[FEATURE] Provide structure paths/breadcrumbs to search results

Classes/Common/Indexer.php

@@ -80,6 +80,13 @@ class Indexer
      */
     protected static array $processedDocs = [];
 
+    /**
+     * @access protected
+     * @static
+     * @var array List of already extracted structure nodes for structure path
+     */
+    protected static array $extractedStructurePathNodes = [];
+
     /**
      * @access protected
      * @static
@@ -371,6 +378,10 @@ protected static function processLogical(Document $document, array $logicalUnit)
                 $solrDoc->setField('toplevel', $logicalUnit['id'] == $doc->getToplevelId());
                 $solrDoc->setField('title', $metadata['title'][0]);
                 $solrDoc->setField('volume', $metadata['volume'][0] ?? '');
+                // extract structure path
+                self::$extractedStructurePathNodes[$logicalUnit['id']] = self::extractStructurePathNodes($doc->tableOfContents, $logicalUnit['id']);
+                $processedStructurePath = self::buildStructurePathData(self::$extractedStructurePathNodes[$logicalUnit['id']], $document->getCurrentDocument()->getToplevelId());
+                $solrDoc->setField('structure_path', json_encode($processedStructurePath, JSON_UNESCAPED_UNICODE));
                 // verify date formatting
                 if (strtotime($metadata['date'][0])) {
                     $solrDoc->setField('date', self::getFormattedDate($metadata['date'][0]));
@@ -465,7 +476,21 @@ protected static function processPhysical(Document $document, int $page, array $
             $solrDoc->setField('type', $physicalUnit['type']);
             $solrDoc->setField('collection', $doc->metadataArray[$doc->getToplevelId()]['collection']);
             $solrDoc->setField('location', $document->getLocation());
-
+            // pick only the deepest structure paths
+            $associatedPaths = [];
+            foreach ($doc->smLinks['p2l'][$physicalUnit['id']] as $logicalId) {
+                $path = self::$extractedStructurePathNodes[$logicalId] ?? [];
+                if (!empty($path)) {
+                    $associatedPaths[$logicalId] = $path;
+                }
+            }
+            $deepestPaths = self::filterDeepestStructurePaths($associatedPaths);
+            $processedStructurePath = [];
+            foreach ($deepestPaths as $path) {
+                $segments = self::buildStructurePathData($path, $document->getCurrentDocument()->getToplevelId());
+                $processedStructurePath[] = json_encode($segments, JSON_UNESCAPED_UNICODE);
+            }
+            $solrDoc->setField('structure_path', $processedStructurePath);
             $solrDoc->setField('fulltext', $fullText);
             if (is_array($doc->metadataArray[$doc->getToplevelId()])) {
                 self::addFaceting($doc, $solrDoc, $physicalUnit);
@@ -728,6 +753,147 @@ private static function removeAppendsFromAuthor(array|string $authors): array|st
         return $authors;
     }
 
