@nicia-ai/typegraph 0.31.0 → 0.32.0

package/dist/{chunk-5CN7B4CZ.cjs.map → chunk-YW7GHMF4.cjs.map}

@@ -1 +1 @@
-{"version":3,"sources":["../src/backend/drizzle/columns/vector.ts","../src/query/dialect/vector/pgvector-strategy.ts"],"names":["customType","assertFiniteEmbedding","sql","vectorPhysicalName","quoteIdentifier","VECTOR_CONTRIBUTION_PREFIX","quotedTableName","vectorScoreExpression","vectorMinScoreCondition"],"mappings":";;;;;;AA0CsBA,iBAAA,CAInB;AAAA,EACD,SAAS,MAAA,EAAQ;AAEf,IAAA,OAAO,MAAA,EAAQ,UAAA,GAAa,CAAA,OAAA,EAAU,MAAA,CAAO,UAAU,CAAA,CAAA,CAAA,GAAM,QAAA;AAAA,EAC/D,CAAA;AAAA,EAEA,SAAS,KAAA,EAAyB;AAEhC,IAAA,OAAO,CAAA,CAAA,EAAI,KAAA,CAAM,IAAA,CAAK,GAAG,CAAC,CAAA,CAAA,CAAA;AAAA,EAC5B,CAAA;AAAA,EAEA,WAAW,KAAA,EAAyB;AAGlC,IAAA,IAAI,KAAA,CAAM,OAAA,CAAQ,KAAK,CAAA,EAAG;AACxB,MAAA,OAAO,KAAA;AAAA,IACT;AAEA,IAAA,MAAM,OAAA,GAAU,KAAA,CAAM,KAAA,CAAM,CAAA,EAAG,EAAE,CAAA;AACjC,IAAA,IAAI,YAAY,EAAA,EAAI;AAClB,MAAA,OAAO,EAAC;AAAA,IACV;AACA,IAAA,OAAO,OAAA,CAAQ,KAAA,CAAM,GAAG,CAAA,CAAE,GAAA,CAAI,CAAC,CAAA,KAAM,MAAA,CAAO,UAAA,CAAW,CAAA,CAAE,IAAA,EAAM,CAAC,CAAA;AAAA,EAClE;AACF,CAAC;AAgBM,SAAS,aAAa,SAAA,EAAsC;AACjE,EAAA,OAAO,CAAA,CAAA,EAAI,SAAA,CAAU,IAAA,CAAK,GAAG,CAAC,CAAA,CAAA,CAAA;AAChC;;;ACzCA,IAAM,YAAA,GAAe,QAAA;AACrB,IAAM,YAAA,GAAe,WAAA;AASd,IAAM,kBAAA,GAAqB,GAAA;AAM3B,SAAS,uBAAuB,QAAA,EAAyB;AAC9D,EAAA,IAAI,aAAa,MAAA,EAAW;AAC5B,EAAA,IAAI,CAAC,MAAA,CAAO,SAAA,CAAU,QAAQ,CAAA,IAAK,YAAY,CAAA,EAAG;AAChD,IAAA,MAAM,IAAI,KAAA,CAAM,CAAA,0CAAA,EAA6C,QAAQ,CAAA,CAAE,CAAA;AAAA,EACzE;AACA,EAAA,IAAI,WAAW,kBAAA,EAAoB;AACjC,IAAA,MAAM,IAAI,UAAA;AAAA,MACR,CAAA,wBAAA,EAAsB,kBAAkB,CAAA,8CAAA,EACjB,kBAAkB,WAAW,QAAQ,CAAA;AAAA,KAC9D;AAAA,EACF;AACF;AAQA,IAAM,uBAAA,GAA0B,IAAA;AAGhC,IAAM,cAAA,GAAiB,EAAA;AAEvB,IAAM,4BAAA,GAA+B,EAAA;AAErC,IAAM,qBAAA,GAAwB,GAAA;AAE9B,IAAM,qBAAA,GAA4C;AAAA,EAChD,SAAA,EAAW,IAAA;AAAA,EACX,OAAA,EAAS,CAAC,QAAA,EAAU,IAAA,EAAM,eAAe,CAAA;AAAA,EACzC,UAAA,EAAY,CAAC,MAAA,EAAQ,SAAA,EAAW,MAAM,CAAA;AAAA,EACtC,aAAA,EAAe;AACjB,CAAA;AAGA,SAAS,aAAa,IAAA,EAA2B;AAC/C,EAAA,OAAO,IAAA,CAAK,SAAA,KAAc,MAAA,IAAU,IAAA,CAAK,SAAA,KAAc,SAAA;AACzD;AAOA,SAAS,cAAc,MAAA,EAA8B;AACnD,EAAA,QAAQ,MAAA;AAAQ,IACd,KAAK,QAAA,EAAU;AACb,MAAA,OAAO,mBAAA;AAAA,IACT;AAAA,IACA,KAAK,IAAA,EAAM;AACT,MAAA,OAAO,eAAA;AAAA,IACT;AAAA,IACA,KAAK,eAAA,EAAiB;AACpB,MAAA,OAAO,eAAA;AAAA,IACT;AAAA,IACA,SAAS;AACP,MAAA,MAAM,WAAA,GAAqB,MAAA;AAC3B,MAAA,MAAM,IAAI,KAAA,CAAM,CAAA,2BAAA,EAA8B,MAAA,CAAO,WAAW,CAAC,CAAA,CAAE,CAAA;AAAA,IACrE;AAAA;AAEJ;AAOA,SAAS,aAAA,CAAc,WAA8B,IAAA,EAAmB;AACtE,EAAAC,uCAAA,CAAsB,WAAW,IAAI,CAAA;AACrC,EAAA,OAAOC,cAAA,CAAA,EAAM,YAAA,CAAa,SAAS,CAAC,CAAA,QAAA,CAAA;AACtC;AAEA,SAAS,kBAAA,CACP,eAAA,EACA,cAAA,EACA,MAAA,EACK;AACL,EAAA,MAAM,KAAA,GAAQ,aAAA,CAAc,cAAA,EAAgB,gBAAgB,CAAA;AAC5D,EAAA,QAAQ,MAAA;AAAQ,IACd,KAAK,QAAA,EAAU;AACb,MAAA,OAAOA,cAAA,CAAA,CAAA,EAAO,eAAe,CAAA,KAAA,EAAQ,KAAK,CAAA,CAAA,CAAA;AAAA,IAC5C;AAAA,IACA,KAAK,IAAA,EAAM;AACT,MAAA,OAAOA,cAAA,CAAA,CAAA,EAAO,eAAe,CAAA,KAAA,EAAQ,KAAK,CAAA,CAAA,CAAA;AAAA,IAC5C;AAAA,IACA,KAAK,eAAA,EAAiB;AACpB,MAAA,OAAOA,cAAA,CAAA,CAAA,EAAO,eAAe,CAAA,KAAA,EAAQ,KAAK,CAAA,CAAA,CAAA;AAAA,IAC5C;AAAA,IACA,SAAS;AACP,MAAA,MAAM,WAAA,GAAqB,MAAA;AAC3B,MAAA,MAAM,IAAI,KAAA,CAAM,CAAA,2BAAA,EAA8B,MAAA,CAAO,WAAW,CAAC,CAAA,CAAE,CAAA;AAAA,IACrE;AAAA;AAEJ;AAEO,IAAM,gBAAA,GAAmC;AAAA,EAC9C,IAAA,EAAM,UAAA;AAAA,EACN,YAAA,EAAc,qBAAA;AAAA,EAEd,SAAA,CAAU,OAAA,EAAS,QAAA,EAAU,SAAA,EAAW;AACtC,IAAA,OAAOC,oCAAA,CAAmB,YAAA,EAAc,OAAA,EAAS,QAAA,EAAU,SAAS,CAAA;AAAA,EACtE,CAAA;AAAA,EAEA,YAAY,IAAA,EAA4C;AACtD,IAAA,MAAM,KAAA,GAAQ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS,CAAA;AACxE,IAAA,MAAM,IAAA,GAAOC,kCAAgB,KAAK,CAAA;AAKlC,IAAA,MAAM,SAAA,GAAY;AAAA,MAChB,8BAA8B,IAAI,CAAA;AAAA;AAAA;AAAA,qBAAA,EAGjB,KAAK,UAAU,CAAA;AAAA;AAAA;AAAA;AAAA,EAAA;AAAA,KAKlC;AAWA,IAAA,OAAO;AAAA,MACL;AAAA,QACE,WAAA,EAAa,GAAGC,4CAA0B,CAAA,CAAA,EAAI,KAAK,QAAQ,CAAA,CAAA,EAAI,KAAK,SAAS,CAAA,CAAA;AAAA,QAC7E,KAAA,EAAO,UAAA;AAAA,QACP,SAAA,EAAW,KAAA;AAAA,QACX,SAAA;AAAA,QACA,aAAA,EAAe;AAAA;AACjB,KACF;AAAA,EACF,CAAA;AAAA,EAEA,WAAA,CAAY,IAAA,EAAM,MAAA,EAA+B,SAAA,EAA2B;AAC1E,IAAA,MAAM,KAAA,GAAQC,iCAAA;AAAA,MACZ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS;AAAA,KAC5D;AACA,IAAA,MAAM,KAAA,GAAQ,aAAA,CAAc,MAAA,CAAO,SAAA,EAAW,WAAW,CAAA;AACzD,IAAA,OAAO;AAAA,MACLJ,cAAA;AAAA,oBAAA,EACgB,KAAK,CAAA;AAAA,gBAAA,EACT,MAAA,CAAO,OAAO,CAAA,EAAA,EAAK,MAAA,CAAO,MAAM,KAAK,KAAK,CAAA,EAAA,EAAK,SAAS,CAAA,EAAA,EAAK,SAAS,CAAA;AAAA;AAAA;AAAA,MAAA;AAAA,KAIpF;AAAA,EACF,CAAA;AAAA,EAEA,WAAA,CAAY,MAAM,MAAA,EAA+C;AAC/D,IAAA,MAAM,KAAA,GAAQI,iCAAA;AAAA,MACZ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS;AAAA,KAC5D;AACA,IAAA,OAAO;AAAA,MACLJ,cAAA;AAAA,oBAAA,EACgB,KAAK;AAAA,2BAAA,EACE,MAAA,CAAO,OAAO,CAAA,iBAAA,EAAoB,MAAA,CAAO,MAAM;AAAA,MAAA;AAAA,KAExE;AAAA,EACF,CAAA;AAAA,EAEA,WAAA,CAAY,MAAM,MAAA,EAAiC;AACjD,IAAA,MAAM,KAAA,GAAQI,iCAAA;AAAA,MACZ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS;AAAA,KAC5D;AACA,IAAA,MAAM,eAAA,GAAkBJ,iBAAM,KAAK,CAAA,YAAA,CAAA;AACnC,IAAA,MAAM,QAAA,GAAW,kBAAA;AAAA,MACf,eAAA;AAAA,MACA,MAAA,CAAO,cAAA;AAAA,MACP,MAAA,CAAO;AAAA,KACT;AACA,IAAA,MAAM,KAAA,GAAQK,uCAAA,CAAsB,QAAA,EAAU,MAAA,CAAO,MAAM,CAAA;AAK3D,IAAA,MAAM,aAAoB,CAACL,cAAA,CAAA,EAAM,KAAK,CAAA,cAAA,EAAiB,MAAA,CAAO,OAAO,CAAA,CAAE,CAAA;AACvE,IAAA,IAAI,MAAA,CAAO,aAAa,MAAA,EAAW;AACjC,MAAA,UAAA,CAAW,IAAA;AAAA,QACTM,yCAAA,CAAwB,QAAA,EAAU,MAAA,CAAO,MAAA,EAAQ,OAAO,QAAQ;AAAA,OAClE;AAAA,IACF;AACA,IAAA,OAAON,cAAA;AAAA,aAAA,EACI,KAAK,0BAA0B,KAAK,CAAA;AAAA,WAAA,EACtC,KAAK;AAAA,YAAA,EACJA,cAAA,CAAI,IAAA,CAAK,UAAA,EAAYA,cAAA,CAAA,KAAA,CAAU,CAAC;AAAA,eAAA,EAC7B,QAAQ,CAAA;AAAA,YAAA,EACX,OAAO,KAAK;AAAA,IAAA,CAAA;AAAA,EAExB,CAAA;AAAA,EAEA,kBAAA,CAAmB,eAAA,EAAiB,cAAA,EAAgB,MAAA,EAAQ;AAC1D,IAAA,OAAO,kBAAA,CAAmB,eAAA,EAAiB,cAAA,EAAgB,MAAM,CAAA;AAAA,EACnE,CAAA;AAAA,EAEA,gBAAA,CAAiB,MAAM,OAAA,EAA0B;AAC/C,IAAA,IAAI,CAAC,YAAA,CAAa,IAAI,CAAA,EAAG,OAAO,MAAA;AAChC,IAAA,OAAOA,cAAA,CAAI,GAAA;AAAA,MACT,gBAAA;AAAA,QACE,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS,CAAA;AAAA,QAC1D,IAAA;AAAA,QACA,SAAS,UAAA,KAAe;AAAA;AAC1B,KACF;AAAA,EACF,CAAA;AAAA,EAEA,eAAe,IAAA,EAAuB;AACpC,IAAA,IAAI,CAAC,YAAA,CAAa,IAAI,CAAA,EAAG,OAAO,MAAA;AAChC,IAAA,MAAM,SAAA,GAAY,kBAAkB,IAAI,CAAA;AACxC,IAAA,OAAOA,eAAI,GAAA,CAAI,CAAA,qBAAA,EAAwBE,iCAAA,CAAgB,SAAS,CAAC,CAAA,CAAE,CAAA;AAAA,EACrE,CAAA;AAAA,EAEA,iBAAiB,IAAA,EAAyB;AACxC,IAAA,MAAM,KAAA,GAAQA,iCAAA;AAAA,MACZ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS;AAAA,KAC5D;AAEA,IAAA,OAAO,CAAC,CAAA,qBAAA,EAAwB,KAAK,CAAA,QAAA,CAAU,CAAA;AAAA,EACjD;AACF;AAEA,SAAS,kBAAkB,IAAA,EAA0B;AACnD,EAAA,OAAOD,oCAAA;AAAA,IACL,YAAA;AAAA,IACA,IAAA,CAAK,OAAA;AAAA,IACL,IAAA,CAAK,QAAA;AAAA,IACL,IAAA,CAAK;AAAA,GACP;AACF;AAEA,SAAS,gBAAA,CACP,KAAA,EACA,IAAA,EACA,UAAA,EACQ;AACR,EAAA,MAAM,SAAA,GAAYC,iCAAA,CAAgB,iBAAA,CAAkB,IAAI,CAAC,CAAA;AACzD,EAAA,MAAM,WAAA,GAAcA,kCAAgB,KAAK,CAAA;AACzC,EAAA,MAAM,OAAA,GAAU,aAAA,CAAc,IAAA,CAAK,MAAM,CAAA;AAEzC,EAAA,MAAM,CAAA,GAAI,IAAA,CAAK,WAAA,EAAa,CAAA,IAAK,cAAA;AACjC,EAAA,MAAM,cAAA,GACJ,IAAA,CAAK,WAAA,EAAa,cAAA,IAAkB,4BAAA;AACtC,EAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,WAAA,EAAa,KAAA,IAAS,qBAAA;AAGzC,EAAA,MAAM,MAAA,GAAS,CAAA,YAAA,EAAe,UAAA,GAAa,eAAA,GAAkB,EAAE,CAAA,cAAA,CAAA;AAE/D,EAAA,QAAQ,KAAK,SAAA;AAAW,IACtB,KAAK,MAAA,EAAQ;AACX,MAAA,OAAO,CAAA,EAAG,MAAM,CAAA,CAAA,EAAI,SAAS,CAAA,IAAA,EAAO,WAAW,CAAA,yBAAA,EAA4B,OAAO,CAAA,YAAA,EAAe,CAAC,CAAA,oBAAA,EAAuB,cAAc,CAAA,EAAA,CAAA;AAAA,IACzI;AAAA,IACA,KAAK,SAAA,EAAW;AACd,MAAA,OAAO,CAAA,EAAG,MAAM,CAAA,CAAA,EAAI,SAAS,OAAO,WAAW,CAAA,4BAAA,EAA+B,OAAO,CAAA,gBAAA,EAAmB,KAAK,CAAA,EAAA,CAAA;AAAA,IAC/G;AAAA,IACA,KAAK,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,KAAA;AAAA,QACR;AAAA,OACF;AAAA,IACF;AAAA,IACA,SAAS;AACP,MAAA,MAAM,cAAqB,IAAA,CAAK,SAAA;AAChC,MAAA,MAAM,IAAI,KAAA,CAAM,CAAA,+BAAA,EAAkC,MAAA,CAAO,WAAW,CAAC,CAAA,CAAE,CAAA;AAAA,IACzE;AAAA;AAEJ","file":"chunk-5CN7B4CZ.cjs","sourcesContent":["/**\n * Custom Drizzle column type for pgvector's VECTOR type.