patch 9.5

RETRIEX_PATCH_9_0F_PROMPT_ROLE_MEASUREMENT_YAML_ONLY_README.md

@@ -0,0 +1,45 @@
+# RetrieX Patch 9.0f - PromptBuilder Role Guard & Measurement Evidence YAML-only
+
+## Scope
+
+This patch continues the PromptBuilder YAML-only migration after Patch 9.0e.
+
+It converts these remaining PromptBuilder truth/product-role related areas to required YAML configuration:
+
+- `role_guard.main_device_request_keywords`
+- `role_guard.main_device_product_keywords`
+- `role_guard.accessory_product_keywords`
+- `technical_product_keywords`
+- `accessory_request_keywords`
+- `sections.measurement_evidence_label`
+- `measurement_evidence_guard.intro_rules`
+- `measurement_evidence_guard.parameters`
+- `technical_product_model_pattern`
+
+## Intent
+
+PHP no longer provides fallback truth for these PromptBuilder areas. If a required YAML value is missing or invalid, `PromptBuilderConfig` now throws a clear `InvalidArgumentException`.
+
+## Not changed
+
+This patch does not change:
+
+- Retrieval logic
+- Commerce logic
+- Shop matching
+- AgentRunner
+- SSE / frontend
+- Prompt wording content in YAML
+
+## Notes
+
+`measurement_evidence_guard.parameters` is now required to contain at least one valid parameter definition. Individual list fields inside a parameter are still normalized as before to preserve behavior.
+
+## Recommended checks
+
+```bash
+php bin/console cache:clear
+php bin/console mto:agent:config:validate
+php bin/console mto:agent:config:audit-source --details
+php bin/console mto:agent:regression:test
+```

src/Config/PromptBuilderConfig.php

@@ -6,151 +6,11 @@ namespace App\Config;
 