+    /**
+     * Extract nodes alongside the structure map in direct line to the target id and return them as flattened array.
+     *
+     * @access private
+     *
+     * @static
+     *
+     * @param array $nodes Tree or Sub-Tree, where the target id should be extracted from if present
+     * @param string $targetId The ID of the logical structure element to be found
+     * @param array $path An intermediate array that keeps track of the current branch that is being looked up
+     *
+     * @return array
+     */
+    private static function extractStructurePathNodes(array $nodes, string $targetId, array $path = []): array
+    {
+        foreach ($nodes as $node) {
+            // remember where we came from
+            $currentPath = array_merge($path, [$node]);
+            if ($node['id'] == $targetId) {
+                return $currentPath;
+            }
+            if (!empty($node['children'])) {
+                $result = self::extractStructurePathNodes($node['children'], $targetId, $currentPath);
+                if ($result) {
+                    return $result;
+                }
+            }
+        }
+        return [];
+    }
+
+    /**
+     * Filters those structure path nodes that are the descending into the structure tree the most and removes any that resemble a "prefix" of another.
+     *
+     * @access private
+     *
+     * @static
+     *
+     * @param array $paths The array containing all structure path nodes associated with a physical page
+     *
+     * @return array
+     */
+    private static function filterDeepestStructurePaths(array $paths): array
+    {
+        if (count($paths) <= 1) {
+            return $paths;
+        }
+
+        $deepestPath = [];
+        foreach ($paths as $currentLogicalId => $currentPath) {
+            $currentIds = array_column($currentPath, 'id');
+            $isPrefix = false;
+
+            foreach ($paths as $comparisonLogicalId => $comparisonPath) {
+                if ($currentLogicalId === $comparisonLogicalId) {
+                    continue;
+                }
+                $comparisonIds = array_column($comparisonPath, 'id');
+                // check if structure path is part/prefix of another structure path
+                if (
+                    count($currentIds) < count($comparisonIds)
+                    && array_slice($comparisonIds, 0, count($currentIds)) === $currentIds
+                ) {
+                    $isPrefix = true;
+                    break;
+                }
+            }
+
+            if (!$isPrefix) {
+                $deepestPath[$currentLogicalId] = $currentPath;
+            }
+        }
+        return $deepestPath;
+    }
+
+    /**
+     * Create the actual array with the required data for the structure path that will be JSON encoded and indexed.
+     *
+     * @access private
+     *
+     * @static
+     *
+     * @param array $path The structure path nodes that shall be processed
+     * @param string $cutoffId The logical id at which ancestors and itself will not be part of the structure path data
+     *
+     * @return array
+     */
+    private static function buildStructurePathData(array $path, string $cutoffId): array
+    {
+        $cutoffIndex = array_search($cutoffId, array_column($path, 'id'));
+        if ($cutoffIndex !== false) {
+            $path = array_slice($path, $cutoffIndex + 1);
+        }
+
+        $segments = [];
+        foreach ($path as $node) {
+            $segments[] = self::buildStructurePathSegments($node);
+        }
+        return $segments;
+    }
+
+    /**
+     * Gets the label or type of a structure path node with corresponding tag
+     *
+     * @access private
+     *
+     * @static
+     *
+     * @param array $node The current node that should be processed
+     *
+     * @return array
+     */
+    private static function buildStructurePathSegments(array $node): array
+    {
+        if (!empty($node['label'])) {
+            return [
+                'label' => $node['label'],
+            ];
+        }
+        if (!empty($node['orderlabel'])) {
+            return [
+                'label' => $node['orderlabel'],
+            ];
+        }
+        if (!empty($node['volume'])) {
+            $value = !empty($node['year'])
+                ? $node['volume'] . ' ' . $node['year']
+                : $node['volume'];
+
+            return [
+                'label' => $value,
+            ];
+        }
+        if (!empty($node['type'])) {
+            return [
+                'type' => $node['type'],
+            ];
+        }
+        return ['label' => ''];
+    }
+
     /**
      * Handle exception.
      *

Classes/Common/Solr/SearchResult/ResultDocument.php

@@ -73,6 +73,12 @@ class ResultDocument
      */
     private ?string $type;
 
+    /**
+     * @access private
+     * @var array The JSON encoded structure path(s)
+     */
+    private array $structurePath = [];
+
     /**
      * @access private
      * @var Page[] All pages in which search phrase was found
@@ -117,6 +123,7 @@ public function __construct(Document $record, array $highlighting, array $fields
         $this->title = $record[$fields['title']];
         $this->toplevel = $record[$fields['toplevel']] ?? false;
         $this->type = $record[$fields['type']];
+        $this->structurePath = $record[$fields['structure_path']] ?? [];
 
         if (!empty($highlighting[$this->id])) {
             $highlightingForRecord = $highlighting[$this->id][$fields['fulltext']];
@@ -225,6 +232,18 @@ public function getType(): ?string
         return $this->type;
     }
 
+    /**
+     * Get the structure path(s)
+     *
+     * @access public
+     *
+     * @return array
+     */
+    public function getStructurePath(): array
+    {
+        return $this->structurePath;
+    }
+
     /**
      * Get all result's pages which contain search phrase.
      *

Classes/Common/Solr/Solr.php

@@ -256,6 +256,7 @@ public static function getFields(): array
             self::$fields['type'] = $solrFields['type'];
             self::$fields['title'] = $solrFields['title'];
             self::$fields['volume'] = $solrFields['volume'];
+            self::$fields['structure_path'] = $solrFields['structurePath'];
             self::$fields['date'] = $solrFields['date'] ?? null;
             self::$fields['thumbnail'] = $solrFields['thumbnail'];
             self::$fields['default'] = $solrFields['default'];

