This commit is contained in:
2026-05-08 00:05:51 +08:00
parent 549b706fcc
commit ed70a140a2
40 changed files with 5590 additions and 36 deletions
@@ -7,6 +7,22 @@ use support\Db;
class ChunkSearchIndexRepository
{
public function resetEmbeddedChunksToPending(?string $archiveUid = null): int
{
$query = Db::table('chunks')
->where('embedding_status', EmbeddingStatus::EMBEDDED);
if ($archiveUid !== null && trim($archiveUid) !== '') {
$query->where('archive_uid', trim($archiveUid));
}
return $query->update([
'search_index_status' => SearchIndexStatus::PENDING,
'search_index_error' => null,
'search_index_updated_at' => null,
]);
}
public function queuePendingArchiveTasks(int $limit): array
{
$statuses = [
@@ -63,6 +79,7 @@ class ChunkSearchIndexRepository
'chunks.created_time',
'chunks.updated_time',
'archives.title',
'archives.summary',
'archives.source',
'archives.author',
'archives.year',
@@ -105,6 +122,7 @@ class ChunkSearchIndexRepository
'page_start' => $row->page_start === null ? null : (int) $row->page_start,
'page_end' => $row->page_end === null ? null : (int) $row->page_end,
'title' => $row->title,
'summary' => $row->summary,
'source' => $row->source,
'author' => $row->author,
'year' => $row->year === null ? null : (int) $row->year,
@@ -16,6 +16,7 @@ class OpenSearchChunkIndex
$index = $this->indexName();
if ($client->indices()->exists(['index' => $index])) {
$this->ensureProperties($client, $index);
return;
}
@@ -64,6 +65,7 @@ class OpenSearchChunkIndex
'page_start' => ['type' => 'integer'],
'page_end' => ['type' => 'integer'],
'title' => $this->textWithKeyword(),
'summary' => ['type' => 'text'],
'source' => $this->textWithKeyword(),
'author' => $this->textWithKeyword(),
'year' => ['type' => 'integer'],
@@ -93,6 +95,31 @@ class OpenSearchChunkIndex
return $this->client ?? (new OpenSearchClientFactory())->make();
}
private function ensureProperties(Client $client, string $index): void
{
$mapping = $client->indices()->getMapping(['index' => $index]);
$existing = $mapping[$index]['mappings']['properties'] ?? [];
$desired = $this->mapping()['mappings']['properties'] ?? [];
$missing = [];
foreach ($desired as $field => $definition) {
if (!array_key_exists($field, $existing)) {
$missing[$field] = $definition;
}
}
if ($missing === []) {
return;
}
$client->indices()->putMapping([
'index' => $index,
'body' => [
'properties' => $missing,
],
]);
}
private function indexName(): string
{
return config('opensearch.indices.chunks', 'proofdb_chunks');
@@ -39,6 +39,7 @@ class OpenSearchSearchService
'fields' => [
'text^4',
'title^3',
'summary^2',
'source^2',
'author^2',
'series^2',
@@ -219,6 +220,7 @@ class OpenSearchSearchService
'page_start' => $source['page_start'] ?? null,
'page_end' => $source['page_end'] ?? null,
'title' => $source['title'] ?? null,
'summary' => $source['summary'] ?? null,
'source' => $source['source'] ?? null,
'author' => $source['author'] ?? null,
'year' => $source['year'] ?? null,
@@ -322,6 +324,7 @@ class OpenSearchSearchService
'page_start',
'page_end',
'title',
'summary',
'source',
'author',
'year',