 final class PromptBuilderConfig
 {
-    private const MAIN_DEVICE_REQUEST_ROLE_KEYWORDS = [
-        'messanlage', 'messanlagen', 'anlage', 'anlagen', 'messgerät', 'messgeraet',
-        'messgeräte', 'messgeraete', 'analysegerät', 'analysegeraet', 'analysegeräte',
-        'analysegeraete', 'analysator', 'analysatoren', 'analyzer', 'gerät', 'geraet',
-        'geräte', 'geraete', 'system', 'systeme', 'monitor', 'monitore', 'controller',
-        'tester', 'pocket tester', 'pockettester', 'handmessgerät', 'handmessgeraet',
-        'überwachungsgerät', 'ueberwachungsgeraet', 'testomat', 'testoamt',
-    ];
-
-    private const MAIN_DEVICE_PRODUCT_ROLE_KEYWORDS = [
-        'messanlage', 'messanlagen', 'messgerät', 'messgeraet', 'messgeräte', 'messgeraete',
-        'analysegerät', 'analysegeraet', 'analysegeräte', 'analysegeraete', 'analysator',
-        'analysatoren', 'analyzer', 'online-analysator', 'online analysator',
-        'online-analysegerät', 'online analysegeraet', 'gerät', 'geraet', 'geräte',
-        'geraete', 'system', 'systeme', 'monitor', 'monitore', 'controller',
-        'tester', 'pocket tester', 'pockettester', 'handmessgerät', 'handmessgeraet',
-        'labor messgerät', 'labor-messgerät', 'labor messgeraet', 'labor-messgeraet',
-        'kombimessgerät', 'kombi-messgerät', 'kombimessgeraet', 'kombi-messgeraet',
-        'überwachungsgerät', 'ueberwachungsgeraet', 'messumformer', 'transmitter',
-        'regler', 'testomat',
-    ];
-
-    private const ACCESSORY_PRODUCT_ROLE_KEYWORDS = [
-        'indikator', 'indikatoren', 'indicator', 'reagenz', 'reagenzien', 'reagent',
-        'zubehör', 'zubehor', 'ersatzteil', 'ersatzteile', 'kit', 'set',
-        'verbrauchsmaterial', 'consumable', 'nachfüll', 'nachfuell', 'refill',
-        'lösung', 'loesung', 'solution', 'teststreifen', 'test strip', 'filter',
-        'pumpenkopf', 'motorblock', 'service set', 'serviceset', 'service-set',
-        'elektrode', 'elektroden', 'electrode', 'electrodes', 'glasschaft-elektrode',
-        'kunststoffschaft-elektrode', 'sensor', 'sensoren', 'sensors', 'sonde',
-        'sonden', 'probe', 'probes', 'messsonde', 'elektrolyt', 'kabel', 'adapter',
-        'ph-indikator', 'ph indikator', 'ph-indikatoren', 'ph indikatoren',
-    ];
-
-    private const TECHNICAL_PRODUCT_KEYWORDS = [
-            'technisch',
-            'technical',
-            'produkt',
-            'product',
-            'gerät',
-            'device',
-            'modell',
-            'model',
-            'messprinzip',
-            'measurement principle',
-            'schnittstelle',
-            'interface',
-            'relais',
-            'relay',
-            'indikator',
-            'indicator',
-            'grenzwert',
-            'threshold',
-            'messbereich',
-            'measurement range',
-            'gemessen',
-            'measured',
-            'minimaler',
-            'minimum',
-            'resthärte',
-            'resthaerte',
-            '°dh',
-            'dh',
-            'spannung',
-            'voltage',
-            'strom',
-            'current',
-            'druck',
-            'pressure',
-            'temperatur',
-            'temperature',
-            'schutzart',
-            'ip',
-            'fehlercode',
-            'error code',
-            'wasserhärte',
-            'hardness',
-            'testomat',
-            'chlor',
-            'chlormessung',
-        ];
-
-    private const MEASUREMENT_EVIDENCE_PARAMETERS = [
-        [
-            'id' => 'ph',
-            'label' => 'pH / pH-Wert',
-            'request_terms' => ['ph', 'pH', 'pH-Wert', 'ph-wert', 'ph wert'],
-            'positive_terms' => ['pH', 'pH-Wert', 'ph-wert', 'ph wert'],
-            'positive_context_terms' => ['Messung', 'messen', 'misst', 'Messbereich', 'Messparameter', 'Messgröße', 'Messgroesse', 'Bestimmung', 'bestimmen', 'Analyse', 'analysiert', 'überwachen', 'ueberwachen', 'Indikator für', 'Indikator fuer', 'Reagenz für', 'Reagenz fuer', 'Sensor', 'Elektrode'],
-            'negative_context_terms' => ['Betriebsbereich', 'Betriebsumgebung', 'Einsatzbedingungen', 'störungsfrei', 'stoerungsfrei', 'pH-Wert bei', 'ph wert bei', 'ph-wert bei', 'bei 20 °C', 'bei 20 °c', 'bei 20°C', 'bei 20°c', 'Reagenzlösung hat', 'Loesung hat', 'Lösung hat'],
-            'non_equivalent_terms' => ['p-Wert', 'p Wert', 'm-Wert', 'minus m-Wert', 'Alkalität', 'Säurekapazität', 'mmol/l'],
-            'safe_no_evidence_answer_de' => 'Ich finde in den bereitgestellten Quellen keinen sicher belegten Testomat für pH-Messung.',
-            'safe_no_accessory_evidence_answer_de' => 'Ich finde in den bereitgestellten Quellen keinen sicher belegten pH-Indikator oder ein pH-Reagenz für Messgeräte.',
-        ],
-        [
-            'id' => 'redox',
-            'label' => 'Redox / ORP',
-            'request_terms' => ['redox', 'orp', 'oxidations-reduktionspotential', 'oxidations reduktionspotential'],
-            'positive_terms' => ['Redox', 'ORP', 'Oxidations-Reduktionspotential', 'Oxidations Reduktionspotential'],
-            'positive_context_terms' => ['Messung', 'messen', 'misst', 'Messbereich', 'Messparameter', 'Messgröße', 'Messgroesse', 'Bestimmung', 'bestimmen', 'Analyse', 'analysiert', 'überwachen', 'ueberwachen', 'Indikator für', 'Indikator fuer', 'Reagenz für', 'Reagenz fuer', 'Sensor', 'Elektrode'],
-            'negative_context_terms' => ['Betriebsbereich', 'Betriebsumgebung', 'Einsatzbedingungen', 'störungsfrei', 'stoerungsfrei'],
-            'non_equivalent_terms' => [],
-            'safe_no_evidence_answer_de' => 'Ich finde in den bereitgestellten Quellen keinen sicher belegten Treffer für Redox-/ORP-Messung.',
-            'safe_no_accessory_evidence_answer_de' => 'Ich finde in den bereitgestellten Quellen keinen sicher belegten Redox-/ORP-Indikator oder ein Redox-/ORP-Reagenz für Messgeräte.',
-        ],
-        [
-            'id' => 'free_chlorine',
-            'label' => 'freies Chlor',
-            'request_terms' => ['freies chlor', 'freiem chlor', 'freien chlor', 'free chlorine'],
-            'positive_terms' => ['freies Chlor', 'freiem Chlor', 'freien Chlor', 'free chlorine'],
-            'positive_context_terms' => ['Messung', 'messen', 'misst', 'Messbereich', 'Messparameter', 'Messgröße', 'Messgroesse', 'Bestimmung', 'bestimmen', 'Analyse', 'analysiert', 'überwachen', 'ueberwachen', 'Indikator für', 'Indikator fuer', 'Reagenz für', 'Reagenz fuer', 'Sensor', 'Elektrode'],
-            'negative_context_terms' => ['Betriebsbereich', 'Betriebsumgebung', 'Einsatzbedingungen', 'störungsfrei', 'stoerungsfrei'],
-            'non_equivalent_terms' => ['Chlor gesamt', 'Gesamtchlor', 'total chlorine'],
-            'safe_no_evidence_answer_de' => 'Ich finde in den bereitgestellten Quellen keinen sicher belegten Treffer für die Messung von freiem Chlor.',
-            'safe_no_accessory_evidence_answer_de' => 'Ich finde in den bereitgestellten Quellen keinen sicher belegten Indikator oder ein Reagenz für die Messung von freiem Chlor.',
-        ],
-    ];
-
-    private const ACCESSORY_REQUEST_KEYWORDS = [
-            'passend',
-            'passende',
-            'passendes',
-            'zubehör',
-            'zubehor',
-            'dazu',
-            'indikator',
-            'indikatoren',
-            'ph-indikator',
-            'ph indikator',
-            'ph-indikatoren',
-            'ph indikatoren',
-            'reagenz',
-            'kit',
-            'set',
-            'zusatz',
-            'ergänzung',
-            'ergaenzung',
-        ];
-
     /**
      * @param array<string, mixed> $config
      */
     public function __construct(
         private readonly array $config = [],
-        private readonly ?DomainVocabularyConfig $vocabulary = null,
     ) {
     }
 
