CType'],
+                $config
+            );
+            $combinedAdditionalContentAndFiles['content'] .= ' ' . $additionalContentAndFiles['content'];
+            $combinedAdditionalContentAndFiles['files'] = array_merge(
+                $combinedAdditionalContentAndFiles['files'],
+                $additionalContentAndFiles['files']
+            );
+        }
+        return [
+            'content' => trim($combinedAdditionalContentAndFiles['content']),
+            'files' => $combinedAdditionalContentAndFiles['files'],
+        ];
+    }
+
+    protected function getContentAndFilesForSingleConfig(array $row, string $cType, array $config): array
     {
         $content = ' ';
         $files = [];
-        foreach ($this->getProcessedConfigsForCType($ttContentRow['CType']) as $config) {
-            $genericRepository = GeneralUtility::makeInstance(GenericRepository::class);
-            $additionalTableContentRows = $genericRepository->findByReferenceField(
+        $genericRepository = GeneralUtility::makeInstance(GenericRepository::class);
+        if (isset($config['parentTable'])) {
+            $additionalRows = $genericRepository->findByReferenceFieldAndParentTable(
                 $config['table'],
+                $config['parentTable'],
                 $config['referenceFieldName'],
-                $ttContentRow['uid']
+                $row['uid']
             );
-            foreach ($additionalTableContentRows as $additionalTableContentRow) {
-                foreach ($config['fields'] as $field) {
-                    $content .= ' ' . ContentUtility::getPlainContentFromContentRow(
-                            $additionalTableContentRow,
-                            $field,
-                            $GLOBALS['TCA'][$config['table']]['columns'][$field]['config']['type'] ?? ''
-                        );
-                    $files = array_merge($files, $this->findLinkedFiles($additionalTableContentRow, $field));
-                }
+        } else {
+            $additionalRows = $genericRepository->findByReferenceField(
+                $config['table'],
+                $config['referenceFieldName'],
+                $row['uid']
+            );
+        }
+        foreach ($additionalRows as $additionalRow) {
+            foreach ($config['fields'] as $field) {
+                $content .= ' ' . ContentUtility::getPlainContentFromContentRow(
+                    $additionalRow,
+                    $field,
+                    $GLOBALS['TCA'][$config['table']]['columns'][$field]['config']['type'] ?? ''
+                );
+                $files = array_merge($files, $this->findLinkedFiles($additionalRow, $field));
+            }
+
+            // Get content from tables which have the current table as parent recursively
+            $subConfigs = $this->getConfigs($cType, $config['table']);
+            foreach ($subConfigs as $subConfig) {
+                $additionalContentAndFiles = $this->getContentAndFilesForSingleConfig(
+                    $additionalRow,
+                    $cType,
+                    $subConfig
+                );
+                $content .= ' ' . $additionalContentAndFiles['content'];
+                $files = array_merge($files, $additionalContentAndFiles['files']);
             }
         }
         return ['content' => trim($content), 'files' => $files];
@@ -108,7 +151,10 @@ protected function parseAndProcessAdditionalTablesConfiguration(): array
         $cTypes = $this->findAllCTypesInConfiguration($tempAdditionalTableConfig);
         $additionalTableConfig = [];
         foreach ($cTypes as $cType) {
-            $additionalTableConfig[$cType] = $this->getUnprocessedConfigsForCType($tempAdditionalTableConfig, $cType);
+            $additionalTableConfig[$cType] = $this->getCombinedConfigsForCTypeFromRawConfig(
+                $tempAdditionalTableConfig,
+                $cType
+            );
         }
         return $additionalTableConfig;
     }
@@ -167,7 +213,7 @@ public function findLinkedFiles(array $contentRow, string $field = 'bodytext'):
      * @param array $additionalTableConfig The additional table configuration to search for cTypes.
      * @return array An array of unique cTypes found in the configuration.
      */
-    protected function findAllCTypesInConfiguration($additionalTableConfig): array
+    protected function findAllCTypesInConfiguration(array $additionalTableConfig): array
     {
         $cTypes = [];
         foreach ($additionalTableConfig as $cType => $config) {
@@ -180,15 +226,15 @@ protected function findAllCTypesInConfiguration($additionalTableConfig): array
     }
 