Classes/Common/Solr/SolrSearch.php

@@ -487,7 +487,7 @@ public function prepare()
         $params['listMetadataRecords'] = [];
 
         // Restrict the fields to the required ones.
-        $params['fields'] = 'uid,id,page,title,thumbnail,partof,toplevel,type';
+        $params['fields'] = 'uid,id,page,title,thumbnail,partof,toplevel,type,structure_path';
 
         if ($this->listedMetadata) {
             foreach ($this->listedMetadata as $metadata) {
@@ -560,6 +560,31 @@ public function submit(int $start, int $rows, bool $processResults = true): void
                             $searchResult['page'] = $doc['page'];
                             $searchResult['thumbnail'] = $doc['thumbnail'];
                             $searchResult['structure'] = $doc['type'];
+                            // create string(s) from structure path(s)
+                            $encodedStructurePaths = $doc['structure_path'] ?? [];
+                            if (!is_array($encodedStructurePaths)) {
+                                $encodedStructurePaths = [$encodedStructurePaths];
+                            }
+                            $structurePathStrings = [];
+                            foreach ($encodedStructurePaths as $jsonString) {
+                                if (!is_string($jsonString) || $jsonString === '') {
+                                    continue;
+                                }
+                                $segments = json_decode($jsonString, true);
+                                if ($segments === null && json_last_error() !== JSON_ERROR_NONE) {
+                                    continue;
+                                }
+                                $structurePathLabels = [];
+                                foreach ($segments as $currentSegment) {
+                                    if (isset($currentSegment['type'])) {
+                                        $structurePathLabels[] = Helper::translate($currentSegment['type'], 'tx_dlf_structures', $this->settings['storagePid']);
+                                    } elseif (!empty($currentSegment['label'])) {
+                                        $structurePathLabels[] = $currentSegment['label'];
+                                    }
+                                }
+                                $structurePathStrings[] = implode(' → ', $structurePathLabels);
+                            }
+                            $searchResult['structure_path'] = $structurePathStrings;
                             $searchResult['title'] = $doc['title'];
                             foreach ($params['listMetadataRecords'] as $indexName => $solrField) {
                                 if (isset($doc['metadata'][$indexName])) {
@@ -901,6 +926,7 @@ private function getDocument(Document $record, array $highlighting, array $field
             'title' => $resultDocument->getTitle(),
             'toplevel' => $resultDocument->getToplevel(),
             'type' => $resultDocument->getType(),
+            'structure_path' => $resultDocument->getStructurePath(),
             'uid' => !empty($resultDocument->getUid()) ? $resultDocument->getUid() : $parameters['uid'],
             'highlight' => $resultDocument->getHighlightsIds(),
         ];

Configuration/ApacheSolr/configsets/dlf/conf/schema.xml

@@ -144,6 +144,8 @@ limitations under the License.
         <!-- Next two fields are mandatory for identifying documents. -->
         <field name="title" type="standard" indexed="true" stored="true" multiValued="false" default="" />
         <field name="volume" type="standard" indexed="true" stored="true" multiValued="false" default="" />
+        <!-- Convenience field to provide context about the path within the logical structure map -->
+        <field name="structure_path" type="string" indexed="false" stored="true" multiValued="true" default="" />
         <!-- The keydate of a resource e.g a newspaper was issued or a letter was written -->
 	    <field name="date" type="daterange" indexed="true" stored="true" multiValued="false" />
         <!-- URL of thumbnail image for the document. -->

Configuration/FlexForms/ListView.xml

@@ -65,6 +65,14 @@
                             <default>0</default>
                         </config>
                     </settings.getTitle>
+                    <settings.getStructurePath>
+                        <exclude>1</exclude>
+                        <label>LLL:EXT:dlf/Resources/Private/Language/locallang_be.xlf:flexform.getStructurePath</label>
+                        <config>
+                            <type>check</type>
+                            <default>0</default>
+                        </config>
+                    </settings.getStructurePath>
                     <settings.basketButton>
                         <onChange>reload</onChange>
                         <exclude>1</exclude>