\n *\n * Provides native support for PostgreSQL vector embeddings without\n * requiring runtime casts in queries.\n *\n * @example\n * ```typescript\n * import { pgTable, text, integer } from \"drizzle-orm/pg-core\";\n * import { vector } from \"./columns/vector\";\n *\n * const embeddings = pgTable(\"embeddings\", {\n *   id: text(\"id\").primaryKey(),\n *   embedding: vector(\"embedding\"),\n *   dimensions: integer(\"dimensions\"),\n * });\n * ```\n */\nimport { customType } from \"drizzle-orm/pg-core\";\n\n/**\n * PostgreSQL vector column type for pgvector extension.\n *\n * Stores embeddings as native VECTOR type, enabling efficient\n * similarity search without runtime casts.\n *\n * The column type is unparameterized (`VECTOR` not `VECTOR(N)`)\n * to support multiple embedding dimensions in a single table.\n * Dimension validation is handled at the application level.\n *\n * @example\n * ```typescript\n * // In table definition\n * embedding: vector(\"embedding\").notNull(),\n *\n * // In queries - no cast needed\n * db.execute(sql`\n *   SELECT * FROM embeddings\n *   WHERE embedding <=> ${formatVector([0.1, 0.2, 0.3])} < 0.5\n * `);\n * ```\n */\nexport const vector = customType<{\n  data: number[];\n  driverData: string;\n  config: { dimensions?: number };\n}>({\n  dataType(config) {\n    // Use parameterized type if dimensions specified, otherwise unparameterized\n    return config?.dimensions ? `vector(${config.dimensions})` : \"vector\";\n  },\n\n  toDriver(value: number[]): string {\n    // Format as pgvector literal: [1.0,2.0,3.0]\n    return `[${value.join(\",\")}]`;\n  },\n\n  fromDriver(value: string): number[] {\n    // pgvector returns '[1,2,3]' format\n    // Handle both string and already-parsed array (some drivers differ)\n    if (Array.isArray(value)) {\n      return value as number[];\n    }\n    // Parse the string representation\n    const content = value.slice(1, -1); // Remove [ and ]\n    if (content === \"\") {\n      return [];\n    }\n    return content.split(\",\").map((s) => Number.parseFloat(s.trim()));\n  },\n});\n\n/**\n * Formats a number array as a pgvector literal string.\n *\n * Use this when building raw SQL queries with embeddings.\n *\n * @example\n * ```typescript\n * const embedding = [0.1, 0.2, 0.3];\n * db.execute(sql`\n *   SELECT * FROM embeddings\n *   WHERE embedding <=> ${formatVector(embedding)}::vector < 0.5\n * `);\n * ```\n */\nexport function formatVector(embedding: readonly number[]): string {\n  return `[${embedding.join(\",\")}]`;\n}\n","/**\n * pgvector strategy.\n *\n * Postgres' pgvector extension is the one engine where the *same* SQL serves\n * brute-force and ANN: `ORDER BY embedding <=> q LIMIT k` scans sequentially\n * with no index and is rewritten to an HNSW/IVFFlat scan the moment a matching\n * index exists — the planner picks it up automatically. So `buildSearch` never\n * branches on `slot.indexType`; it emits one relevance scan and lets Postgres\n * decide. (Contrast libSQL/sqlite-vec, whose ANN needs distinct syntax.)\n *\n * Storage is per-`(nodeKind, fieldPath)`: a `vector(N)` column carrying the\n * field's fixed dimension `N`, with the operator-class-typed ANN index folded\n * into the same table contribution so materialization builds table + index\n * together. The typed column means the index does not need the legacy\n * `::vector(N)` cast-in-index escape hatch the old shared table relied on.\n *\n * Metric support is the full pgvector set — `cosine`, `l2`, `inner_product` —\n * advertised as data on `capabilities`, mirroring `vector_cosine_ops` /\n * `vector_l2_ops` / `vector_ip_ops` and the `<=>` / `<->` / `<#>` operators.\n *\n * Mirrors the `libsqlVectorStrategy` / `sqliteVecStrategy` structure while\n * letting Postgres' planner pick the ANN index when one exists.\n */\nimport { type SQL, sql } from \"drizzle-orm\";\n\nimport { formatVector } from \"../../../backend/drizzle/columns/vector\";\nimport { quotedTableName } from \"../../../backend/drizzle/operations/shared\";\nimport { type StrategyTableContribution } from \"../../../backend/table-contribution\";\nimport {\n  type DeleteEmbeddingParams,\n  type UpsertEmbeddingParams,\n  type VectorCapabilities,\n  type VectorMetric,\n  type VectorSearchParams,\n} from \"../../../backend/types\";\nimport {\n  assertFiniteEmbedding,\n  quoteIdentifier,\n  VECTOR_CONTRIBUTION_PREFIX,\n  vectorMinScoreCondition,\n  vectorPhysicalName,\n  vectorScoreExpression,\n  type VectorSlot,\n  type VectorStrategy,\n} from \"../vector-strategy\";\n\n/** Physical-name prefixes for the per-field table and its ANN index. */\nconst TABLE_PREFIX = \"tg_vec\";\nconst INDEX_PREFIX = \"tg_vecidx\";\n\n/**\n * pgvector defines `hnsw.ef_search` as an integer GUC with a valid range\n * of 1..1000; `SET hnsw.ef_search = 1001` errors at the server. The\n * backend validates a per-search `efSearch` against this ceiling before\n * inlining it into `SET LOCAL`, surfacing a clear message instead of a raw\n * pgvector error.\n */\nexport const MAX_HNSW_EF_SEARCH = 1000;\n\n/**\n * Validates a per-search `efSearch` override against pgvector's\n * `hnsw.ef_search` valid range. `undefined` (no override) is accepted.\n */\nexport function assertPgvectorEfSearch(efSearch?: number): void {\n  if (efSearch === undefined) return;\n  if (!Number.isInteger(efSearch) || efSearch <= 0) {\n    throw new Error(`efSearch must be a positive integer, got: ${efSearch}`);\n  }\n  if (efSearch > MAX_HNSW_EF_SEARCH) {\n    throw new RangeError(\n      `efSearch must be ≤ ${MAX_HNSW_EF_SEARCH} (pgvector's hnsw.ef_search ` +\n        `valid range is 1..${MAX_HNSW_EF_SEARCH}), got: ${efSearch}`,\n    );\n  }\n}\n\n/**\n * pgvector's `vector` type caps at 16,000 dimensions for a stored column\n * (2,000 for an indexed column, but the column-type ceiling is what governs\n * storage; recall degradation beyond 2,000 is a tuning concern, not a hard\n * limit advertised here).\n */\nconst PGVECTOR_MAX_DIMENSIONS = 16_000;\n\n/** Default HNSW `m` (max connections per layer) — pgvector's own default. */\nconst DEFAULT_HNSW_M = 16;\n/** Default HNSW `ef_construction` (build candidate list) — pgvector's own default. */\nconst DEFAULT_HNSW_EF_CONSTRUCTION = 64;\n/** Default IVFFlat `lists` (inverted lists) — pgvector's own default. */\nconst DEFAULT_IVFFLAT_LISTS = 100;\n\nconst PGVECTOR_CAPABILITIES: VectorCapabilities = {\n  supported: true,\n  metrics: [\"cosine\", \"l2\", \"inner_product\"],\n  indexTypes: [\"hnsw\", \"ivfflat\", \"none\"],\n  maxDimensions: PGVECTOR_MAX_DIMENSIONS,\n};\n\n/** Whether a slot's declared index type materializes a real pgvector ANN index. */\nfunction usesAnnIndex(slot: VectorSlot): boolean {\n  return slot.indexType === \"hnsw\" || slot.indexType === \"ivfflat\";\n}\n\n/**\n * pgvector operator class for a metric's ANN index — `vector_cosine_ops` /\n * `vector_l2_ops` / `vector_ip_ops`, mirroring the `<=>` / `<->` / `<#>`\n * distance operators this strategy emits.\n */\nfunction operatorClass(metric: VectorMetric): string {\n  switch (metric) {\n    case \"cosine\": {\n      return \"vector_cosine_ops\";\n    }\n    case \"l2\": {\n      return \"vector_l2_ops\";\n    }\n    case \"inner_product\": {\n      return \"vector_ip_ops\";\n    }\n    default: {\n      const _exhaustive: never = metric;\n      throw new Error(`Unsupported vector metric: ${String(_exhaustive)}`);\n    }\n  }\n}\n\n/**\n * Formats an embedding as a `::vector` literal. The value is bound as a\n * parameter (not interpolated) — Postgres parses the `[a,b,c]` text and the\n * cast types it. Finiteness is validated first so a NaN names its index.