🐛 More fixes

Author: willscottuk

app/Integrations/Untappd/UntappdPlugin.php

@@ -96,9 +96,9 @@ public static function getDomain(): string
     public static function getActionTypes(): array
     {
         return [
-            'drank_beer' => [
+            'drank' => [
                 'icon' => 'fas.beer-mug-empty',
-                'display_name' => 'Drank Beer',
+                'display_name' => 'Drank',
                 'description' => 'Drank a beer',
                 'display_with_object' => true,
                 'value_unit' => null,

app/Jobs/Data/Goodreads/GoodreadsRssData.php

@@ -3,6 +3,7 @@
 namespace App\Jobs\Data\Goodreads;
 
 use App\Jobs\Base\BaseProcessingJob;
+use App\Models\EventObject;
 use Carbon\Carbon;
 use DOMDocument;
 use DOMXPath;
@@ -62,16 +63,20 @@ protected function process(): void
 
             // Build target (book)
             $bookTitle = $descriptionData['bookTitle'] ?? $parsedData['bookTitle'] ?? 'Unknown Book';
+            $bookUrl = $descriptionData['bookUrl'] ?? $link;
+            $bookId = $this->extractBookId($bookUrl);
+
             $target = [
                 'concept' => 'document',
                 'type' => 'goodreads_book',
                 'title' => $bookTitle,
                 'content' => null,
                 'metadata' => [
+                    'book_id' => $bookId,
                     'author' => $descriptionData['authorName'] ?? null,
                     'author_url' => $descriptionData['authorUrl'] ?? null,
                 ],
-                'url' => $descriptionData['bookUrl'] ?? $link,
+                'url' => $bookUrl,
                 'image_url' => $descriptionData['coverUrl'] ?? null,
                 'time' => $pubDate ? Carbon::parse($pubDate) : now(),
             ];
@@ -146,6 +151,47 @@ protected function process(): void
         }
     }
 
+    /**
+     * Override createOrUpdateObject to handle book deduplication by book_id
+     */
+    protected function createOrUpdateObject(array $objectData): EventObject
+    {
+        // For Goodreads books, check if object with same book_id already exists
+        if ($objectData['type'] === 'goodreads_book' && isset($objectData['metadata']['book_id'])) {
+            $bookId = $objectData['metadata']['book_id'];
+
+            // Find existing book by book_id in metadata
+            $existingBook = EventObject::where('user_id', $this->integration->user_id)
+                ->where('concept', $objectData['concept'])
+                ->where('type', $objectData['type'])
+                ->whereJsonContains('metadata->book_id', $bookId)
+                ->first();
+
+            if ($existingBook) {
+                // Keep the longer title (handles truncated titles from RSS feed)
+                $newTitle = $objectData['title'];
+                $existingTitle = $existingBook->title;
+                $titleToKeep = mb_strlen($newTitle) > mb_strlen($existingTitle) ? $newTitle : $existingTitle;
+
+                // Update the existing book with new data
+                $existingBook->update([
+                    'time' => $objectData['time'] ?? now(),
+                    'title' => $titleToKeep,
+                    'content' => $objectData['content'] ?? null,
+                    'metadata' => array_merge($existingBook->metadata ?? [], $objectData['metadata'] ?? []),
+                    'url' => $objectData['url'] ?? $existingBook->url,
+                    'media_url' => $objectData['image_url'] ?? $existingBook->media_url,
+                    'embeddings' => $objectData['embeddings'] ?? $existingBook->embeddings,
+                ]);
+
+                return $existingBook;
+            }
+        }
+
+        // Fall back to parent method for other object types
+        return parent::createOrUpdateObject($objectData);
+    }
+
     /**
      * Parse the RSS item title to extract action type and book information
      */
@@ -316,4 +362,22 @@ private function getFullSizeCoverUrl(?string $url): ?string
         // Remove size suffixes like _SX98_, _SY475_, etc.
         return preg_replace('/\._[A-Z]{2}\d+_\./', '.', $url);
     }
+
+    /**
+     * Extract book ID from Goodreads book URL
+     * Example: https://www.goodreads.com/book/show/25792894-kings-rising -> 25792894
+     */
+    private function extractBookId(?string $url): ?string
+    {
+        if (! $url) {
+            return null;
+        }
+
+        // Match pattern: /book/show/{book_id}-{slug} or /book/show/{book_id}
+        if (preg_match('/\/book\/show\/(\d+)(?:-|$)/', $url, $matches)) {
+            return $matches[1];
+        }
+
+        return null;
+    }
 }

app/Jobs/Data/Untappd/UntappdRssData.php

@@ -126,7 +126,7 @@ protected function process(): void
                 'source_id' => $sourceId,
                 'time' => $pubDate ? Carbon::parse($pubDate) : now(),
                 'domain' => 'health',
-                'action' => 'drank_beer',
+                'action' => 'drank',
                 'value' => null,
                 'value_multiplier' => 1,
                 'value_unit' => null,

app/Jobs/OAuth/Untappd/UntappdRssPull.php

@@ -36,11 +36,28 @@ protected function fetchData(): array
             throw new Exception('RSS fetch failed: ' . $result['error']);
         }
 
-        // Parse XML
-        $xml = simplexml_load_string($result['html']);
+        $xmlContent = $result['html'];
+
+        // Extract XML from <pre> tags if wrapped in HTML
+        if (str_starts_with(trim($xmlContent), '<!DOCTYPE') || str_starts_with(trim($xmlContent), '<html')) {
+            // Try to extract content from <pre> tags
+            if (preg_match('/<pre[^>]*>(.*?)<\/pre>/is', $xmlContent, $matches)) {
+                $xmlContent = html_entity_decode($matches[1], ENT_QUOTES | ENT_HTML5);
+            } else {
+                throw new Exception('Received HTML instead of XML and could not extract from <pre> tags. Response starts with: ' . substr($xmlContent, 0, 200));
+            }
+        }
+
+        // Parse XML with error handling
+        libxml_use_internal_errors(true);
+        $xml = simplexml_load_string($xmlContent);
 
         if ($xml === false) {
-            throw new Exception('Failed to parse RSS XML');
+            $errors = libxml_get_errors();
+            $errorMessages = array_map(fn ($error) => trim($error->message), $errors);
+            libxml_clear_errors();
+
+            throw new Exception('Failed to parse RSS XML: ' . implode(', ', $errorMessages) . '. Content preview: ' . substr($xmlContent, 0, 500));
         }
 
         $items = [];

tests/Unit/Integrations/UntappdRssDataTest.php

@@ -57,12 +57,12 @@ public function it_parses_beer_checkin_with_brewery()
 
         $this->assertDatabaseHas('events', [
             'service' => 'untappd',
-            'action' => 'drank_beer',
+            'action' => 'drank',
             'domain' => 'health',
         ]);
 
         $event = Event::where('service', 'untappd')->first();
-        $this->assertEquals('drank_beer', $event->action);
+        $this->assertEquals('drank', $event->action);
         $this->assertNull($event->value);
 
         // Check actor
@@ -217,8 +217,8 @@ public function it_handles_multiple_checkins_in_one_batch()
         $events = Event::where('service', 'untappd')->get();
         $this->assertCount(3, $events);
 
-        // All should be drank_beer actions
-        $this->assertTrue($events->every(fn ($e) => $e->action === 'drank_beer'));
+        // All should be drank actions
+        $this->assertTrue($events->every(fn ($e) => $e->action === 'drank'));
 
         // Check that one has a venue
         $eventWithVenue = $events->first(fn ($e) => ! empty($e->target->metadata['venue']));