@@ -214,19 +74,7 @@ final class PromptBuilderConfig
         return $this->getRequiredInt('technical_product_keyword_match_threshold');
     }
 
-    private function getInt(string $path, int $default): int
-    {
-        $value = $this->getValue($path, $default);
 
-        return is_numeric($value) ? (int) $value : $default;
-    }
-
-    private function getFloat(string $path, float $default): float
-    {
-        $value = $this->getValue($path, $default);
-
-        return is_numeric($value) ? (float) $value : $default;
-    }
 
     private function getRequiredInt(string $path): int
     {
@@ -250,18 +98,6 @@ final class PromptBuilderConfig
         return (float) $value;
     }
 
-    private function getString(string $path, string $default): string
-    {
-        $value = $this->getValue($path, $default);
-
-        if (!is_scalar($value)) {
-            return $default;
-        }
-
-        $value = (string) $value;
-
-        return $value !== '' ? $value : $default;
-    }
 
     private function getRequiredString(string $path): string
     {
@@ -279,33 +115,6 @@ final class PromptBuilderConfig
         return $value;
     }
 
-    /**
-     * @return string[]
-     */
-    private function getStringList(string $path, array $default): array
-    {
-        $value = $this->getValue($path, $default);
-
-        if (!is_array($value)) {
-            return $default;
-        }
-
-        $out = [];
-        foreach ($value as $item) {
-            if (!is_scalar($item)) {
-                continue;
-            }
-
-            $item = trim((string) $item);
-            if ($item === '' || in_array($item, $out, true)) {
-                continue;
-            }
-
-            $out[] = $item;
-        }
-
-        return $out !== [] ? $out : $default;
-    }
 
     /**
      * @return string[]
@@ -371,28 +180,7 @@ final class PromptBuilderConfig
         return $out;
     }
 
-    /**
-     * @return string[]
-     */
-    private function vocabularyView(string $path, array $fallback): array
-    {
-        return $this->vocabulary?->view($path, $fallback) ?? $fallback;
-    }
 
-    private function getValue(string $path, mixed $default): mixed
-    {
-        $current = $this->config;
-
-        foreach (explode('.', $path) as $segment) {
-            if (!is_array($current) || !array_key_exists($segment, $current)) {
-                return $default;
-            }
-
-            $current = $current[$segment];
-        }
-
-        return $current;
-    }
 
     private function getOptionalValue(string $path): mixed
     {
@@ -737,7 +525,7 @@ final class PromptBuilderConfig
      */
     public function getMainDeviceRequestRoleKeywords(): array
     {
-        return $this->getStringList('role_guard.main_device_request_keywords', self::MAIN_DEVICE_REQUEST_ROLE_KEYWORDS);
+        return $this->getRequiredStringList('role_guard.main_device_request_keywords');
     }
 
     /**
@@ -745,7 +533,7 @@ final class PromptBuilderConfig
      */
     public function getMainDeviceProductRoleKeywords(): array
     {
-        return $this->getStringList('role_guard.main_device_product_keywords', self::MAIN_DEVICE_PRODUCT_ROLE_KEYWORDS);
+        return $this->getRequiredStringList('role_guard.main_device_product_keywords');
     }
 
     /**
@@ -753,7 +541,7 @@ final class PromptBuilderConfig
      */
     public function getAccessoryProductRoleKeywords(): array
     {
-        return $this->getStringList('role_guard.accessory_product_keywords', self::ACCESSORY_PRODUCT_ROLE_KEYWORDS);
+        return $this->getRequiredStringList('role_guard.accessory_product_keywords');
     }
 
     /**
@@ -761,10 +549,7 @@ final class PromptBuilderConfig
      */
     public function getTechnicalProductKeywords(): array
     {
-        return $this->getStringList(
+        return $this->getRequiredStringList('technical_product_keywords');
-            'technical_product_keywords',
-            $this->vocabularyView('prompt.technical_product_keywords', self::TECHNICAL_PRODUCT_KEYWORDS)
-        );
     }
 
     /**
@@ -772,15 +557,12 @@ final class PromptBuilderConfig
      */
     public function getAccessoryRequestKeywords(): array
     {
-        return $this->getStringList(
+        return $this->getRequiredStringList('accessory_request_keywords');
-            'accessory_request_keywords',
-            $this->vocabularyView('prompt.accessory_request_keywords', self::ACCESSORY_REQUEST_KEYWORDS)
-        );
     }
 
     public function getMeasurementEvidenceSectionLabel(): string
     {
-        return $this->getString('sections.measurement_evidence_label', 'MEASUREMENT PARAMETER EVIDENCE CHECK');
+        return $this->getRequiredString('sections.measurement_evidence_label');
     }
 
     /**
@@ -788,11 +570,7 @@ final class PromptBuilderConfig
      */
     public function getMeasurementEvidenceIntroRules(): array
     {
-        return $this->getStringList('measurement_evidence_guard.intro_rules', [
+        return $this->getRequiredStringList('measurement_evidence_guard.intro_rules');
-            '- This block is generated from the current user question and is stricter than broad product-selection wording.',
-            '- For measurement-parameter questions, technical suitability requires explicit positive evidence for the requested parameter in the same source record.',
-            '- Similar water-treatment parameters, abbreviations, units, product families, search queries, or ranking positions are not enough.',
-        ]);
     }
 
     /**
@@ -800,10 +578,10 @@ final class PromptBuilderConfig
      */
     public function getMeasurementEvidenceParameters(): array
     {
-        $value = $this->getValue('measurement_evidence_guard.parameters', self::MEASUREMENT_EVIDENCE_PARAMETERS);
+        $value = $this->getRequiredValue('measurement_evidence_guard.parameters');
 
         if (!is_array($value)) {
-            return self::MEASUREMENT_EVIDENCE_PARAMETERS;
+            throw new \InvalidArgumentException('RetrieX prompt config value "measurement_evidence_guard.parameters" must be a list of parameter definitions.');
         }
 
         $out = [];
@@ -837,7 +615,11 @@ final class PromptBuilderConfig
             ];
         }
 
-        return $out !== [] ? $out : self::MEASUREMENT_EVIDENCE_PARAMETERS;
+        if ($out === []) {
+            throw new \InvalidArgumentException('RetrieX prompt config value "measurement_evidence_guard.parameters" must contain at least one valid parameter definition.');
+        }
+
+        return $out;
     }
 
     /**
@@ -866,6 +648,6 @@ final class PromptBuilderConfig
 
     public function getTechnicalProductModelPattern(): string
     {
-        return $this->getString('technical_product_model_pattern', '/\b[\p{L}]{2,}\s?\d{2,5}\b/u');
+        return $this->getRequiredString('technical_product_model_pattern');
     }
 }