@delali/narsil 0.3.1-next.72 → 0.3.1-next.73

package/README.md

@@ -106,7 +106,7 @@ Every hit holds the document, its id, and its BM25 score. Beside the hits, `resu
 
 **Geosearch.** [Geo filters](https://github.com/assetcorp/narsil/blob/main/docs/geosearch.md) match documents by radius, using Haversine or Vincenty distance, or by polygon containment. You can combine a geo filter with every other query feature.
 
-**Storage.** [Persistence adapters](https://github.com/assetcorp/narsil/blob/main/docs/persistence-and-durability.md) store indexes on the filesystem, in IndexedDB, in memory, or in a backend of your own. With durability on, the engine keeps a write-ahead log with periodic checkpoints and recovers on its own after a crash. A snapshot captures a whole index as one portable byte array.
+**Storage.** [Persistence adapters](https://github.com/assetcorp/narsil/blob/main/docs/persistence-and-durability.md) store indexes on the filesystem, in IndexedDB, in memory, or in a backend of your own. With durability on, the engine keeps a write-ahead log with periodic checkpoints and recovers on its own after a crash. You can stream a whole index into a portable snapshot archive, or export its documents as JSON Lines to move the index across releases.
 
 **Scale.** The engine routes each document to a [partition](https://github.com/assetcorp/narsil/blob/main/docs/partitions-and-workers.md) by a deterministic hash. When you call `rebalance()`, the engine reshapes the partitions online and buffers incoming writes in a write-ahead queue. Once an index holds 1,000 documents, the engine serves keyword queries from worker copies on half its worker threads, while the HTTP server receives requests on the threads that hold those copies.
 
@@ -124,7 +124,7 @@ Every hit holds the document, its id, and its BM25 score. Beside the hits, `resu
 | [Hybrid search](https://github.com/assetcorp/narsil/blob/main/docs/hybrid-search.md) | Reciprocal rank fusion and linear blending of text and vector rankings |
 | [Geosearch](https://github.com/assetcorp/narsil/blob/main/docs/geosearch.md) | Radius and polygon filters, and the two distance formulas |
 | [Embedding adapters](https://github.com/assetcorp/narsil/blob/main/docs/embedding-adapters.md) | Automatic embedding on insert and query, named adapters, the bundled ones, and custom ones |
-| [Persistence and durability](https://github.com/assetcorp/narsil/blob/main/docs/persistence-and-durability.md) | Storage backends, the write-ahead log, checkpoints, recovery, and snapshots |
+| [Persistence and durability](https://github.com/assetcorp/narsil/blob/main/docs/persistence-and-durability.md) | Storage backends, the write-ahead log, checkpoints, recovery, snapshots, document exports, and indexes from an earlier release |
 | [Partitions and workers](https://github.com/assetcorp/narsil/blob/main/docs/partitions-and-workers.md) | Partition routing, online rebalancing, worker copies, and multi-instance invalidation |
 | [Language support](https://github.com/assetcorp/narsil/blob/main/docs/language-support.md) | The 107 language modules, analysis revisions and rebuilds, and named tokenizers and stop words |
 | [HTTP server](https://github.com/assetcorp/narsil/blob/main/docs/http-server.md) | Wrapping an engine in a REST API, every route it serves, and long-running tasks |

package/dist/adapters/filesystem.mjs

@@ -1,4 +1,4 @@
-import { NarsilError, ErrorCodes } from '../chunk-RV77FR2F.mjs';
+import { NarsilError, ErrorCodes } from '../chunk-I64LXKO3.mjs';
 
 // src/persistence/filesystem.ts
 var cachedFs = null;

package/dist/adapters/indexeddb.mjs

@@ -1,4 +1,4 @@
-import { NarsilError, ErrorCodes } from '../chunk-RV77FR2F.mjs';
+import { NarsilError, ErrorCodes } from '../chunk-I64LXKO3.mjs';
 
 // src/persistence/constants.ts
 var DEFAULT_INDEXEDDB_NAME = "narsil";

package/dist/{capabilities-D6gjjQyA.d.ts → capabilities-CYeAWYxQ.d.ts}

@@ -1,6 +1,6 @@
 import { E as EmbeddingAdapter } from './adapters-79OIa4xm.js';
-import { C as ClusterNamespace } from './config-BLguB-om.js';
-import { h as TaskStore } from './tasks-BECd0b__.js';
+import { C as ClusterNamespace } from './config-Bv8etwUZ.js';
+import { h as TaskStore } from './tasks-DpgCpW1x.js';
 