\n */\nfunction vectorLiteral(embedding: readonly number[], name: string): SQL {\n  assertFiniteEmbedding(embedding, name);\n  return sql`${formatVector(embedding)}::vector`;\n}\n\nfunction distanceExpression(\n  embeddingColumn: SQL,\n  queryEmbedding: readonly number[],\n  metric: VectorMetric,\n): SQL {\n  const query = vectorLiteral(queryEmbedding, \"queryEmbedding\");\n  switch (metric) {\n    case \"cosine\": {\n      return sql`(${embeddingColumn} <=> ${query})`;\n    }\n    case \"l2\": {\n      return sql`(${embeddingColumn} <-> ${query})`;\n    }\n    case \"inner_product\": {\n      return sql`(${embeddingColumn} <#> ${query})`;\n    }\n    default: {\n      const _exhaustive: never = metric;\n      throw new Error(`Unsupported vector metric: ${String(_exhaustive)}`);\n    }\n  }\n}\n\nexport const pgvectorStrategy: VectorStrategy = {\n  name: \"pgvector\",\n  capabilities: PGVECTOR_CAPABILITIES,\n\n  tableName(graphId, nodeKind, fieldPath) {\n    return vectorPhysicalName(TABLE_PREFIX, graphId, nodeKind, fieldPath);\n  },\n\n  ownedTables(slot): readonly StrategyTableContribution[] {\n    const table = this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath);\n    const name = quoteIdentifier(table);\n\n    // No standalone graph_id index: the PRIMARY KEY (graph_id, node_id) already\n    // covers `WHERE graph_id = ?` via its leading column, so a separate index\n    // would be pure write amplification.\n    const createDdl = [\n      `CREATE TABLE IF NOT EXISTS ${name} (\n  \"graph_id\" TEXT NOT NULL,\n  \"node_id\" TEXT NOT NULL,\n  \"embedding\" vector(${slot.dimensions}) NOT NULL,\n  \"created_at\" TIMESTAMPTZ NOT NULL,\n  \"updated_at\" TIMESTAMPTZ NOT NULL,\n  PRIMARY KEY (\"graph_id\", \"node_id\")\n);`,\n    ];\n\n    // The HNSW/IVFFlat index is intentionally NOT created here. pgvector\n    // similarity SQL is planner-driven (`ORDER BY embedding <=> q LIMIT k`\n    // uses the index when present, sequential-scans when not), so the index\n    // is a pure materialization concern: `buildCreateIndex` builds it through\n    // `materializeIndexes()` with the field's declared `m`/`ef_construction`/\n    // `lists`. Building it eagerly here would bake in default tuning (the\n    // write-ensure slot has no `indexParams`) and `CREATE INDEX IF NOT EXISTS`\n    // would then mask the tuned index materialization would emit.\n\n    return [\n      {\n        logicalName: `${VECTOR_CONTRIBUTION_PREFIX}:${slot.nodeKind}.${slot.fieldPath}`,\n        owner: \"pgvector\",\n        tableName: table,\n        createDdl,\n        runtimeEnsure: true,\n      },\n    ];\n  },\n\n  buildUpsert(slot, params: UpsertEmbeddingParams, timestamp): readonly SQL[] {\n    const table = quotedTableName(\n      this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n    );\n    const value = vectorLiteral(params.embedding, \"embedding\");\n    return [\n      sql`\n        INSERT INTO ${table} (\"graph_id\", \"node_id\", \"embedding\", \"created_at\", \"updated_at\")\n        VALUES (${params.graphId}, ${params.nodeId}, ${value}, ${timestamp}, ${timestamp})\n        ON CONFLICT (\"graph_id\", \"node_id\")\n        DO UPDATE SET \"embedding\" = EXCLUDED.\"embedding\", \"updated_at\" = EXCLUDED.\"updated_at\"\n      `,\n    ];\n  },\n\n  buildDelete(slot, params: DeleteEmbeddingParams): readonly SQL[] {\n    const table = quotedTableName(\n      this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n    );\n    return [\n      sql`\n        DELETE FROM ${table}\n        WHERE \"graph_id\" = ${params.graphId} AND \"node_id\" = ${params.nodeId}\n      `,\n    ];\n  },\n\n  buildSearch(slot, params: VectorSearchParams): SQL {\n    const table = quotedTableName(\n      this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n    );\n    const embeddingColumn = sql`${table}.\"embedding\"`;\n    const distance = distanceExpression(\n      embeddingColumn,\n      params.queryEmbedding,\n      params.metric,\n    );\n    const score = vectorScoreExpression(distance, params.metric);\n\n    // Same SQL for brute-force and ANN: with a matching HNSW/IVFFlat index the\n    // Postgres planner rewrites this `ORDER BY distance LIMIT k` into an index\n    // scan automatically, so the strategy never branches on `slot.indexType`.\n    const conditions: SQL[] = [sql`${table}.\"graph_id\" = ${params.graphId}`];\n    if (params.minScore !== undefined) {\n      conditions.push(\n        vectorMinScoreCondition(distance, params.metric, params.minScore),\n      );\n    }\n    return sql`\n      SELECT ${table}.\"node_id\" AS node_id, ${score} AS score\n      FROM ${table}\n      WHERE ${sql.join(conditions, sql` AND `)}\n      ORDER BY ${distance} ASC\n      LIMIT ${params.limit}\n    `;\n  },\n\n  distanceExpression(embeddingColumn, queryEmbedding, metric) {\n    return distanceExpression(embeddingColumn, queryEmbedding, metric);\n  },\n\n  buildCreateIndex(slot, options): SQL | undefined {\n    if (!usesAnnIndex(slot)) return undefined;\n    return sql.raw(\n      pgvectorIndexDdl(\n        this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n        slot,\n        options?.concurrent === true,\n      ),\n    );\n  },\n\n  buildDropIndex(slot): SQL | undefined {\n    if (!usesAnnIndex(slot)) return undefined;\n    const indexName = pgvectorIndexName(slot);\n    return sql.raw(`DROP INDEX IF EXISTS ${quoteIdentifier(indexName)}`);\n  },\n\n  buildDropStorage(slot): readonly string[] {\n    const table = quoteIdentifier(\n      this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n    );\n    // CASCADE drops the ANN index along with the table.\n    return [`DROP TABLE IF EXISTS ${table} CASCADE`];\n  },\n};\n\nfunction pgvectorIndexName(slot: VectorSlot): string {\n  return vectorPhysicalName(\n    INDEX_PREFIX,\n    slot.graphId,\n    slot.nodeKind,\n    slot.fieldPath,\n  );\n}\n\nfunction pgvectorIndexDdl(\n  table: string,\n  slot: VectorSlot,\n  concurrent: boolean,\n): string {\n  const indexName = quoteIdentifier(pgvectorIndexName(slot));\n  const quotedTable = quoteIdentifier(table);\n  const opClass = operatorClass(slot.metric);\n  // Honor the field's declared tuning; fall back to pgvector defaults.\n  const m = slot.indexParams?.m ?? DEFAULT_HNSW_M;\n  const efConstruction =\n    slot.indexParams?.efConstruction ?? DEFAULT_HNSW_EF_CONSTRUCTION;\n  const lists = slot.indexParams?.lists ?? DEFAULT_IVFFLAT_LISTS;\n  // CONCURRENTLY builds without taking a write-blocking lock on the live table\n  // (materializeIndexes passes concurrent on Postgres). Run outside a tx.\n  const create = `CREATE INDEX${concurrent ? \" CONCURRENTLY\" : \"\"} IF NOT EXISTS`;\n\n  switch (slot.indexType) {\n    case \"hnsw\": {\n      return `${create} ${indexName} ON ${quotedTable} USING hnsw (\"embedding\" ${opClass}) WITH (m = ${m}, ef_construction = ${efConstruction});`;\n    }\n    case \"ivfflat\": {\n      return `${create} ${indexName} ON ${quotedTable} USING ivfflat (\"embedding\" ${opClass}) WITH (lists = ${lists});`;\n    }\n    case \"none\": {\n      throw new Error(\n        \"pgvectorIndexDdl called for a brute-force-only slot (indexType 'none').\",\n      );\n    }\n    default: {\n      const _exhaustive: never = slot.indexType;\n      throw new Error(`Unsupported vector index type: ${String(_exhaustive)}`);\n    }\n  }\n}\n"]}
+{"version":3,"sources":["../src/backend/drizzle/columns/vector.ts","../src/query/dialect/vector/pgvector-strategy.ts"],"names":["customType","assertFiniteEmbedding","sql","vectorPhysicalName","quoteIdentifier","VECTOR_CONTRIBUTION_PREFIX","quotedTableName","vectorScoreExpression","vectorMinScoreCondition"],"mappings":";;;;;;AA0CsBA,iBAAA,CAInB;AAAA,EACD,SAAS,MAAA,EAAQ;AAEf,IAAA,OAAO,MAAA,EAAQ,UAAA,GAAa,CAAA,OAAA,EAAU,MAAA,CAAO,UAAU,CAAA,CAAA,CAAA,GAAM,QAAA;AAAA,EAC/D,CAAA;AAAA,EAEA,SAAS,KAAA,EAAyB;AAEhC,IAAA,OAAO,CAAA,CAAA,EAAI,KAAA,CAAM,IAAA,CAAK,GAAG,CAAC,CAAA,CAAA,CAAA;AAAA,EAC5B,CAAA;AAAA,EAEA,WAAW,KAAA,EAAyB;AAGlC,IAAA,IAAI,KAAA,CAAM,OAAA,CAAQ,KAAK,CAAA,EAAG;AACxB,MAAA,OAAO,KAAA;AAAA,IACT;AAEA,IAAA,MAAM,OAAA,GAAU,KAAA,CAAM,KAAA,CAAM,CAAA,EAAG,EAAE,CAAA;AACjC,IAAA,IAAI,YAAY,EAAA,EAAI;AAClB,MAAA,OAAO,EAAC;AAAA,IACV;AACA,IAAA,OAAO,OAAA,CAAQ,KAAA,CAAM,GAAG,CAAA,CAAE,GAAA,CAAI,CAAC,CAAA,KAAM,MAAA,CAAO,UAAA,CAAW,CAAA,CAAE,IAAA,EAAM,CAAC,CAAA;AAAA,EAClE;AACF,CAAC;AAgBM,SAAS,aAAa,SAAA,EAAsC;AACjE,EAAA,OAAO,CAAA,CAAA,EAAI,SAAA,CAAU,IAAA,CAAK,GAAG,CAAC,CAAA,CAAA,CAAA;AAChC;;;ACzCA,IAAM,YAAA,GAAe,QAAA;AACrB,IAAM,YAAA,GAAe,WAAA;AASd,IAAM,kBAAA,GAAqB,GAAA;AAM3B,SAAS,uBAAuB,QAAA,EAAyB;AAC9D,EAAA,IAAI,aAAa,MAAA,EAAW;AAC5B,EAAA,IAAI,CAAC,MAAA,CAAO,SAAA,CAAU,QAAQ,CAAA,IAAK,YAAY,CAAA,EAAG;AAChD,IAAA,MAAM,IAAI,KAAA,CAAM,CAAA,0CAAA,EAA6C,QAAQ,CAAA,CAAE,CAAA;AAAA,EACzE;AACA,EAAA,IAAI,WAAW,kBAAA,EAAoB;AACjC,IAAA,MAAM,IAAI,UAAA;AAAA,MACR,CAAA,wBAAA,EAAsB,kBAAkB,CAAA,8CAAA,EACjB,kBAAkB,WAAW,QAAQ,CAAA;AAAA,KAC9D;AAAA,EACF;AACF;AAQA,IAAM,uBAAA,GAA0B,IAAA;AAGhC,IAAM,cAAA,GAAiB,EAAA;AAEvB,IAAM,4BAAA,GAA+B,EAAA;AAErC,IAAM,qBAAA,GAAwB,GAAA;AAE9B,IAAM,qBAAA,GAA4C;AAAA,EAChD,SAAA,EAAW,IAAA;AAAA,EACX,OAAA,EAAS,CAAC,QAAA,EAAU,IAAA,EAAM,eAAe,CAAA;AAAA,EACzC,UAAA,EAAY,CAAC,MAAA,EAAQ,SAAA,EAAW,MAAM,CAAA;AAAA,EACtC,aAAA,EAAe;AACjB,CAAA;AAGA,SAAS,aAAa,IAAA,EAA2B;AAC/C,EAAA,OAAO,IAAA,CAAK,SAAA,KAAc,MAAA,IAAU,IAAA,CAAK,SAAA,KAAc,SAAA;AACzD;AAOA,SAAS,cAAc,MAAA,EAA8B;AACnD,EAAA,QAAQ,MAAA;AAAQ,IACd,KAAK,QAAA,EAAU;AACb,MAAA,OAAO,mBAAA;AAAA,IACT;AAAA,IACA,KAAK,IAAA,EAAM;AACT,MAAA,OAAO,eAAA;AAAA,IACT;AAAA,IACA,KAAK,eAAA,EAAiB;AACpB,MAAA,OAAO,eAAA;AAAA,IACT;AAAA,IACA,SAAS;AACP,MAAA,MAAM,WAAA,GAAqB,MAAA;AAC3B,MAAA,MAAM,IAAI,KAAA,CAAM,CAAA,2BAAA,EAA8B,MAAA,CAAO,WAAW,CAAC,CAAA,CAAE,CAAA;AAAA,IACrE;AAAA;AAEJ;AAOA,SAAS,aAAA,CAAc,WAA8B,IAAA,EAAmB;AACtE,EAAAC,uCAAA,CAAsB,WAAW,IAAI,CAAA;AACrC,EAAA,OAAOC,cAAA,CAAA,EAAM,YAAA,CAAa,SAAS,CAAC,CAAA,QAAA,CAAA;AACtC;AAEA,SAAS,kBAAA,CACP,eAAA,EACA,cAAA,EACA,MAAA,EACK;AACL,EAAA,MAAM,KAAA,GAAQ,aAAA,CAAc,cAAA,EAAgB,gBAAgB,CAAA;AAC5D,EAAA,QAAQ,MAAA;AAAQ,IACd,KAAK,QAAA,EAAU;AACb,MAAA,OAAOA,cAAA,CAAA,CAAA,EAAO,eAAe,CAAA,KAAA,EAAQ,KAAK,CAAA,CAAA,CAAA;AAAA,IAC5C;AAAA,IACA,KAAK,IAAA,EAAM;AACT,MAAA,OAAOA,cAAA,CAAA,CAAA,EAAO,eAAe,CAAA,KAAA,EAAQ,KAAK,CAAA,CAAA,CAAA;AAAA,IAC5C;AAAA,IACA,KAAK,eAAA,EAAiB;AACpB,MAAA,OAAOA,cAAA,CAAA,CAAA,EAAO,eAAe,CAAA,KAAA,EAAQ,KAAK,CAAA,CAAA,CAAA;AAAA,IAC5C;AAAA,IACA,SAAS;AACP,MAAA,MAAM,WAAA,GAAqB,MAAA;AAC3B,MAAA,MAAM,IAAI,KAAA,CAAM,CAAA,2BAAA,EAA8B,MAAA,CAAO,WAAW,CAAC,CAAA,CAAE,CAAA;AAAA,IACrE;AAAA;AAEJ;AAEO,IAAM,gBAAA,GAAmC;AAAA,EAC9C,IAAA,EAAM,UAAA;AAAA,EACN,YAAA,EAAc,qBAAA;AAAA,EAEd,SAAA,CAAU,OAAA,EAAS,QAAA,EAAU,SAAA,EAAW;AACtC,IAAA,OAAOC,oCAAA,CAAmB,YAAA,EAAc,OAAA,EAAS,QAAA,EAAU,SAAS,CAAA;AAAA,EACtE,CAAA;AAAA,EAEA,YAAY,IAAA,EAA4C;AACtD,IAAA,MAAM,KAAA,GAAQ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS,CAAA;AACxE,IAAA,MAAM,IAAA,GAAOC,kCAAgB,KAAK,CAAA;AAKlC,IAAA,MAAM,SAAA,GAAY;AAAA,MAChB,8BAA8B,IAAI,CAAA;AAAA;AAAA;AAAA,qBAAA,EAGjB,KAAK,UAAU,CAAA;AAAA;AAAA;AAAA;AAAA,EAAA;AAAA,KAKlC;AAWA,IAAA,OAAO;AAAA,MACL;AAAA,QACE,WAAA,EAAa,GAAGC,4CAA0B,CAAA,CAAA,EAAI,KAAK,QAAQ,CAAA,CAAA,EAAI,KAAK,SAAS,CAAA,CAAA;AAAA,QAC7E,KAAA,EAAO,UAAA;AAAA,QACP,SAAA,EAAW,KAAA;AAAA,QACX,SAAA;AAAA,QACA,aAAA,EAAe;AAAA;AACjB,KACF;AAAA,EACF,CAAA;AAAA,EAEA,WAAA,CAAY,IAAA,EAAM,MAAA,EAA+B,SAAA,EAA2B;AAC1E,IAAA,MAAM,KAAA,GAAQC,iCAAA;AAAA,MACZ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS;AAAA,KAC5D;AACA,IAAA,MAAM,KAAA,GAAQ,aAAA,CAAc,MAAA,CAAO,SAAA,EAAW,WAAW,CAAA;AACzD,IAAA,OAAO;AAAA,MACLJ,cAAA;AAAA,oBAAA,EACgB,KAAK,CAAA;AAAA,gBAAA,EACT,MAAA,CAAO,OAAO,CAAA,EAAA,EAAK,MAAA,CAAO,MAAM,KAAK,KAAK,CAAA,EAAA,EAAK,SAAS,CAAA,EAAA,EAAK,SAAS,CAAA;AAAA;AAAA;AAAA,MAAA;AAAA,KAIpF;AAAA,EACF,CAAA;AAAA,EAEA,WAAA,CAAY,MAAM,MAAA,EAA+C;AAC/D,IAAA,MAAM,KAAA,GAAQI,iCAAA;AAAA,MACZ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS;AAAA,KAC5D;AACA,IAAA,OAAO;AAAA,MACLJ,cAAA;AAAA,oBAAA,EACgB,KAAK;AAAA,2BAAA,EACE,MAAA,CAAO,OAAO,CAAA,iBAAA,EAAoB,MAAA,CAAO,MAAM;AAAA,MAAA;AAAA,KAExE;AAAA,EACF,CAAA;AAAA,EAEA,WAAA,CAAY,MAAM,MAAA,EAAiC;AACjD,IAAA,MAAM,KAAA,GAAQI,iCAAA;AAAA,MACZ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS;AAAA,KAC5D;AACA,IAAA,MAAM,eAAA,GAAkBJ,iBAAM,KAAK,CAAA,YAAA,CAAA;AACnC,IAAA,MAAM,QAAA,GAAW,kBAAA;AAAA,MACf,eAAA;AAAA,MACA,MAAA,CAAO,cAAA;AAAA,MACP,MAAA,CAAO;AAAA,KACT;AACA,IAAA,MAAM,KAAA,GAAQK,uCAAA,CAAsB,QAAA,EAAU,MAAA,CAAO,MAAM,CAAA;AAK3D,IAAA,MAAM,aAAoB,CAACL,cAAA,CAAA,EAAM,KAAK,CAAA,cAAA,EAAiB,MAAA,CAAO,OAAO,CAAA,CAAE,CAAA;AACvE,IAAA,IAAI,MAAA,CAAO,aAAa,MAAA,EAAW;AACjC,MAAA,UAAA,CAAW,IAAA;AAAA,QACTM,yCAAA,CAAwB,QAAA,EAAU,MAAA,CAAO,MAAA,EAAQ,OAAO,QAAQ;AAAA,OAClE;AAAA,IACF;AACA,IAAA,OAAON,cAAA;AAAA,aAAA,EACI,KAAK,0BAA0B,KAAK,CAAA;AAAA,WAAA,EACtC,KAAK;AAAA,YAAA,EACJA,cAAA,CAAI,IAAA,CAAK,UAAA,EAAYA,cAAA,CAAA,KAAA,CAAU,CAAC;AAAA,eAAA,EAC7B,QAAQ,CAAA;AAAA,YAAA,EACX,OAAO,KAAK;AAAA,IAAA,CAAA;AAAA,EAExB,CAAA;AAAA,EAEA,kBAAA,CAAmB,eAAA,EAAiB,cAAA,EAAgB,MAAA,EAAQ;AAC1D,IAAA,OAAO,kBAAA,CAAmB,eAAA,EAAiB,cAAA,EAAgB,MAAM,CAAA;AAAA,EACnE,CAAA;AAAA,EAEA,gBAAA,CAAiB,MAAM,OAAA,EAA0B;AAC/C,IAAA,IAAI,CAAC,YAAA,CAAa,IAAI,CAAA,EAAG,OAAO,MAAA;AAChC,IAAA,OAAOA,cAAA,CAAI,GAAA;AAAA,MACT,gBAAA;AAAA,QACE,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS,CAAA;AAAA,QAC1D,IAAA;AAAA,QACA,SAAS,UAAA,KAAe;AAAA;AAC1B,KACF;AAAA,EACF,CAAA;AAAA,EAEA,eAAe,IAAA,EAAuB;AACpC,IAAA,IAAI,CAAC,YAAA,CAAa,IAAI,CAAA,EAAG,OAAO,MAAA;AAChC,IAAA,MAAM,SAAA,GAAY,kBAAkB,IAAI,CAAA;AACxC,IAAA,OAAOA,eAAI,GAAA,CAAI,CAAA,qBAAA,EAAwBE,iCAAA,CAAgB,SAAS,CAAC,CAAA,CAAE,CAAA;AAAA,EACrE,CAAA;AAAA,EAEA,iBAAiB,IAAA,EAAyB;AACxC,IAAA,MAAM,KAAA,GAAQA,iCAAA;AAAA,MACZ,KAAK,SAAA,CAAU,IAAA,CAAK,SAAS,IAAA,CAAK,QAAA,EAAU,KAAK,SAAS;AAAA,KAC5D;AAEA,IAAA,OAAO,CAAC,CAAA,qBAAA,EAAwB,KAAK,CAAA,QAAA,CAAU,CAAA;AAAA,EACjD;AACF;AAEA,SAAS,kBAAkB,IAAA,EAA0B;AACnD,EAAA,OAAOD,oCAAA;AAAA,IACL,YAAA;AAAA,IACA,IAAA,CAAK,OAAA;AAAA,IACL,IAAA,CAAK,QAAA;AAAA,IACL,IAAA,CAAK;AAAA,GACP;AACF;AAEA,SAAS,gBAAA,CACP,KAAA,EACA,IAAA,EACA,UAAA,EACQ;AACR,EAAA,MAAM,SAAA,GAAYC,iCAAA,CAAgB,iBAAA,CAAkB,IAAI,CAAC,CAAA;AACzD,EAAA,MAAM,WAAA,GAAcA,kCAAgB,KAAK,CAAA;AACzC,EAAA,MAAM,OAAA,GAAU,aAAA,CAAc,IAAA,CAAK,MAAM,CAAA;AAEzC,EAAA,MAAM,CAAA,GAAI,IAAA,CAAK,WAAA,EAAa,CAAA,IAAK,cAAA;AACjC,EAAA,MAAM,cAAA,GACJ,IAAA,CAAK,WAAA,EAAa,cAAA,IAAkB,4BAAA;AACtC,EAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,WAAA,EAAa,KAAA,IAAS,qBAAA;AAGzC,EAAA,MAAM,MAAA,GAAS,CAAA,YAAA,EAAe,UAAA,GAAa,eAAA,GAAkB,EAAE,CAAA,cAAA,CAAA;AAE/D,EAAA,QAAQ,KAAK,SAAA;AAAW,IACtB,KAAK,MAAA,EAAQ;AACX,MAAA,OAAO,CAAA,EAAG,MAAM,CAAA,CAAA,EAAI,SAAS,CAAA,IAAA,EAAO,WAAW,CAAA,yBAAA,EAA4B,OAAO,CAAA,YAAA,EAAe,CAAC,CAAA,oBAAA,EAAuB,cAAc,CAAA,EAAA,CAAA;AAAA,IACzI;AAAA,IACA,KAAK,SAAA,EAAW;AACd,MAAA,OAAO,CAAA,EAAG,MAAM,CAAA,CAAA,EAAI,SAAS,OAAO,WAAW,CAAA,4BAAA,EAA+B,OAAO,CAAA,gBAAA,EAAmB,KAAK,CAAA,EAAA,CAAA;AAAA,IAC/G;AAAA,IACA,KAAK,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,KAAA;AAAA,QACR;AAAA,OACF;AAAA,IACF;AAAA,IACA,SAAS;AACP,MAAA,MAAM,cAAqB,IAAA,CAAK,SAAA;AAChC,MAAA,MAAM,IAAI,KAAA,CAAM,CAAA,+BAAA,EAAkC,MAAA,CAAO,WAAW,CAAC,CAAA,CAAE,CAAA;AAAA,IACzE;AAAA;AAEJ","file":"chunk-YW7GHMF4.cjs","sourcesContent":["/**\n * Custom Drizzle column type for pgvector's VECTOR type.\n *\n * Provides native support for PostgreSQL vector embeddings without\n * requiring runtime casts in queries.\n *\n * @example\n * ```typescript\n * import { pgTable, text, integer } from \"drizzle-orm/pg-core\";\n * import { vector } from \"./columns/vector\";\n *\n * const embeddings = pgTable(\"embeddings\", {\n *   id: text(\"id\").primaryKey(),\n *   embedding: vector(\"embedding\"),\n *   dimensions: integer(\"dimensions\"),\n * });\n * ```\n */\nimport { customType } from \"drizzle-orm/pg-core\";\n\n/**\n * PostgreSQL vector column type for pgvector extension.\n *\n * Stores embeddings as native VECTOR type, enabling efficient\n * similarity search without runtime casts.\n *\n * The column type is unparameterized (`VECTOR` not `VECTOR(N)`)\n * to support multiple embedding dimensions in a single table.\n * Dimension validation is handled at the application level.\n *\n * @example\n * ```typescript\n * // In table definition\n * embedding: vector(\"embedding\").notNull(),\n *\n * // In queries - no cast needed\n * db.execute(sql`\n *   SELECT * FROM embeddings\n *   WHERE embedding <=> ${formatVector([0.1, 0.2, 0.3])} < 0.5\n * `);\n * ```\n */\nexport const vector = customType<{\n  data: number[];\n  driverData: string;\n  config: { dimensions?: number };\n}>({\n  dataType(config) {\n    // Use parameterized type if dimensions specified, otherwise unparameterized\n    return config?.dimensions ? `vector(${config.dimensions})` : \"vector\";\n  },\n\n  toDriver(value: number[]): string {\n    // Format as pgvector literal: [1.0,2.0,3.0]\n    return `[${value.join(\",\")}]`;\n  },\n\n  fromDriver(value: string): number[] {\n    // pgvector returns '[1,2,3]' format\n    // Handle both string and already-parsed array (some drivers differ)\n    if (Array.isArray(value)) {\n      return value as number[];\n    }\n    // Parse the string representation\n    const content = value.slice(1, -1); // Remove [ and ]\n    if (content === \"\") {\n      return [];\n    }\n    return content.split(\",\").map((s) => Number.parseFloat(s.trim()));\n  },\n});\n\n/**\n * Formats a number array as a pgvector literal string.\n *\n * Use this when building raw SQL queries with embeddings.\n *\n * @example\n * ```typescript\n * const embedding = [0.1, 0.2, 0.3];\n * db.execute(sql`\n *   SELECT * FROM embeddings\n *   WHERE embedding <=> ${formatVector(embedding)}::vector < 0.5\n * `);\n * ```\n */\nexport function formatVector(embedding: readonly number[]): string {\n  return `[${embedding.join(\",\")}]`;\n}\n","/**\n * pgvector strategy.\n *\n * Postgres' pgvector extension is the one engine where the *same* SQL serves\n * brute-force and ANN: `ORDER BY embedding <=> q LIMIT k` scans sequentially\n * with no index and is rewritten to an HNSW/IVFFlat scan the moment a matching\n * index exists — the planner picks it up automatically. So `buildSearch` never\n * branches on `slot.indexType`; it emits one relevance scan and lets Postgres\n * decide. (Contrast libSQL/sqlite-vec, whose ANN needs distinct syntax.)\n *\n * Storage is per-`(nodeKind, fieldPath)`: a `vector(N)` column carrying the\n * field's fixed dimension `N`, with the operator-class-typed ANN index folded\n * into the same table contribution so materialization builds table + index\n * together. The typed column means the index does not need the legacy\n * `::vector(N)` cast-in-index escape hatch the old shared table relied on.\n *\n * Metric support is the full pgvector set — `cosine`, `l2`, `inner_product` —\n * advertised as data on `capabilities`, mirroring `vector_cosine_ops` /\n * `vector_l2_ops` / `vector_ip_ops` and the `<=>` / `<->` / `<#>` operators.\n *\n * Mirrors the `libsqlVectorStrategy` / `sqliteVecStrategy` structure while\n * letting Postgres' planner pick the ANN index when one exists.\n */\nimport { type SQL, sql } from \"drizzle-orm\";\n\nimport { formatVector } from \"../../../backend/drizzle/columns/vector\";\nimport { quotedTableName } from \"../../../backend/drizzle/operations/shared\";\nimport { type StrategyTableContribution } from \"../../../backend/table-contribution\";\nimport {\n  type DeleteEmbeddingParams,\n  type UpsertEmbeddingParams,\n  type VectorCapabilities,\n  type VectorMetric,\n  type VectorSearchParams,\n} from \"../../../backend/types\";\nimport {\n  assertFiniteEmbedding,\n  quoteIdentifier,\n  VECTOR_CONTRIBUTION_PREFIX,\n  vectorMinScoreCondition,\n  vectorPhysicalName,\n  vectorScoreExpression,\n  type VectorSlot,\n  type VectorStrategy,\n} from \"../vector-strategy\";\n\n/** Physical-name prefixes for the per-field table and its ANN index. */\nconst TABLE_PREFIX = \"tg_vec\";\nconst INDEX_PREFIX = \"tg_vecidx\";\n\n/**\n * pgvector defines `hnsw.ef_search` as an integer GUC with a valid range\n * of 1..1000; `SET hnsw.ef_search = 1001` errors at the server. The\n * backend validates a per-search `efSearch` against this ceiling before\n * inlining it into `SET LOCAL`, surfacing a clear message instead of a raw\n * pgvector error.\n */\nexport const MAX_HNSW_EF_SEARCH = 1000;\n\n/**\n * Validates a per-search `efSearch` override against pgvector's\n * `hnsw.ef_search` valid range. `undefined` (no override) is accepted.\n */\nexport function assertPgvectorEfSearch(efSearch?: number): void {\n  if (efSearch === undefined) return;\n  if (!Number.isInteger(efSearch) || efSearch <= 0) {\n    throw new Error(`efSearch must be a positive integer, got: ${efSearch}`);\n  }\n  if (efSearch > MAX_HNSW_EF_SEARCH) {\n    throw new RangeError(\n      `efSearch must be ≤ ${MAX_HNSW_EF_SEARCH} (pgvector's hnsw.ef_search ` +\n        `valid range is 1..${MAX_HNSW_EF_SEARCH}), got: ${efSearch}`,\n    );\n  }\n}\n\n/**\n * pgvector's `vector` type caps at 16,000 dimensions for a stored column\n * (2,000 for an indexed column, but the column-type ceiling is what governs\n * storage; recall degradation beyond 2,000 is a tuning concern, not a hard\n * limit advertised here).\n */\nconst PGVECTOR_MAX_DIMENSIONS = 16_000;\n\n/** Default HNSW `m` (max connections per layer) — pgvector's own default. */\nconst DEFAULT_HNSW_M = 16;\n/** Default HNSW `ef_construction` (build candidate list) — pgvector's own default. */\nconst DEFAULT_HNSW_EF_CONSTRUCTION = 64;\n/** Default IVFFlat `lists` (inverted lists) — pgvector's own default. */\nconst DEFAULT_IVFFLAT_LISTS = 100;\n\nconst PGVECTOR_CAPABILITIES: VectorCapabilities = {\n  supported: true,\n  metrics: [\"cosine\", \"l2\", \"inner_product\"],\n  indexTypes: [\"hnsw\", \"ivfflat\", \"none\"],\n  maxDimensions: PGVECTOR_MAX_DIMENSIONS,\n};\n\n/** Whether a slot's declared index type materializes a real pgvector ANN index. */\nfunction usesAnnIndex(slot: VectorSlot): boolean {\n  return slot.indexType === \"hnsw\" || slot.indexType === \"ivfflat\";\n}\n\n/**\n * pgvector operator class for a metric's ANN index — `vector_cosine_ops` /\n * `vector_l2_ops` / `vector_ip_ops`, mirroring the `<=>` / `<->` / `<#>`\n * distance operators this strategy emits.\n */\nfunction operatorClass(metric: VectorMetric): string {\n  switch (metric) {\n    case \"cosine\": {\n      return \"vector_cosine_ops\";\n    }\n    case \"l2\": {\n      return \"vector_l2_ops\";\n    }\n    case \"inner_product\": {\n      return \"vector_ip_ops\";\n    }\n    default: {\n      const _exhaustive: never = metric;\n      throw new Error(`Unsupported vector metric: ${String(_exhaustive)}`);\n    }\n  }\n}\n\n/**\n * Formats an embedding as a `::vector` literal. The value is bound as a\n * parameter (not interpolated) — Postgres parses the `[a,b,c]` text and the\n * cast types it. Finiteness is validated first so a NaN names its index.\n */\nfunction vectorLiteral(embedding: readonly number[], name: string): SQL {\n  assertFiniteEmbedding(embedding, name);\n  return sql`${formatVector(embedding)}::vector`;\n}\n\nfunction distanceExpression(\n  embeddingColumn: SQL,\n  queryEmbedding: readonly number[],\n  metric: VectorMetric,\n): SQL {\n  const query = vectorLiteral(queryEmbedding, \"queryEmbedding\");\n  switch (metric) {\n    case \"cosine\": {\n      return sql`(${embeddingColumn} <=> ${query})`;\n    }\n    case \"l2\": {\n      return sql`(${embeddingColumn} <-> ${query})`;\n    }\n    case \"inner_product\": {\n      return sql`(${embeddingColumn} <#> ${query})`;\n    }\n    default: {\n      const _exhaustive: never = metric;\n      throw new Error(`Unsupported vector metric: ${String(_exhaustive)}`);\n    }\n  }\n}\n\nexport const pgvectorStrategy: VectorStrategy = {\n  name: \"pgvector\",\n  capabilities: PGVECTOR_CAPABILITIES,\n\n  tableName(graphId, nodeKind, fieldPath) {\n    return vectorPhysicalName(TABLE_PREFIX, graphId, nodeKind, fieldPath);\n  },\n\n  ownedTables(slot): readonly StrategyTableContribution[] {\n    const table = this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath);\n    const name = quoteIdentifier(table);\n\n    // No standalone graph_id index: the PRIMARY KEY (graph_id, node_id) already\n    // covers `WHERE graph_id = ?` via its leading column, so a separate index\n    // would be pure write amplification.\n    const createDdl = [\n      `CREATE TABLE IF NOT EXISTS ${name} (\n  \"graph_id\" TEXT NOT NULL,\n  \"node_id\" TEXT NOT NULL,\n  \"embedding\" vector(${slot.dimensions}) NOT NULL,\n  \"created_at\" TIMESTAMPTZ NOT NULL,\n  \"updated_at\" TIMESTAMPTZ NOT NULL,\n  PRIMARY KEY (\"graph_id\", \"node_id\")\n);`,\n    ];\n\n    // The HNSW/IVFFlat index is intentionally NOT created here. pgvector\n    // similarity SQL is planner-driven (`ORDER BY embedding <=> q LIMIT k`\n    // uses the index when present, sequential-scans when not), so the index\n    // is a pure materialization concern: `buildCreateIndex` builds it through\n    // `materializeIndexes()` with the field's declared `m`/`ef_construction`/\n    // `lists`. Building it eagerly here would bake in default tuning (the\n    // write-ensure slot has no `indexParams`) and `CREATE INDEX IF NOT EXISTS`\n    // would then mask the tuned index materialization would emit.\n\n    return [\n      {\n        logicalName: `${VECTOR_CONTRIBUTION_PREFIX}:${slot.nodeKind}.${slot.fieldPath}`,\n        owner: \"pgvector\",\n        tableName: table,\n        createDdl,\n        runtimeEnsure: true,\n      },\n    ];\n  },\n\n  buildUpsert(slot, params: UpsertEmbeddingParams, timestamp): readonly SQL[] {\n    const table = quotedTableName(\n      this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n    );\n    const value = vectorLiteral(params.embedding, \"embedding\");\n    return [\n      sql`\n        INSERT INTO ${table} (\"graph_id\", \"node_id\", \"embedding\", \"created_at\", \"updated_at\")\n        VALUES (${params.graphId}, ${params.nodeId}, ${value}, ${timestamp}, ${timestamp})\n        ON CONFLICT (\"graph_id\", \"node_id\")\n        DO UPDATE SET \"embedding\" = EXCLUDED.\"embedding\", \"updated_at\" = EXCLUDED.\"updated_at\"\n      `,\n    ];\n  },\n\n  buildDelete(slot, params: DeleteEmbeddingParams): readonly SQL[] {\n    const table = quotedTableName(\n      this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n    );\n    return [\n      sql`\n        DELETE FROM ${table}\n        WHERE \"graph_id\" = ${params.graphId} AND \"node_id\" = ${params.nodeId}\n      `,\n    ];\n  },\n\n  buildSearch(slot, params: VectorSearchParams): SQL {\n    const table = quotedTableName(\n      this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n    );\n    const embeddingColumn = sql`${table}.\"embedding\"`;\n    const distance = distanceExpression(\n      embeddingColumn,\n      params.queryEmbedding,\n      params.metric,\n    );\n    const score = vectorScoreExpression(distance, params.metric);\n\n    // Same SQL for brute-force and ANN: with a matching HNSW/IVFFlat index the\n    // Postgres planner rewrites this `ORDER BY distance LIMIT k` into an index\n    // scan automatically, so the strategy never branches on `slot.indexType`.\n    const conditions: SQL[] = [sql`${table}.\"graph_id\" = ${params.graphId}`];\n    if (params.minScore !== undefined) {\n      conditions.push(\n        vectorMinScoreCondition(distance, params.metric, params.minScore),\n      );\n    }\n    return sql`\n      SELECT ${table}.\"node_id\" AS node_id, ${score} AS score\n      FROM ${table}\n      WHERE ${sql.join(conditions, sql` AND `)}\n      ORDER BY ${distance} ASC\n      LIMIT ${params.limit}\n    `;\n  },\n\n  distanceExpression(embeddingColumn, queryEmbedding, metric) {\n    return distanceExpression(embeddingColumn, queryEmbedding, metric);\n  },\n\n  buildCreateIndex(slot, options): SQL | undefined {\n    if (!usesAnnIndex(slot)) return undefined;\n    return sql.raw(\n      pgvectorIndexDdl(\n        this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n        slot,\n        options?.concurrent === true,\n      ),\n    );\n  },\n\n  buildDropIndex(slot): SQL | undefined {\n    if (!usesAnnIndex(slot)) return undefined;\n    const indexName = pgvectorIndexName(slot);\n    return sql.raw(`DROP INDEX IF EXISTS ${quoteIdentifier(indexName)}`);\n  },\n\n  buildDropStorage(slot): readonly string[] {\n    const table = quoteIdentifier(\n      this.tableName(slot.graphId, slot.nodeKind, slot.fieldPath),\n    );\n    // CASCADE drops the ANN index along with the table.\n    return [`DROP TABLE IF EXISTS ${table} CASCADE`];\n  },\n};\n\nfunction pgvectorIndexName(slot: VectorSlot): string {\n  return vectorPhysicalName(\n    INDEX_PREFIX,\n    slot.graphId,\n    slot.nodeKind,\n    slot.fieldPath,\n  );\n}\n\nfunction pgvectorIndexDdl(\n  table: string,\n  slot: VectorSlot,\n  concurrent: boolean,\n): string {\n  const indexName = quoteIdentifier(pgvectorIndexName(slot));\n  const quotedTable = quoteIdentifier(table);\n  const opClass = operatorClass(slot.metric);\n  // Honor the field's declared tuning; fall back to pgvector defaults.\n  const m = slot.indexParams?.m ?? DEFAULT_HNSW_M;\n  const efConstruction =\n    slot.indexParams?.efConstruction ?? DEFAULT_HNSW_EF_CONSTRUCTION;\n  const lists = slot.indexParams?.lists ?? DEFAULT_IVFFLAT_LISTS;\n  // CONCURRENTLY builds without taking a write-blocking lock on the live table\n  // (materializeIndexes passes concurrent on Postgres). Run outside a tx.\n  const create = `CREATE INDEX${concurrent ? \" CONCURRENTLY\" : \"\"} IF NOT EXISTS`;\n\n  switch (slot.indexType) {\n    case \"hnsw\": {\n      return `${create} ${indexName} ON ${quotedTable} USING hnsw (\"embedding\" ${opClass}) WITH (m = ${m}, ef_construction = ${efConstruction});`;\n    }\n    case \"ivfflat\": {\n      return `${create} ${indexName} ON ${quotedTable} USING ivfflat (\"embedding\" ${opClass}) WITH (lists = ${lists});`;\n    }\n    case \"none\": {\n      throw new Error(\n        \"pgvectorIndexDdl called for a brute-force-only slot (indexType 'none').\",\n      );\n    }\n    default: {\n      const _exhaustive: never = slot.indexType;\n      throw new Error(`Unsupported vector index type: ${String(_exhaustive)}`);\n    }\n  }\n}\n"]}

package/dist/{ddl-BJcPrB4L.d.ts → ddl-C8e8zDAE.d.ts}

@@ -1,6 +1,6 @@
-import { F as FulltextStrategy } from './types-BKQljXom.js';
-import { P as PostgresTables } from './postgres-BdIb7tS_.js';
-import { S as SqliteTables } from './sqlite-aM_PAz-O.js';
+import { F as FulltextStrategy } from './types-Dk8r09KY.js';
+import { P as PostgresTables } from './postgres-Dhom7iiX.js';
+import { S as SqliteTables } from './sqlite-BfLsCEqO.js';
 
 /**
  * Generates all DDL statements for the given SQLite tables.

package/dist/{ddl-CPZ7mB7X.d.cts → ddl-CwyG4HoQ.d.cts}

@@ -1,6 +1,6 @@
-import { F as FulltextStrategy } from './types-C1CYxWrL.cjs';
-import { P as PostgresTables } from './postgres-4J2tmauh.cjs';
-import { S as SqliteTables } from './sqlite-Gatk_iw_.cjs';
+import { F as FulltextStrategy } from './types-Dvmn-eal.cjs';
+import { P as PostgresTables } from './postgres-jgXmD45C.cjs';
+import { S as SqliteTables } from './sqlite-DcQ_BCUW.cjs';
 
 /**
  * Generates all DDL statements for the given SQLite tables.

package/dist/graph-merge/index.cjs

@@ -1,18 +1,18 @@
 'use strict';
 
-var chunkRZ4DH4E7_cjs = require('../chunk-RZ4DH4E7.cjs');
+var chunk2LAT5BIK_cjs = require('../chunk-2LAT5BIK.cjs');
 require('../chunk-HTGDWIDE.cjs');
-var chunk3JBJEYWE_cjs = require('../chunk-3JBJEYWE.cjs');
-var chunk5DTPBFX5_cjs = require('../chunk-5DTPBFX5.cjs');
-var chunk7JGLPAFV_cjs = require('../chunk-7JGLPAFV.cjs');
-require('../chunk-6SA7UIQJ.cjs');
+var chunkSB65GJNY_cjs = require('../chunk-SB65GJNY.cjs');
+var chunkLAZGHOZW_cjs = require('../chunk-LAZGHOZW.cjs');
+var chunkEEYV6UOE_cjs = require('../chunk-EEYV6UOE.cjs');
+require('../chunk-3QRBXYT5.cjs');
 require('../chunk-6RMLO5O4.cjs');
 require('../chunk-YQGLVWB2.cjs');
 require('../chunk-ZFWNO3GE.cjs');
 var chunkKTIEB7VH_cjs = require('../chunk-KTIEB7VH.cjs');
 require('../chunk-6G35I3VW.cjs');
 var chunkRP7YITHX_cjs = require('../chunk-RP7YITHX.cjs');
-require('../chunk-V4IQYOQ4.cjs');
+require('../chunk-CFNT3ITG.cjs');
 var chunk73KASJGO_cjs = require('../chunk-73KASJGO.cjs');
 var chunkJIEJCTDS_cjs = require('../chunk-JIEJCTDS.cjs');
 require('../chunk-FLRCVI2G.cjs');
@@ -228,8 +228,8 @@ function byId(left, right) {
 }
 async function diffAgainstBase(baseStore, forkStore) {
   const graph = baseStore.graph;
-  const nodeKinds = chunk7JGLPAFV_cjs.getNodeKinds(graph);
-  const edgeKinds = chunk7JGLPAFV_cjs.getEdgeKinds(graph);
+  const nodeKinds = chunkEEYV6UOE_cjs.getNodeKinds(graph);
+  const edgeKinds = chunkEEYV6UOE_cjs.getEdgeKinds(graph);
   const newNodes = [];
   const modifiedNodes = [];
   const deletedNodes = [];
@@ -311,19 +311,16 @@ async function readActiveSchemaVersion(backend, graphId) {
 }
 async function computeSchemaComponent(store) {
   const version = await readActiveSchemaVersion(store.backend, store.graphId);
-  return chunk5DTPBFX5_cjs.computeSchemaHash(chunk5DTPBFX5_cjs.serializeSchema(store.graph, version));
+  return chunkLAZGHOZW_cjs.computeSchemaHash(chunkLAZGHOZW_cjs.serializeSchema(store.graph, version));
 }
 function byDigestEntry(left, right) {
   const byKind = compareStrings(left.kind, right.kind);
   return byKind === 0 ? compareStrings(left.id, right.id) : byKind;
 }
 var CONTENT_FINGERPRINT_BYTES = 16;
-async function computeContentFingerprint(store) {
-  const backend = store.backend;
-  const graphId = store.graphId;
-  const graph = store.graph;
-  const nodeKinds = chunk7JGLPAFV_cjs.getNodeKinds(graph);
-  const edgeKinds = chunk7JGLPAFV_cjs.getEdgeKinds(graph);
+async function computeContentComponent(backend, graphId, graph) {
+  const nodeKinds = chunkEEYV6UOE_cjs.getNodeKinds(graph);
+  const edgeKinds = chunkEEYV6UOE_cjs.getEdgeKinds(graph);
   const nodeDigest = [];
   for (const kind of nodeKinds) {
     const rows = await enumerateAllNodes(backend, graphId, kind);
@@ -369,12 +366,16 @@ async function computeContentFingerprint(store) {
 async function computeBaseVersion(store) {
   const [schemaComponent, contentComponent] = await Promise.all([
     computeSchemaComponent(store),
-    computeContentFingerprint(store)
+    computeContentComponent(store.backend, store.graphId, store.graph)
   ]);
   return asBaseVersion(
     `${schemaComponent}${TOKEN_SEPARATOR}${contentComponent}`
   );
 }
+function contentComponentOf(version) {
+  const separatorIndex = version.indexOf(TOKEN_SEPARATOR);
+  return separatorIndex === -1 ? version : version.slice(separatorIndex + 1);
+}
 
 // src/graph-merge/errors.ts
 var MERGE_ERROR_CODES = {
@@ -445,17 +446,17 @@ var CLONE_IMPORT_BATCH_SIZE = 1e3;
 function cloneWorkingCopyStrategy(makeBackend) {
   return {
     create: async (baseStore) => {
-      const data = await chunk3JBJEYWE_cjs.exportGraph(baseStore, {
+      const data = await chunkSB65GJNY_cjs.exportGraph(baseStore, {
         includeMeta: true,
         includeDeleted: false
       });
       const backend = await makeBackend();
       try {
-        const [freshStore] = await chunkRZ4DH4E7_cjs.createStoreWithSchema(
+        const [freshStore] = await chunk2LAT5BIK_cjs.createStoreWithSchema(
           baseStore.graph,
           backend
         );
-        const result = await chunk3JBJEYWE_cjs.importGraph(freshStore, data, {
+        const result = await chunkSB65GJNY_cjs.importGraph(freshStore, data, {
           onConflict: "error",
           onUnknownProperty: "error",
           validateReferences: true,
@@ -483,7 +484,7 @@ function cloneWorkingCopyStrategy(makeBackend) {
 async function branch(baseStore, makeBackend, options, strategy) {
   try {
     const base = await computeBaseVersion(baseStore);
-    const id = options?.id ?? asBranchId(chunkRZ4DH4E7_cjs.generateId());
+    const id = options?.id ?? asBranchId(chunk2LAT5BIK_cjs.generateId());
     const workingCopyStrategy = strategy ?? cloneWorkingCopyStrategy(makeBackend);
     const store = await workingCopyStrategy.create(baseStore);
     return chunkKTIEB7VH_cjs.ok({ id, base, store });
@@ -512,7 +513,7 @@ function constraintSignature(node, constraint) {
       return void 0;
     }
   }
-  return `${constraint.name}${KEY_PART_SEPARATOR}${chunkRZ4DH4E7_cjs.computeUniqueKey(
+  return `${constraint.name}${KEY_PART_SEPARATOR}${chunk2LAT5BIK_cjs.computeUniqueKey(
     props,
     constraint.fields,
     constraint.collation
@@ -2055,7 +2056,7 @@ function normalizeMergeOptions(options = {}) {
     ...provenanceWeights === void 0 ? {} : { provenanceWeights }
   };
 }
-var Provenance = chunk5DTPBFX5_cjs.defineNode("Provenance", {
+var Provenance = chunkLAZGHOZW_cjs.defineNode("Provenance", {
   schema: zod.z.object({
     targetGraphId: zod.z.string(),
     role: zod.z.enum(["node", "edge"]),
@@ -2069,14 +2070,14 @@ function provenanceGraphId(targetGraphId) {
   return `${targetGraphId}::merge-provenance`;
 }
 function buildProvenanceGraph(targetGraphId) {
-  return chunk7JGLPAFV_cjs.defineGraph({
+  return chunkEEYV6UOE_cjs.defineGraph({
     id: provenanceGraphId(targetGraphId),
     nodes: { Provenance: { type: Provenance } },
     edges: {}
   });
 }
 async function openProvenanceStore(backend, targetGraphId) {
-  const [store] = await chunkRZ4DH4E7_cjs.createStoreWithSchema(
+  const [store] = await chunk2LAT5BIK_cjs.createStoreWithSchema(
     buildProvenanceGraph(targetGraphId),
     backend
   );
@@ -2507,6 +2508,48 @@ async function stageBranches(baseStore, branches) {
   };
 }
 
+// src/graph-merge/tx-retry.ts
+var RETRYABLE_SQLSTATES = /* @__PURE__ */ new Set(["40001", "40P01"]);
+var RETRYABLE_MESSAGE_PATTERN = /could not serialize access|deadlock detected/i;
+var MAX_COMMIT_ATTEMPTS = 3;
+function isRetryableTxConflict(error) {
+  const seen = /* @__PURE__ */ new Set();
+  let current = error;
+  while (current !== void 0 && current !== null && typeof current === "object" && !seen.has(current)) {
+    seen.add(current);
+    const candidate = current;
+    if (typeof candidate.code === "string" && RETRYABLE_SQLSTATES.has(candidate.code)) {
+      return true;
+    }
+    if (typeof candidate.message === "string" && RETRYABLE_MESSAGE_PATTERN.test(candidate.message)) {
+      return true;
+    }
+    current = candidate.cause;
+  }
+  return false;
+}
+async function withTxConflictRetry(commit) {
+  let lastConflict;
+  for (let attempt = 1; attempt <= MAX_COMMIT_ATTEMPTS; attempt += 1) {
+    try {
+      return await commit();
+    } catch (error) {
+      if (!isRetryableTxConflict(error)) {
+        throw error;
+      }
+      lastConflict = error;
+    }
+  }
+  throw new MergeError(
+    `Merge commit aborted by transaction conflicts (serialization failure or deadlock) on ${MAX_COMMIT_ATTEMPTS} consecutive attempts; giving up.`,
+    {
+      cause: lastConflict,
+      details: { attempts: MAX_COMMIT_ATTEMPTS },
+      suggestion: "Reduce concurrent writes to the merge target, or serialize merges against it."
+    }
+  );
+}
+
 // src/graph-merge/type-reconcile.ts
 var INCOMPATIBLE_TYPES_FLAG_REASON = "type-reconcile:incompatible-kinds";
 function distinctKinds(kinds) {
@@ -3226,20 +3269,48 @@ async function applyMergePlan(plan, nodesApi, edgesApi) {
   }
   return { nodes: committedNodeIds.size, edges: committedEdges };
 }
-async function commitPlan(target, plan) {
+async function commitPlan(target, plan, expectedBaseVersion) {
   if (!target.backend.capabilities.transactions) {
     throw new MergeError(
       "merge() requires a transaction-capable target backend. The merged plan (canonical upserts, soft-deletes, edge upserts) must commit atomically; a non-transactional fallback would leave a partially-merged graph on a mid-commit failure. (mergeIncremental() enforces the same requirement.)",
       { details: { capability: "transactions" } }
     );
   }
-  return target.transaction(
-    (tx) => applyMergePlan(
-      plan,
-      tx.nodes,
-      tx.edges
-    )
+  return withTxConflictRetry(
+    () => target.transaction(async (tx) => {
+      if (expectedBaseVersion !== void 0) {
+        await assertTargetUnchanged(tx.backend, target, expectedBaseVersion);
+      }
+      return applyMergePlan(
+        plan,
+        tx.nodes,
+        tx.edges
+      );
+    }, MERGE_COMMIT_TX_OPTIONS)
+  );
+}
+var MERGE_COMMIT_TX_OPTIONS = {
+  isolationLevel: "serializable"
+};
+async function assertTargetUnchanged(txBackend, target, expectedBaseVersion) {
+  const liveContent = await computeContentComponent(
+    txBackend,
+    target.graphId,
+    target.graph
   );
+  const expectedContent = contentComponentOf(expectedBaseVersion);
+  if (liveContent !== expectedContent) {
+    throw new BaseVersionMismatchError(
+      "The merge target was modified between the base@V check and the commit transaction; the resolved plan no longer describes the live target and was not applied.",
+      {
+        details: {
+          expectedContentFingerprint: expectedContent,
+          liveContentFingerprint: liveContent
+        },
+        suggestion: "Re-run the merge (and re-branch if the divergence is real), or serialize writers against merges on this target."
+      }
+    );
+  }
 }
 function nodeCollection(nodes, kind) {
   const collection = nodes[kind];
@@ -3277,7 +3348,7 @@ async function validateBaseVersions(target, branches) {
       );
     }
   }
-  return chunkKTIEB7VH_cjs.ok();
+  return chunkKTIEB7VH_cjs.ok(targetVersion);
 }
 function tryNormalize(optionsInput) {
   try {
@@ -3286,7 +3357,7 @@ function tryNormalize(optionsInput) {
     return chunkKTIEB7VH_cjs.err(new MergeError("Invalid merge options.", { cause: error }));
   }
 }
-async function resolveMerge(store, target, branches, options, useBaseSources, incremental) {
+async function resolveMerge(store, target, branches, options, useBaseSources, incremental, expectedBaseVersion) {
   let targetBranchSeen = false;
   for (const branch2 of branches) {
     if (branch2.id === BASE_PROVENANCE_BRANCH) {
@@ -3357,7 +3428,16 @@ async function resolveMerge(store, target, branches, options, useBaseSources, in
       subClassClosure,
       preferredBranchId
     );
-    const merged = incremental === void 0 ? await commitPlan(target, plan) : await commitIncrementalPlan(target, plan);
+    const merged = incremental === void 0 ? await commitPlan(target, plan, expectedBaseVersion) : await commitIncrementalPlan(target, plan, {
+      stagedNewByKind: newNodesByKind(staging),
+      options,
+      introspectionKinds,
+      // The committed (kind, id) keys the base sources matched at PLAN time;
+      // anything NEW the in-tx re-probe surfaces is a window write.
+      plannedBaseMatchKeys: new Set(
+        candidates.data.baseMembers.map((member) => mergeKeyOf(member))
+      )
+    });
     const provenance = options.provenance ? buildProvenanceIndex(plan.provenanceRecords) : { byBranch: () => ({ nodeIds: [], edgeIds: [] }) };
     const warnings = [...plan.warnings];
     let provenancePersisted;
@@ -3418,7 +3498,15 @@ async function merge(store, branches, optionsInput = {}) {
   if (chunkKTIEB7VH_cjs.isErr(precondition)) {
     return chunkKTIEB7VH_cjs.err(precondition.error);
   }
-  return resolveMerge(store, target, branches, options, false);
+  return resolveMerge(
+    store,
+    target,
+    branches,
+    options,
+    false,
+    void 0,
+    precondition.data
+  );
 }
 async function validateForkPointVersions(forkPoint, branches) {
   const forkVersion = await computeBaseVersion(forkPoint);
@@ -3651,13 +3739,12 @@ async function validateIncrementalEdgeWrites(target, edgesApi, plan) {
   }
   const notFound = [...plannedKindById.keys()].filter((id) => !existingById.has(id)).sort((left, right) => compareStrings(left, right));
   if (notFound.length > 0) {
-    const perKind = await Promise.all(
-      Object.keys(target.graph.edges).map(
-        (kind) => edgeCollection(edgesApi, kind).getByIds(notFound, INCLUDE_TOMBSTONES)
-      )
-    );
     const crossKindById = /* @__PURE__ */ new Map();
-    for (const rows of perKind) {
+    for (const kind of Object.keys(target.graph.edges)) {
+      const rows = await edgeCollection(edgesApi, kind).getByIds(
+        notFound,
+        INCLUDE_TOMBSTONES
+      );
       for (const [index, id] of notFound.entries()) {
         const row = rows[index];
         if (row !== void 0 && !crossKindById.has(id)) {
@@ -3725,20 +3812,77 @@ async function validateIncrementalEdgeWrites(target, edgesApi, plan) {
     }
   }
 }
-async function commitIncrementalPlan(target, plan) {
+async function collectBaseMatchKeys(lookupStore, guard) {
+  const matched = /* @__PURE__ */ new Set();
+  for (const [kind, stagedNodes] of guard.stagedNewByKind) {
+    const resolveConfig = guard.options.resolve[kind];
+    if (resolveConfig === void 0) {
+      continue;
+    }
+    const collection = lookupStore.nodes[kind];
+    if (collection === void 0 || stagedNodes.length === 0) {
+      continue;
+    }
+    const items = stagedNodes.map((staged) => ({ props: staged.node.props }));
+    for (const constraint of uniqueConstraintsFor(
+      guard.introspectionKinds,
+      kind
+    )) {
+      const matches = await collection.bulkFindByConstraint(
+        constraint.name,
+        items
+      );
+      for (const base of matches) {
+        if (base !== void 0) {
+          matched.add(mergeKeyOf(base));
+        }
+      }
+    }
+    if (resolveConfig.blockIndex !== void 0) {
+      const matchesByItem = await collection.bulkFindByIndex(
+        resolveConfig.blockIndex,
+        items
+      );
+      for (const perItem of matchesByItem) {
+        for (const base of perItem) {
+          matched.add(mergeKeyOf(base));
+        }
+      }
+    }
+  }
+  return matched;
+}
+async function assertBaseResolutionStable(lookupStore, guard) {
+  const liveKeys = await collectBaseMatchKeys(lookupStore, guard);
+  const appeared = [...liveKeys].filter((key) => !guard.plannedBaseMatchKeys.has(key)).sort((left, right) => compareMergeKeys(left, right));
+  if (appeared.length === 0) {
+    return;
+  }
+  throw new BaseVersionMismatchError(
+    `mergeIncremental() resolved its branch additions against the target as of planning, but ${appeared.length} committed row(s) matching a branch addition's identity key (a unique constraint or block index) were inserted before the commit transaction. Committing the plan would create duplicate entities the base-source resolution would otherwise have collapsed.`,
+    {
+      details: { appeared },
+      suggestion: "Re-run mergeIncremental(); the re-planned merge resolves the branch additions against the now-committed rows."
+    }
+  );
+}
+async function commitIncrementalPlan(target, plan, guard) {
   if (!target.backend.capabilities.transactions) {
     throw new MergeError(
       "mergeIncremental() requires a transaction-capable target backend. Incremental writes must preflight existing rows and commit atomically; non-transactional fallback would allow partial graph writes.",
       { details: { capability: "transactions" } }
     );
   }
-  return target.transaction(async (tx) => {
-    const nodesApi = tx.nodes;
-    const edgesApi = tx.edges;
-    await validateIncrementalNodeWrites(target, nodesApi, plan);
-    await validateIncrementalEdgeWrites(target, edgesApi, plan);
-    return applyMergePlan(plan, nodesApi, edgesApi);
-  });
+  return withTxConflictRetry(
+    () => target.transaction(async (tx) => {
+      const nodesApi = tx.nodes;
+      const edgesApi = tx.edges;
+      await assertBaseResolutionStable(tx, guard);
+      await validateIncrementalNodeWrites(target, nodesApi, plan);
+      await validateIncrementalEdgeWrites(target, edgesApi, plan);
+      return applyMergePlan(plan, nodesApi, edgesApi);
+    }, MERGE_COMMIT_TX_OPTIONS)
+  );
 }
 async function mergeIncremental(args) {
   const { forkPoint, target, branches } = args;