p23

RETRIEX_PATCH_23_COMMERCE_CLEANUP_PROFILE_WIRING_README.md

@@ -0,0 +1,26 @@
+# RetrieX Patch 23 - Commerce Cleanup Profile Wiring
+
+## Ziel
+
+Commerce nutzt erstmals das zentrale Language-Cleanup-Profil `commerce_query`.
+
+## Änderungen
+
+- `config/retriex/commerce.yaml` erhält `cleanup_profile: commerce_query`.
+- `CommerceQueryParserConfig` liest den Profilnamen aus YAML.
+- `CommerceQueryParser` kombiniert Profil-Phrasen/Stopwords mit den bestehenden Legacy-Listen.
+- `phrases_to_remove` und `filter_search_tokens` bleiben bewusst erhalten.
+- `RetriexEffectiveConfigProvider` validiert, dass das referenzierte Profil existiert.
+
+## Wichtig
+
+Dieser Patch entfernt noch keine alten Listen. Er verdrahtet nur die neue zentrale Struktur mit Commerce.
+
+## Pflichtchecks
+
+```bash
+bin/console mto:agent:config:validate
+bin/console mto:agent:regression:test
+bin/console mto:agent:config:audit-source --details
+bin/console mto:agent:config:audit-patterns --details
+```

config/retriex/commerce.yaml

@@ -14,6 +14,8 @@ parameters:
   # Commerce query parser configuration.
   # YAML is the only operative source of truth; PHP must not contain parser defaults.
   retriex.commerce_query.config:
+    cleanup_profile: commerce_query
+
     known_brands:
       - heyl
       - horiba

src/Commerce/CommerceQueryParser.php

@@ -7,6 +7,7 @@ namespace App\Commerce;
 use App\Commerce\Dto\CommerceSearchQuery;
 use App\Config\CommerceIntentConfig;
 use App\Config\CommerceQueryParserConfig;
+use App\Config\LanguageCleanupConfig;
 use App\Knowledge\Retrieval\QueryCleaner;
 use App\Knowledge\Text\TextNormalizer;
 
@@ -17,6 +18,7 @@ final readonly class CommerceQueryParser
         private QueryCleaner $queryCleaner,
         private CommerceQueryParserConfig $config,
         private CommerceIntentConfig $intentConfig,
+        private LanguageCleanupConfig $languageCleanupConfig,
     ) {
     }
 
@@ -177,7 +179,7 @@ final readonly class CommerceQueryParser
 
         $text = $this->wrapForPhraseReplacement($prompt);
 
-        foreach ($this->config->getPhrasesToRemove() as $phrase) {
+        foreach ($this->getCommercePhrasesToRemove() as $phrase) {
             $normalizedPhrase = $this->normalize((string) $phrase);
 
             if ($normalizedPhrase === '') {
@@ -500,6 +502,47 @@ final readonly class CommerceQueryParser
         return preg_replace($this->config->getWhitespaceCollapsePattern(), ' ', $text) ?? $text;
     }
 
+    /** @return string[] */
+    private function getCommercePhrasesToRemove(): array
+    {
+        return $this->mergeUniqueTokens(
+            $this->languageCleanupConfig->getPhrasesForProfile($this->config->getCleanupProfile()),
+            $this->config->getPhrasesToRemove()
+        );
+    }
+
+    /** @return string[] */
+    private function getCommerceFilterSearchTokens(): array
+    {
+        return $this->mergeUniqueTokens(
+            $this->languageCleanupConfig->getStopWordsForProfile($this->config->getCleanupProfile()),
+            $this->config->getFilterSearchTokens()
+        );
+    }
+
+    /**
+     * @param string[] $left
+     * @param string[] $right
+     * @return string[]
+     */
+    private function mergeUniqueTokens(array $left, array $right): array
+    {
+        $out = [];
+
+        foreach ([$left, $right] as $list) {
+            foreach ($list as $token) {
+                $token = trim(mb_strtolower((string) $token, 'UTF-8'));
+                if ($token === '' || in_array($token, $out, true)) {
+                    continue;
+                }
+
+                $out[] = $token;
+            }
+        }
+
+        return $out;
+    }
+
     private function isSearchControlToken(string $token): bool
     {
         $token = trim(mb_strtolower($token));
@@ -508,7 +551,7 @@ final readonly class CommerceQueryParser
             return true;
         }
 
-        if (in_array($token, $this->config->getFilterSearchTokens(), true)) {
+        if (in_array($token, $this->getCommerceFilterSearchTokens(), true)) {
             return true;
         }
 

src/Config/CommerceQueryParserConfig.php

@@ -16,6 +16,12 @@ final class CommerceQueryParserConfig
     ) {
     }
 
+
+    public function getCleanupProfile(): string
+    {
+        return $this->string('cleanup_profile');
+    }
+
     /** @return string[] */
     public function getKnownBrands(): array
     {

src/Config/RetriexEffectiveConfigProvider.php

@@ -25,6 +25,7 @@ final readonly class RetriexEffectiveConfigProvider
         private SalesIntentConfig $salesIntentConfig,
         private ShopServiceConfig $shopServiceConfig,
         private StopWordsConfig $stopWordsConfig,
+        private LanguageCleanupConfig $languageCleanupConfig,
         private QueryEnricherConfig $queryEnricherConfig,
         private GovernanceConfig $governanceConfig,
         private CatalogIntentConfig $catalogIntentConfig,
@@ -607,6 +608,7 @@ final readonly class RetriexEffectiveConfigProvider
     private function commerceQueryConfig(): array
     {
         return [
+            'cleanup_profile' => $this->commerceQueryParserConfig->getCleanupProfile(),
             'known_brands' => $this->commerceQueryParserConfig->getKnownBrands(),
             'phrases_to_remove' => $this->commerceQueryParserConfig->getPhrasesToRemove(),
             'filter_search_tokens' => $this->commerceQueryParserConfig->getFilterSearchTokens(),
@@ -1212,6 +1214,13 @@ final readonly class RetriexEffectiveConfigProvider
             $this->validateCommerceQueryPatterns($patterns, $errors, $warnings);
         }
 
+        $cleanupProfile = $commerceQuery['cleanup_profile'] ?? null;
+        if (!is_string($cleanupProfile) || trim($cleanupProfile) === '') {
+            $errors[] = 'commerce_query.cleanup_profile must be a non-empty string.';
+        } elseif (!in_array($cleanupProfile, $this->languageCleanupConfig->getCleanupProfileNames(), true)) {
+            $errors[] = 'commerce_query.cleanup_profile references unknown language cleanup profile: ' . $cleanupProfile . '.';
+        }
+
         $measurementPattern = $patterns['measurement_value_token'] ?? null;
         $filterTokens = $commerceQuery['filter_search_tokens'] ?? [];
         foreach ($this->governanceConfig->getRegressionProtectedMeasurementValues() as $measurementValue) {