 /**
  * Context handed to {@link OnRequestHook} for every inbound request, captured
@@ -70,8 +70,15 @@ interface CorsOptions {
 interface ServerLimits {
     /** Cap for JSON request bodies (single doc, batch, search). Excess → 413. */
     maxBodyBytes?: number;
-    /** Cap for the NDJSON import stream and binary restore body. Excess → 413. */
+    /** Cap for the NDJSON import stream. Excess → 413. */
     maxImportBytes?: number;
+    /** The server accepts a snapshot archive for `POST /indexes/{name}/restore`
+     * and a document export for `POST /indexes/{name}/import` of up to this many
+     * bytes, and it answers a larger upload with 413. The server writes each
+     * upload to a temporary file as the upload arrives, so it needs the same
+     * memory for an upload of any size. The default is 64 GiB, and 0 accepts an
+     * upload of any size. */
+    maxRestoreBytes?: number;
     /** Cap for a single NDJSON line, so one unterminated line cannot exhaust memory. Excess → 413. */
     maxLineBytes?: number;
     /** Documents handed to the engine per batch during NDJSON import; the loop yields between batches. */
@@ -255,6 +262,14 @@ declare const REBUILD_ANALYSIS_CAPABILITY = "indexes.rebuildAnalysis";
  * @public
  */
 declare const INDEX_LIFECYCLE_CAPABILITY = "indexes.lifecycle";
+/**
+ * Marks a server that answers `GET /indexes/{name}/export` with a stream of a
+ * document export and imports an export sent to `POST /indexes/{name}/import`
+ * in a task.
+ *
+ * @public
+ */
+declare const DOCUMENT_EXPORT_CAPABILITY = "indexes.documentExport";
 /**
  * Every capability this server announces at `/capabilities`.
  *
@@ -266,4 +281,4 @@ declare const INDEX_LIFECYCLE_CAPABILITY = "indexes.lifecycle";
  */
 declare const SERVER_CAPABILITIES: readonly string[];
 
-export { ASYNC_IMPORT_CAPABILITY as A, type CorsOptions as C, type ErrorEnvelope as E, INDEX_LIFECYCLE_CAPABILITY as I, type NarsilServer as N, type OnRequestHook as O, REBUILD_ANALYSIS_CAPABILITY as R, SERVER_CAPABILITIES as S, TASK_CANCEL_CAPABILITY as T, TASK_FILTER_CAPABILITY as a, type ServerOptions as b, type RequestContext as c, type RequestDenial as d, type ServerLimits as e };
+export { ASYNC_IMPORT_CAPABILITY as A, type CorsOptions as C, DOCUMENT_EXPORT_CAPABILITY as D, type ErrorEnvelope as E, INDEX_LIFECYCLE_CAPABILITY as I, type NarsilServer as N, type OnRequestHook as O, REBUILD_ANALYSIS_CAPABILITY as R, SERVER_CAPABILITIES as S, TASK_CANCEL_CAPABILITY as T, TASK_FILTER_CAPABILITY as a, type ServerOptions as b, type RequestContext as c, type RequestDenial as d, type ServerLimits as e };

package/dist/{chunk-Y52WZWAU.mjs → chunk-2WLD2NFW.mjs}

@@ -1,8 +1,8 @@
-import { readSharedConfigFields, restoredEmbedding, restoredPartitionConfig, MAX_PARTITION_COUNT, MAX_REPLICATION_FACTOR } from './chunk-CVPA5Q7R.mjs';
-import { MAX_INDEX_NAME_LENGTH, INDEX_NAME_PATTERN } from './chunk-XZ4PAREY.mjs';
-import { getLanguage } from './chunk-KZTI25D3.mjs';
+import { readSharedConfigFields, restoredEmbedding, restoredPartitionConfig, MAX_PARTITION_COUNT, MAX_REPLICATION_FACTOR } from './chunk-AFKH4ZK3.mjs';
+import { getLanguage } from './chunk-4MQ5F4IL.mjs';
+import { MAX_INDEX_NAME_LENGTH, INDEX_NAME_PATTERN } from './chunk-AK3BLGCI.mjs';
 import { compareCodePoints } from './chunk-LJYTW7GT.mjs';
-import { NarsilError, ErrorCodes } from './chunk-RV77FR2F.mjs';
+import { NarsilError, ErrorCodes } from './chunk-I64LXKO3.mjs';
 import { encode, decode } from '@msgpack/msgpack';
 
 // src/distribution/cluster/index-settings.ts

package/dist/{chunk-QR4OKRVY.mjs → chunk-3YKWSO54.mjs}

@@ -1,7 +1,7 @@
-import { isWordIndexedBase, flattenSchema, MAX_LATITUDE_DEGREES, MAX_LONGITUDE_DEGREES, MIN_POLYGON_POINTS, parsedTypeOf, isStringValuedBase, withSortableOption, isGeopointOrVectorType, MAX_SORT_FIELDS, MAX_SORT_FIELD_NAME_LENGTH, MAX_CURSOR_LENGTH, MAX_CURSOR_ANCHOR_CODE_POINTS, RESULT_WINDOW, DEFAULT_FUZZY_PREFIX_LENGTH, EARTH_RADIUS_METERS, RADIUS_BOX_MARGIN_RATIO, RADIUS_BOX_MARGIN_RADIANS, FULL_TURN_DEGREES, ANTIMERIDIAN_SPAN_DEGREES, DEFAULT_PAGE_SIZE, PROTOTYPE_POLLUTION_KEYS, DEG_TO_RAD, LOCATION_CODE_BITS_PER_AXIS, isLatitude, isLongitude, isWordIndexedFieldType, PREFIX_MAX_EXPANSIONS, MAX_COVERING_CELLS } from './chunk-M6XF5HDE.mjs';
+import { isWordIndexedBase, flattenSchema, MAX_LATITUDE_DEGREES, MAX_LONGITUDE_DEGREES, MIN_POLYGON_POINTS, parsedTypeOf, isStringValuedBase, withSortableOption, isGeopointOrVectorType, MAX_SORT_FIELDS, MAX_SORT_FIELD_NAME_LENGTH, MAX_CURSOR_LENGTH, MAX_CURSOR_ANCHOR_CODE_POINTS, RESULT_WINDOW, DEFAULT_FUZZY_PREFIX_LENGTH, EARTH_RADIUS_METERS, RADIUS_BOX_MARGIN_RATIO, RADIUS_BOX_MARGIN_RADIANS, FULL_TURN_DEGREES, ANTIMERIDIAN_SPAN_DEGREES, DEFAULT_PAGE_SIZE, PROTOTYPE_POLLUTION_KEYS, DEG_TO_RAD, LOCATION_CODE_BITS_PER_AXIS, isLatitude, isLongitude, isWordIndexedFieldType, PREFIX_MAX_EXPANSIONS, MAX_COVERING_CELLS } from './chunk-AK3BLGCI.mjs';
 import { DEFAULT_PATTERN_WORK_CAP, PATTERN_TEST_OPERATORS, bitsetIsEmpty, MAX_PATTERN_TEXT_CODE_POINTS, MAX_PATTERN_POSITIONS, createBitSet, bitsetSet, PATTERN_WORK_FLUSH_UNITS, MATCHER_STATE_BUDGET_BYTES } from './chunk-ZKRV5M2P.mjs';
 import { compareCodePoints, sortModeOf, defaultSortMode, SORT_VALUE_MAX_CODE_POINTS, toReducedSortValue, readSortField, compareComparableKeys, toComparableSortValue, exceedsCodePoints, holdsUnpairedSurrogate, isSortMode, SORT_MODES, compareSortValues, foldedCodePointCount } from './chunk-LJYTW7GT.mjs';
-import { createNarsilError, ErrorCodes, NarsilError } from './chunk-RV77FR2F.mjs';
+import { createNarsilError, ErrorCodes, NarsilError } from './chunk-I64LXKO3.mjs';
 
 // src/core/tokenizer/constants.ts
 var TOKEN_CACHE_SIZE_FLOOR = 5e4;
@@ -777,6 +777,7 @@ function getOrCreateFieldNameIndex(table, fieldName) {
   table.indexMap.set(fieldName, idx);
   return idx;
 }
+var VECTOR_PATTERN = /^vector\[(\d+)]$/;
 function getNestedValue(obj, path) {
   const parts = path.split(".");
   let current = obj;
@@ -809,6 +810,10 @@ function textFieldsChanged(oldDoc, newDoc, fields) {
   }
   return false;
 }
+function parseVectorDimension(fieldType) {
+  const match = VECTOR_PATTERN.exec(fieldType);
+  return match ? Number.parseInt(match[1], 10) : null;
+}
 function tokenizeOptions(options) {
   return {
     stem: true,
@@ -2573,4 +2578,4 @@ function setNestedValue(obj, path, value) {
   current[segments[segments.length - 1]] = value;
 }
 
-export { GEO_UNIT_METRES, MAX_FILTER_ARRAY_SIZE, MAX_FILTER_DEPTH, MAX_FILTER_TESTS, applyPagination, applyProjection, applySorting, boundedLevenshtein, clampRowCount, clearNormalizationCache, cloneProjected, collectQueryTermStats, computeFacets, configureNormalizationCache, countsWithoutScores, coveringRanges, createPatternWorkMeter, decodePageCursor, encodePageCursor, everyValueFacetConfig, fanOutMatchCount, fanOutQuery, filterTestCount, fnv1a, fnv1aBytes, fnv1aUint32s, fulltextMatches, fuzzyPrefixOf, fuzzyTermMatches, getFieldValueByInternalId, getFlatSchema, getNestedValue, getNormalizationCacheSize, getOrCreateFieldNameIndex, joinPatternWorkMeter, kWayMerge, locationCode, mergeFacets, mergePartitionStats, normalizeSort, oversampledFacetConfig, oversampledShardSize, partitionsIn, polygonBoxes, polygonContainment, projectionKeepsField, pruneStatsToQueryTerms, radiusBoxes, readSortValues, requireMatchingCursor, requireSortableFields, requireValidFilter, requireValidOperands, requireWithinResultWindow, resetNormalizationCache, resolveProjection, sanitizeGlobalStats, setNestedValue, sortModesOf, sortSignatureEntry, sortSignatureOf, textFieldsChanged, tokenize, tokenizeIterator, tokenizeOptions, undeclaredFieldOnStrictIndex, undeclaredFilterField, visitTokens };
+export { GEO_UNIT_METRES, MAX_FILTER_ARRAY_SIZE, MAX_FILTER_DEPTH, MAX_FILTER_TESTS, applyPagination, applyProjection, applySorting, boundedLevenshtein, clampRowCount, clearNormalizationCache, cloneProjected, collectQueryTermStats, computeFacets, configureNormalizationCache, countsWithoutScores, coveringRanges, createPatternWorkMeter, decodePageCursor, encodePageCursor, everyValueFacetConfig, fanOutMatchCount, fanOutQuery, filterTestCount, fnv1a, fnv1aBytes, fnv1aUint32s, fulltextMatches, fuzzyPrefixOf, fuzzyTermMatches, getFieldValueByInternalId, getFlatSchema, getNestedValue, getNormalizationCacheSize, getOrCreateFieldNameIndex, joinPatternWorkMeter, kWayMerge, locationCode, mergeFacets, mergePartitionStats, normalizeSort, oversampledFacetConfig, oversampledShardSize, parseVectorDimension, partitionsIn, polygonBoxes, polygonContainment, projectionKeepsField, pruneStatsToQueryTerms, radiusBoxes, readSortValues, requireMatchingCursor, requireSortableFields, requireValidFilter, requireValidOperands, requireWithinResultWindow, resetNormalizationCache, resolveProjection, sanitizeGlobalStats, setNestedValue, sortModesOf, sortSignatureEntry, sortSignatureOf, textFieldsChanged, tokenize, tokenizeIterator, tokenizeOptions, undeclaredFieldOnStrictIndex, undeclaredFilterField, visitTokens };

package/dist/{chunk-KZTI25D3.mjs → chunk-4MQ5F4IL.mjs}

@@ -1,5 +1,5 @@
 import { english } from './chunk-TGP5M5R7.mjs';
-import { NarsilError, ErrorCodes } from './chunk-RV77FR2F.mjs';
+import { NarsilError, ErrorCodes } from './chunk-I64LXKO3.mjs';
 
 // src/languages/registry.ts
 var languages = /* @__PURE__ */ new Map();

package/dist/{chunk-CVPA5Q7R.mjs → chunk-AFKH4ZK3.mjs}

@@ -1,5 +1,5 @@
-import { validateRequiredFields } from './chunk-M6XF5HDE.mjs';
-import { ErrorCodes, NarsilError } from './chunk-RV77FR2F.mjs';
+import { requireDocumentWithinSizeLimit, validateRequiredFields } from './chunk-AK3BLGCI.mjs';
+import { ErrorCodes, NarsilError } from './chunk-I64LXKO3.mjs';
 
 // src/narsil/internals.ts
 var engineCores = /* @__PURE__ */ new WeakMap();
@@ -244,7 +244,10 @@ function providedDocId(document) {
   return typeof id === "string" && id.length > 0 ? id : void 0;
 }
 function requireDocumentObject(document) {
-  if (document !== null && typeof document === "object" && !Array.isArray(document)) return;
+  if (document !== null && typeof document === "object" && !Array.isArray(document)) {
+    requireDocumentWithinSizeLimit(document);
+    return;
+  }
   const received = Array.isArray(document) ? "a list" : document === null ? "null" : `a ${typeof document}`;
   throw new NarsilError(
     ErrorCodes.DOC_VALIDATION_FAILED,
@@ -578,9 +581,9 @@ function readSharedConfigFields(source, reject) {
     ...typeof source.patternValueLimit === "number" ? { patternValueLimit: source.patternValueLimit } : {}
   };
 }
-function restoredConfigFields(envelope) {
-  const shared = readSharedConfigFields(envelope, rejectSnapshot);
-  const partitions = restoredPartitionConfig(envelope.partitionConfig);
+function restoredConfigFields(envelope, reject = rejectSnapshot) {
+  const shared = readSharedConfigFields(envelope, reject);
+  const partitions = restoredPartitionConfig(envelope.partitionConfig, reject);
   return {
     ...shared,
     ...partitions !== void 0 ? { partitions } : {}

package/dist/{chunk-M6XF5HDE.mjs → chunk-AK3BLGCI.mjs}

@@ -1,6 +1,7 @@
-import { MAX_M } from './chunk-P2ZZN32C.mjs';
+import { EARLIER_RELEASE_MIN_M, MIN_M, MAX_M } from './chunk-PU43WDY4.mjs';
 import { holdsUnpairedSurrogate, exceedsCodePoints } from './chunk-LJYTW7GT.mjs';
-import { NarsilError, ErrorCodes } from './chunk-RV77FR2F.mjs';
+import { NarsilError, ErrorCodes } from './chunk-I64LXKO3.mjs';
+import { encode } from '@msgpack/msgpack';
 
 // src/schema/constants.ts
 var MAX_NESTING_DEPTH = 4;
@@ -8,6 +9,10 @@ var MAX_DOCUMENT_NESTING_DEPTH = 32;
 var DEFAULT_PATTERN_VALUE_LIMIT = 8192;
 var MIN_PATTERN_VALUE_LIMIT = 1;
 var MAX_PATTERN_VALUE_LIMIT = 65536;
+var MAX_VECTOR_DIMENSION = 4096;
+var MAX_STORED_VECTOR_DIMENSION = 65535;
+var MAX_TEXT_FIELDS = 65536;
+var MAX_DOCUMENT_BYTES = 16777216;
 
 // src/schema/validator/shared.ts
 var VECTOR_PATTERN = /^vector\[(\d+)]$/;
@@ -57,11 +62,11 @@ function parseBase(baseName, path, type) {
     invalidType(path, type, `Field "${path}" has an unsupported type: "${type}"`);
   }
   const dimension = Number.parseInt(vectorMatch[1], 10);
-  if (dimension <= 0) {
+  if (dimension <= 0 || dimension > MAX_STORED_VECTOR_DIMENSION) {
     throw new NarsilError(
       ErrorCodes.SCHEMA_INVALID_VECTOR_DIMENSION,
-      `Vector field "${path}" must have a positive dimension, got ${dimension}`,
-      { field: path, dimension }
+      `Vector field "${path}" has ${dimension} dimensions, but a vector field holds from 1 to ${MAX_STORED_VECTOR_DIMENSION}`,
+      { field: path, dimension, maximum: MAX_STORED_VECTOR_DIMENSION }
     );
   }
   return { base: "vector", dimension };
@@ -194,9 +199,30 @@ function validateSchema(schema) {
   if (Object.keys(schema).length === 0) {
     throw new NarsilError(ErrorCodes.SCHEMA_INVALID_TYPE, "Schema must define at least one field");
   }
-  return validateSchemaFields(schema, 1, "");
+  const validated = validateSchemaFields(schema, 1, "");
+  requireTextFieldCount(schemaFieldsOf(validated).filter((field) => isWordIndexedBase(field.base)).length);
+  return validated;
+}
+function requireTextFieldCount(count) {
+  if (count <= MAX_TEXT_FIELDS) return;
+  throw new NarsilError(
+    ErrorCodes.SCHEMA_INVALID_TYPE,
+    `The schema declares ${count} string and string[] fields, but a schema holds at most ${MAX_TEXT_FIELDS}`,
+    { textFields: count, maximum: MAX_TEXT_FIELDS }
+  );
+}
+function requireCreatableSchema(schema) {
+  for (const field of schemaFieldsOf(schema)) {
+    if (field.base !== "vector" || field.dimension <= MAX_VECTOR_DIMENSION) continue;
+    throw new NarsilError(
+      ErrorCodes.SCHEMA_INVALID_VECTOR_DIMENSION,
+      `Vector field "${field.path}" has ${field.dimension} dimensions, but a new index takes at most ${MAX_VECTOR_DIMENSION}`,
+      { field: field.path, dimension: field.dimension, maximum: MAX_VECTOR_DIMENSION }
+    );
+  }
 }
 function requireValidFieldTypes(flatSchema) {
+  let textFields = 0;
   for (const [path, type] of Object.entries(flatSchema)) {
     if (typeof type !== "string") {
       throw new NarsilError(
@@ -205,8 +231,9 @@ function requireValidFieldTypes(flatSchema) {
         { field: path, type: String(type) }
       );
     }
-    parseFieldType(type, path);
+    if (isWordIndexedBase(parseFieldType(type, path).base)) textFields += 1;
   }
+  requireTextFieldCount(textFields);
 }
 function flattenRecursive(schema, prefix, result) {
   for (const [field, type] of Object.entries(schema)) {
@@ -474,7 +501,11 @@ function walkStorable(value, path, depth, ancestors, skip) {
     reject(path, `holds a ${kind}, which cannot be stored or recovered`);
   }
   const container = value;
-  if (container instanceof Date || container instanceof Uint8Array) return;
+  if (container instanceof Date) {
+    if (Number.isNaN(container.getTime())) reject(path, "holds an invalid date, which the engine cannot store");
+    return;
+  }
+  if (container instanceof Uint8Array) return;
   const isArray = Array.isArray(container);
   if (!isArray && !isPlainObject2(container)) {
     reject(
@@ -628,6 +659,71 @@ function validateRequiredFields(document, required) {
     }
   }
 }
+function holdsNegativeZero(value) {
+  const pending = [value];
+  while (pending.length > 0) {
+    const next = pending.pop();
+    if (typeof next === "number") {
+      if (Object.is(next, -0)) return true;
+      continue;
+    }
+    if (next === null || typeof next !== "object" || ArrayBuffer.isView(next)) continue;
+    for (const child of Array.isArray(next) ? next : Object.values(next)) pending.push(child);
+  }
+  return false;
+}
+function encodeDocument(document) {
+  return encode(document, holdsNegativeZero(document) ? { forceIntegerToFloat: true } : void 0);
+}
+
+// src/schema/validator/document-size.ts
+var LARGEST_HEADER_BYTES = 5;
+var LARGEST_NUMBER_BYTES = 9;
+var LARGEST_TIMESTAMP_BYTES = 15;
+var LARGEST_UTF8_BYTES_PER_UNIT = 3;
+function upperBoundBytes(value, budget, depth) {
+  if (value === null || value === void 0 || typeof value === "boolean") return 1;
+  if (typeof value === "number") return LARGEST_NUMBER_BYTES;
+  if (typeof value === "string") return LARGEST_HEADER_BYTES + value.length * LARGEST_UTF8_BYTES_PER_UNIT;
+  if (value instanceof Date) return LARGEST_TIMESTAMP_BYTES;
+  if (ArrayBuffer.isView(value)) return LARGEST_HEADER_BYTES + value.byteLength;
+  if (typeof value !== "object" || depth > MAX_DOCUMENT_NESTING_DEPTH) return LARGEST_NUMBER_BYTES;
+  let total = LARGEST_HEADER_BYTES;
+  if (Array.isArray(value)) {
+    for (const item of value) {
+      total += upperBoundBytes(item, budget - total, depth + 1);
+      if (total > budget) return total;
+    }
+    return total;
+  }
+  const record = value;
+  for (const key of Object.keys(record)) {
+    total += LARGEST_HEADER_BYTES + key.length * LARGEST_UTF8_BYTES_PER_UNIT;
+    total += upperBoundBytes(record[key], budget - total, depth + 1);
+    if (total > budget) return total;
+  }
+  return total;
+}
+function messagePackUpperBoundBytes(value) {
+  return upperBoundBytes(value, Number.POSITIVE_INFINITY, 0);
+}
+function encodedBytesOf(document) {
+  try {
+    return encodeDocument(document).byteLength;
+  } catch {
+    return null;
+  }
+}
+function requireDocumentWithinSizeLimit(document) {
+  if (upperBoundBytes(document, MAX_DOCUMENT_BYTES, 0) <= MAX_DOCUMENT_BYTES) return;
+  const encodedBytes = encodedBytesOf(document);
+  if (encodedBytes === null || encodedBytes <= MAX_DOCUMENT_BYTES) return;
+  throw new NarsilError(
+    ErrorCodes.DOC_VALIDATION_FAILED,
+    `The document takes ${encodedBytes} bytes as MessagePack, vector values included, but a document may take at most ${MAX_DOCUMENT_BYTES}`,
+    { bytes: encodedBytes, maximum: MAX_DOCUMENT_BYTES }
+  );
+}
 
 // src/schema/validator/vector-promotion.ts
 var QUANTIZATION_MODES = /* @__PURE__ */ new Set(["osq8", "osq4", "osq2", "osq1", "none"]);
@@ -651,6 +747,13 @@ function validateVectorStorage(config, filesystemDurability) {
     });
   }
 }
+function raiseEarlierReleaseGraphSetting(config) {
+  if (typeof config !== "object" || config === null) return config;
+  const hnsw = config.hnswConfig;
+  if (typeof hnsw !== "object" || hnsw === null) return config;
+  if (hnsw.m !== EARLIER_RELEASE_MIN_M) return config;
+  return { ...config, hnswConfig: { ...hnsw, m: MIN_M } };
+}
 function validateVectorPromotion(config) {
   if (config === void 0) return;
   if (config.threshold !== void 0 && !isPositiveInteger(config.threshold)) {
@@ -666,8 +769,8 @@ function validateVectorPromotion(config) {
   }
   const hnsw = config.hnswConfig;
   if (hnsw === void 0) return;
-  if (hnsw.m !== void 0 && (!isPositiveInteger(hnsw.m) || hnsw.m > MAX_M)) {
-    fail(`vectorPromotion.hnswConfig.m must be a positive integer no greater than ${MAX_M}`, { m: hnsw.m });
+  if (hnsw.m !== void 0 && (!Number.isInteger(hnsw.m) || hnsw.m < MIN_M || hnsw.m > MAX_M)) {
+    fail(`vectorPromotion.hnswConfig.m must be a whole number from ${MIN_M} to ${MAX_M}`, { m: hnsw.m });
   }
   if (hnsw.efConstruction !== void 0 && !isPositiveInteger(hnsw.efConstruction)) {
     fail("vectorPromotion.hnswConfig.efConstruction must be a positive integer", {
@@ -698,4 +801,31 @@ var DEFAULT_LINEAR_ALPHA = 0.5;
 var MIN_LINEAR_ALPHA = 0;
 var MAX_LINEAR_ALPHA = 1;
 
-export { ANTIMERIDIAN_SPAN_DEGREES, DEFAULT_FUZZY_PREFIX_LENGTH, DEFAULT_LINEAR_ALPHA, DEFAULT_MAX_PER_GROUP, DEFAULT_PAGE_SIZE, DEFAULT_RANK_CONSTANT, DEG_TO_RAD, EARTH_RADIUS_METERS, FULL_TURN_DEGREES, LOCATION_CODE_BITS_PER_AXIS, MAX_COVERING_CELLS, MAX_CURSOR_ANCHOR_CODE_POINTS, MAX_CURSOR_LENGTH, MAX_LATITUDE_DEGREES, MAX_LINEAR_ALPHA, MAX_LONGITUDE_DEGREES, MAX_NESTING_DEPTH, MAX_PATTERN_VALUE_LIMIT, MAX_PREFIX_LENGTH, MAX_SORT_FIELDS, MAX_SORT_FIELD_NAME_LENGTH, MAX_TOLERANCE, MIN_LINEAR_ALPHA, MIN_PATTERN_VALUE_LIMIT, MIN_POLYGON_POINTS, MIN_RANK_CONSTANT, PREFIX_MAX_EXPANSIONS, PROTOTYPE_POLLUTION_KEYS, RADIUS_BOX_MARGIN_RADIANS, RADIUS_BOX_MARGIN_RATIO, RESULT_WINDOW, VECTOR_PATTERN, VINCENTY_CONVERGENCE_THRESHOLD, VINCENTY_MAX_ITERATIONS, extractVectorFieldsFromSchema, flattenSchema, isGeopointOrVectorType, isLatitude, isListBase, isLongitude, isQuantizationMode, isSingleStringFieldType, isStorageMode, isStringValuedBase, isWordIndexedBase, isWordIndexedFieldType, parsedTypeOf, requireValidFieldTypes, schemaFieldsOf, validateDocument, validateDocumentStrict, validateRequiredFields, validateSchema, validateVectorPromotion, validateVectorStorage, withSortableOption };
+// src/engine/constants.ts
+var MIN_DOCUMENTS_FOR_SEGMENTS = 64;
+var BATCH_CHUNK_SIZE = 1e3;
+var DEFAULT_LIMIT = DEFAULT_PAGE_SIZE;
+var DEFAULT_OFFSET = 0;
+var INDEX_NAME_PATTERN = /^[a-zA-Z0-9][a-zA-Z0-9._-]*$/;
+var ABSOLUTE_WINDOWS_PATH_PATTERN = /^[a-zA-Z]:[\\/]/;
+var MODULE_URL_SCHEME_PATTERN = /^(file|https?|data|node):/;
+var MAX_INDEX_NAME_LENGTH = 255;
+var MAX_DOC_ID_LENGTH = 512;
+var EXPORT_CHUNK_BYTES = 1048576;
+var MAX_EXPORT_LINE_BYTES = 117443592;
+var MAX_INDEX_CONFIGURATION_BYTES = 16777216;
+var IMPORT_BATCH_DOCUMENTS = 1e3;
+var MAX_IMPORT_REFUSALS_REPORTED = 1e3;
+var MIN_SWEEP_INTERVAL_MS = 100;
+var MAX_SWEEP_INTERVAL_MS = 6e4;
+var HEAP_PRESSURE_WARNING_FRACTION = 0.9;
+var HEAP_PRESSURE_REARM_FRACTION = 0.8;
+var STATISTICS_BROADCAST_INTERVAL_MS = 5e3;
+var FOREIGN_STATS_TTL_MS = 6e4;
+var DEFAULT_SUGGEST_LIMIT = 10;
+var MAX_SUGGEST_LIMIT = 100;
+var SUGGEST_OVERSAMPLE_FACTOR = 1.5;
+var SUGGEST_OVERSAMPLE_PADDING = 10;
+var MAX_SUGGEST_SCATTER_LIMIT = Math.ceil(MAX_SUGGEST_LIMIT * SUGGEST_OVERSAMPLE_FACTOR) + SUGGEST_OVERSAMPLE_PADDING;
+
+export { ABSOLUTE_WINDOWS_PATH_PATTERN, ANTIMERIDIAN_SPAN_DEGREES, BATCH_CHUNK_SIZE, DEFAULT_FUZZY_PREFIX_LENGTH, DEFAULT_LIMIT, DEFAULT_LINEAR_ALPHA, DEFAULT_MAX_PER_GROUP, DEFAULT_OFFSET, DEFAULT_PAGE_SIZE, DEFAULT_RANK_CONSTANT, DEFAULT_SUGGEST_LIMIT, DEG_TO_RAD, EARTH_RADIUS_METERS, EXPORT_CHUNK_BYTES, FOREIGN_STATS_TTL_MS, FULL_TURN_DEGREES, HEAP_PRESSURE_REARM_FRACTION, HEAP_PRESSURE_WARNING_FRACTION, IMPORT_BATCH_DOCUMENTS, INDEX_NAME_PATTERN, LOCATION_CODE_BITS_PER_AXIS, MAX_COVERING_CELLS, MAX_CURSOR_ANCHOR_CODE_POINTS, MAX_CURSOR_LENGTH, MAX_DOCUMENT_NESTING_DEPTH, MAX_DOC_ID_LENGTH, MAX_EXPORT_LINE_BYTES, MAX_IMPORT_REFUSALS_REPORTED, MAX_INDEX_CONFIGURATION_BYTES, MAX_INDEX_NAME_LENGTH, MAX_LATITUDE_DEGREES, MAX_LINEAR_ALPHA, MAX_LONGITUDE_DEGREES, MAX_NESTING_DEPTH, MAX_PATTERN_VALUE_LIMIT, MAX_PREFIX_LENGTH, MAX_SORT_FIELDS, MAX_SORT_FIELD_NAME_LENGTH, MAX_STORED_VECTOR_DIMENSION, MAX_SUGGEST_LIMIT, MAX_SUGGEST_SCATTER_LIMIT, MAX_SWEEP_INTERVAL_MS, MAX_TOLERANCE, MIN_DOCUMENTS_FOR_SEGMENTS, MIN_LINEAR_ALPHA, MIN_PATTERN_VALUE_LIMIT, MIN_POLYGON_POINTS, MIN_RANK_CONSTANT, MIN_SWEEP_INTERVAL_MS, MODULE_URL_SCHEME_PATTERN, PREFIX_MAX_EXPANSIONS, PROTOTYPE_POLLUTION_KEYS, RADIUS_BOX_MARGIN_RADIANS, RADIUS_BOX_MARGIN_RATIO, RESULT_WINDOW, STATISTICS_BROADCAST_INTERVAL_MS, SUGGEST_OVERSAMPLE_FACTOR, SUGGEST_OVERSAMPLE_PADDING, VECTOR_PATTERN, VINCENTY_CONVERGENCE_THRESHOLD, VINCENTY_MAX_ITERATIONS, encodeDocument, extractVectorFieldsFromSchema, flattenSchema, isGeopointOrVectorType, isLatitude, isListBase, isLongitude, isQuantizationMode, isSingleStringFieldType, isStorageMode, isStringValuedBase, isWordIndexedBase, isWordIndexedFieldType, messagePackUpperBoundBytes, parsedTypeOf, raiseEarlierReleaseGraphSetting, requireCreatableSchema, requireDocumentWithinSizeLimit, requireValidFieldTypes, schemaFieldsOf, validateDocument, validateDocumentStrict, validateRequiredFields, validateSchema, validateVectorPromotion, validateVectorStorage, withSortableOption };