Resources/Private/Language/de.locallang_be.xlf

@@ -17,6 +17,10 @@
                 <source><![CDATA[Show only documents from the selected collection(s)]]></source>
                 <target><![CDATA[Nur Dokumente der ausgewählten Kollektion(en) berücksichtigen]]></target>
             </trans-unit>
+            <trans-unit id="flexform.getStructurePath" approved="yes">
+                <source><![CDATA[Show breadcrumb/path to result location within the structure map]]></source>
+                <target><![CDATA[Breadcrumb/Pfad des Treffers innerhalb des Strukturbaums anzeigen]]></target>
+            </trans-unit>
             <trans-unit id="flexform.getTitle" approved="yes">
                 <source><![CDATA[Show title of parent document if document has no title itself]]></source>
                 <target><![CDATA[Bei Bedarf Titel des übergeordneten Dokuments anzeigen]]></target>

Resources/Private/Language/de.locallang_labels.xlf

@@ -821,6 +821,10 @@
                 <target>Solr-Schema-Feld "volume" : Volume field is mandatory for identifying documents (Standard ist "volume")</target>
                 <source>Solr Schema Field "volume" : Volume field is mandatory for identifying documents (default is "volume")</source>
             </trans-unit>
+            <trans-unit id="config.solr.fields.structurePath">
+                <target>Solr-Schema-Feld "structure_path" : Field providing context about the location of a resource in the structure map (Standard ist "structure_path")</target>
+                <source>Solr Schema Field "structure_path" : Field providing context about the location of a resource in the structure map (default is "structure_path")</source>
+            </trans-unit>
             <trans-unit id="config.solr.fields.date">
                 <target>Solr Schema Field "date" : The date a resource was issued or created. Used for datesearch (Standard ist "date")</target>
                 <source>Solr Schema Field "date" : The date a resource was issued or created. Used for datesearch (default is "date")</source>

Resources/Private/Language/de.locallang_metadata.xlf

@@ -93,6 +93,10 @@
                 <source><![CDATA[Shelfmark]]></source>
                 <target><![CDATA[Signatur]]></target>
             </trans-unit>
+            <trans-unit id="metadata.structure_path" approved="yes">
+                <source><![CDATA[Structure Path]]></source>
+                <target><![CDATA[Strukturpfad]]></target>
+            </trans-unit>
             <trans-unit id="metadata.terms" approved="yes">
                 <source><![CDATA[Terms of Use]]></source>
                 <target><![CDATA[Nutzungsbedingungen]]></target>

Resources/Private/Language/locallang_be.xlf

@@ -14,6 +14,12 @@
             <trans-unit id="flexform.excludeOtherCollections">
                 <source><![CDATA[Show only documents from the selected collection(s)]]></source>
             </trans-unit>
+            <trans-unit id="flexform.getStructurePath">
+                <source><![CDATA[Show breadcrumb/path to result location within the structure map]]></source>
+            </trans-unit>
+            <trans-unit id="flexform.getTitle">
+                <source><![CDATA[Show title of parent document if document has no title itself]]></source>
+            </trans-unit>
             <trans-unit id="flexform.library">
                 <source><![CDATA[Providing library]]></source>
             </trans-unit>
@@ -356,9 +362,6 @@
             <trans-unit id="plugins.listview.flexform.limit">
                 <source><![CDATA[Documents per page]]></source>
             </trans-unit>
-            <trans-unit id="flexform.getTitle">
-                <source><![CDATA[Show title of parent document if document has no title itself]]></source>
-            </trans-unit>
             <trans-unit id="plugins.collection.title">
                 <source><![CDATA[Kitodo: Collection]]></source>
             </trans-unit>

Resources/Private/Language/locallang_labels.xlf

@@ -617,6 +617,9 @@
             <trans-unit id="config.solr.fields.volume">
                 <source>Solr Schema Field "volume" : Volume field is mandatory for identifying documents (default is "volume")</source>
             </trans-unit>
+            <trans-unit id="config.solr.fields.structurePath">
+                <source>Solr Schema Field "structure_path" : Field providing context about the location of a resource in the structure map (default is "structure_path")</source>
+            </trans-unit>
             <trans-unit id="config.solr.fields.date">
                 <source>Solr Schema Field "date" : The date a resource was issued or created. Used for datesearch (default is "date")</source>
             </trans-unit>