     /**
-     * Retrieves an array of unprocessed configurations for a given content type. Unprocessed means
-     * the CTypes names my have indexes in it ("my_ctype.1", "my_ctype.2") and each configuration
-     * configures only one table.
+     * Retrieves an array of configurations for a given content type. Expects the additional table config in ini
+     * format which may have multiple configurations for the same CType with indexes in it
+     *  ("my_ctype.1", "my_ctype.2") and combines them into one array.
      *
      * @param array $additionalTableConfig Array containing configurations for various content types.
-     * @param string $cType The content type for which unprocessed configurations are to be fetched.
-     * @return array Array containing unprocessed configurations specific to the given content type.
+     * @param string $cType The content type for which configurations are to be fetched.
+     * @return array Array containing combined configurations specific to the given content type.
      */
-    protected function getUnprocessedConfigsForCType(array $additionalTableConfig, string $cType): array
+    protected function getCombinedConfigsForCTypeFromRawConfig(array $additionalTableConfig, string $cType): array
     {
         $configs = [];
         foreach ($additionalTableConfig as $currentCType => $currentConfig) {
@@ -202,16 +248,20 @@ protected function getUnprocessedConfigsForCType(array $additionalTableConfig, s
         return $configs;
     }
 
-
     /**
-     * Retrieves processed configurations for a given content type. The result is a multidimensional array, in the first
-     * level there's one element per table.
+     * Retrieves processed configurations for a given content type and, if given, the parent table.
      *
      * @param string $cType The content type identifier.
      * @return array The processed configurations associated with the specified content type.
      */
-    protected function getProcessedConfigsForCType(string $cType): array
+    protected function getConfigs(string $cType, string $parentTable = ''): array
     {
-        return $this->processedAdditionalTableConfig[$cType] ?? $this->processedAdditionalTableConfig[$cType] ?? [];
+        $configs = $this->processedAdditionalTableConfig[$cType] ?? $this->processedAdditionalTableConfig[$cType] ?? [];
+        foreach ($configs as $key => $config) {
+            if ($parentTable != '' && $config['parentTable'] !== $parentTable) {
+                unset($configs[$key]);
+            }
+        }
+        return $configs;
     }
 }
diff --git a/Documentation/Indexing/IndexerTypes/Pages.rst b/Documentation/Indexing/IndexerTypes/Pages.rst
@@ -127,10 +127,19 @@ table
     This is the table that holds the content.
 
 referenceFieldName
-    This ist the field that holds the relation to the tt_content record (the
+    This is the field that holds the relation to the tt_content record (the
     UID of the record). In EXT:bootstrap_package it is named `tt_content`,
     in EXT:mask it is named `parentid`.
 
+parentTable
+    The parent table is an optional setting. It's only necessary if you want to index
+    sub-elements of EXT:mask. For example If you have repeating elements in a mask
+    element which themselves have repeating elements. You can define the parent table
+    for the sub-elements here (see example below). Indexing will be done recursively.
+    If set the database query will contain a "WHERE parenttable = ..." condition. This
+    column exists in content elements from EXT:mask but not in content elements
+    from EXT:bootstrap_package.
+
 fields[]
     A list of database fields which should be indexed. If the field is
     configured as type "file" in the TCA the indexer will check if it links
@@ -165,15 +174,34 @@ index mask elements (remember to also add
     referenceFieldName = parentid
     fields[] = tx_mask_content_item
 
+This is an example how to add multiple additional tables for the same CType.
+
+.. code-block:: ini
+
+    [mask_mytest]
+    table = tx_mask_repeating1
+    referenceFieldName = parentid
+    fields[] = tx_mask_name
+
+    [mask_mytest.1]
+    table = tx_mask_repeating2
+    referenceFieldName = parentid
+    fields[] = tx_mask_title
+
+This is an example how to index sub-elements of additional tables (note the `parentTable` configuration line).
+
+.. code-block:: ini
+
     [mask_mytest]
     table = tx_mask_repeating1
     referenceFieldName = parentid
-    fields[] = tx_mask_string1
+    fields[] = tx_mask_name
 
     [mask_mytest.1]
     table = tx_mask_repeating2
+    parentTable = tx_mask_repeating1
     referenceFieldName = parentid
-    fields[] = tx_mask_string2
+    fields[] = tx_mask_title
 
 This is an example for a some mask elements: