p59 + p60

This commit is contained in:
team 1
2026-05-07 07:52:52 +02:00
parent 56646a0c3b
commit 87c2134e6c
20 changed files with 808 additions and 1256 deletions

View File

@@ -14,6 +14,7 @@ final class NdjsonHybridRetrieverConfig
public function __construct(
private array $config = [],
private ?DomainVocabularyConfig $vocabulary = null,
private ?GenreConfig $genreConfig = null,
) {
}
@@ -146,7 +147,8 @@ final class NdjsonHybridRetrieverConfig
/** @return array<string, string[]> */
public function exactSelectionTokenVariantPrefixes(): array
{
return $this->requiredStringListMap('exact_selection_token_variant_prefixes');
return $this->genreStringListMap('retrieval_and_language.exact_selection.token_variant_prefixes')
?: $this->requiredStringListMap('exact_selection_token_variant_prefixes');
}
/** @return string[] */
@@ -158,43 +160,50 @@ final class NdjsonHybridRetrieverConfig
/** @return string[] */
public function exactSelectionIndicatorQuestionTokens(): array
{
return $this->requiredStringList('exact_selection_indicator_question_tokens');
return $this->genreStringList('retrieval_and_language.exact_selection.indicator_question_tokens')
?: $this->requiredStringList('exact_selection_indicator_question_tokens');
}
/** @return string[] */
public function exactSelectionIndicatorQuestionPhrases(): array
{
return $this->requiredStringList('exact_selection_indicator_question_phrases');
return $this->genreStringList('retrieval_and_language.exact_selection.indicator_question_phrases')
?: $this->requiredStringList('exact_selection_indicator_question_phrases');
}
/** @return string[] */
public function exactSelectionIndicatorTableHeadingPatterns(): array
{
return $this->requiredStringList('exact_selection_indicator_table_heading_patterns');
return $this->genreStringList('retrieval_and_language.exact_selection.indicator_table_heading_patterns')
?: $this->requiredStringList('exact_selection_indicator_table_heading_patterns');
}
/** @return string[] */
public function exactSelectionIndicatorTableHeaderPatterns(): array
{
return $this->requiredStringList('exact_selection_indicator_table_header_patterns');
return $this->genreStringList('retrieval_and_language.exact_selection.indicator_table_header_patterns')
?: $this->requiredStringList('exact_selection_indicator_table_header_patterns');
}
/** @return string[] */
public function exactSelectionIndicatorTableRowPatterns(): array
{
return $this->requiredStringList('exact_selection_indicator_table_row_patterns');
return $this->genreStringList('retrieval_and_language.exact_selection.indicator_table_row_patterns')
?: $this->requiredStringList('exact_selection_indicator_table_row_patterns');
}
/** @return string[] */
public function exactSelectionIndicatorTableRequiredPrimaryTerms(): array
{
return $this->requiredStringList('exact_selection_indicator_table_required_primary_terms');
return $this->genreStringList('retrieval_and_language.exact_selection.indicator_table_required_primary_terms')
?: $this->requiredStringList('exact_selection_indicator_table_required_primary_terms');
}
/** @return string[] */
public function exactSelectionIndicatorTableRequiredContextTerms(): array
{
return $this->requiredStringList('exact_selection_indicator_table_required_context_terms');
return $this->genreStringList('retrieval_and_language.exact_selection.indicator_table_required_context_terms')
?: $this->requiredStringList('exact_selection_indicator_table_required_context_terms');
}
/** @return string[] */
@@ -370,6 +379,23 @@ final class NdjsonHybridRetrieverConfig
];
}
/** @return string[] */
private function genreStringList(string $path): array
{
return $this->genreConfig?->getValueStringList($path) ?? [];
}
/** @return array<string, string[]> */
private function genreStringListMap(string $path): array
{
$value = $this->genreConfig?->getValueArray($path) ?? [];
if ($value === []) {
return [];
}
return $this->normalizeStringListMap($value);
}
private function requiredInt(string $key, int $min = PHP_INT_MIN, ?int $max = null): int
{
$value = $this->requiredValue($key);
@@ -458,6 +484,35 @@ final class NdjsonHybridRetrieverConfig
return $out;
}
/** @return array<string, string[]> */
private function normalizeStringListMap(array $value): array
{
$out = [];
foreach ($value as $mapKey => $items) {
if (!is_string($mapKey) || trim($mapKey) === '' || !is_array($items)) {
continue;
}
$cleanItems = [];
foreach ($items as $item) {
if (!is_scalar($item)) {
continue;
}
$item = trim((string) $item);
if ($item !== '' && !in_array($item, $cleanItems, true)) {
$cleanItems[] = $item;
}
}
if ($cleanItems !== []) {
$out[trim($mapKey)] = $cleanItems;
}
}
return $out;
}
/**
* @return array<string, string[]>
*/