@classytic/mongokit 3.2.0 → 3.2.2

package/dist/utils/{index.d.ts → index.d.mts}

@@ -1,14 +1,7 @@
-export { d as buildCrudSchemasFromModel, b as buildCrudSchemasFromMongooseSchema, k as configureLogger, j as createError, c as createFieldPreset, l as createMemoryCache, f as filterResponseData, g as getFieldsForUser, e as getImmutableFields, a as getMongooseProjection, h as getSystemManagedFields, i as isFieldUpdateAllowed, v as validateUpdateBody } from '../mongooseToJsonSchema-CaRF_bCN.js';
-import { S as SelectSpec, e as PopulateSpec, f as SortSpec } from '../types-Jni1KgkP.js';
-import 'mongoose';
-
-/**
- * Cache Key Utilities
- *
- * Generates deterministic, collision-free cache keys for MongoDB queries.
- * Key design inspired by Next.js cache tags and best practices from Stripe/Meta.
- */
+import { G as PopulateSpec, at as SortSpec, et as SelectSpec } from "../types-D-gploPr.mjs";
+import { a as isFieldUpdateAllowed, c as configureLogger, d as filterResponseData, f as getFieldsForUser, i as getSystemManagedFields, l as createError, n as buildCrudSchemasFromMongooseSchema, o as validateUpdateBody, p as getMongooseProjection, r as getImmutableFields, s as createMemoryCache, t as buildCrudSchemasFromModel, u as createFieldPreset } from "../mongooseToJsonSchema-Wbvjfwkn.mjs";
 
+//#region src/utils/cache-keys.d.ts
 /**
  * Generate cache key for getById operations
  *
@@ -29,8 +22,8 @@ declare function byIdKey(prefix: string, model: string, id: string): string;
  * // => 'mk:one:User:a1b2c3d4'
  */
 declare function byQueryKey(prefix: string, model: string, query: Record<string, unknown>, options?: {
-    select?: SelectSpec;
-    populate?: PopulateSpec;
+  select?: SelectSpec;
+  populate?: PopulateSpec;
 }): string;
 /**
  * Generate cache key for paginated list queries
@@ -46,13 +39,13 @@ declare function byQueryKey(prefix: string, model: string, query: Record<string,
  * // => 'mk:list:User:1:e5f6g7h8'
  */
 declare function listQueryKey(prefix: string, model: string, version: number, params: {
-    filters?: Record<string, unknown>;
-    sort?: SortSpec;
-    page?: number;
-    limit?: number;
-    after?: string;
-    select?: SelectSpec;
-    populate?: PopulateSpec;
+  filters?: Record<string, unknown>;
+  sort?: SortSpec;
+  page?: number;
+  limit?: number;
+  after?: string;
+  select?: SelectSpec;
+  populate?: PopulateSpec;
 }): string;
 /**
  * Generate cache key for collection version tag
@@ -78,5 +71,5 @@ declare function modelPattern(prefix: string, model: string): string;
  * Format: {prefix}:list:{model}:*
  */
 declare function listPattern(prefix: string, model: string): string;
-
-export { byIdKey, byQueryKey, listPattern, listQueryKey, modelPattern, versionKey };
+//#endregion
+export { buildCrudSchemasFromModel, buildCrudSchemasFromMongooseSchema, byIdKey, byQueryKey, configureLogger, createError, createFieldPreset, createMemoryCache, filterResponseData, getFieldsForUser, getImmutableFields, getMongooseProjection, getSystemManagedFields, isFieldUpdateAllowed, listPattern, listQueryKey, modelPattern, validateUpdateBody, versionKey };

package/dist/utils/index.mjs

@@ -0,0 +1,5 @@
+import { i as createError, t as configureLogger } from "../logger-D8ily-PP.mjs";
+import { a as modelPattern, c as filterResponseData, i as listQueryKey, l as getFieldsForUser, n as byQueryKey, o as versionKey, r as listPattern, s as createFieldPreset, t as byIdKey, u as getMongooseProjection } from "../cache-keys-C8Z9B5sw.mjs";
+import { a as isFieldUpdateAllowed, i as getSystemManagedFields, n as buildCrudSchemasFromMongooseSchema, o as validateUpdateBody, r as getImmutableFields, s as createMemoryCache, t as buildCrudSchemasFromModel } from "../mongooseToJsonSchema-COdDEkIJ.mjs";
+
+export { buildCrudSchemasFromModel, buildCrudSchemasFromMongooseSchema, byIdKey, byQueryKey, configureLogger, createError, createFieldPreset, createMemoryCache, filterResponseData, getFieldsForUser, getImmutableFields, getMongooseProjection, getSystemManagedFields, isFieldUpdateAllowed, listPattern, listQueryKey, modelPattern, validateUpdateBody, versionKey };

package/package.json

@@ -1,36 +1,36 @@
 {
   "name": "@classytic/mongokit",
-  "version": "3.2.0",
+  "version": "3.2.2",
   "description": "Production-grade MongoDB repositories with zero dependencies - smart pagination, events, and plugins",
   "type": "module",
   "sideEffects": false,
-  "main": "./dist/index.js",
-  "module": "./dist/index.js",
-  "types": "./dist/index.d.ts",
+  "main": "./dist/index.mjs",
+  "module": "./dist/index.mjs",
+  "types": "./dist/index.d.mts",
   "exports": {
     ".": {
-      "types": "./dist/index.d.ts",
-      "default": "./dist/index.js"
+      "types": "./dist/index.d.mts",
+      "default": "./dist/index.mjs"
     },
     "./pagination": {
-      "types": "./dist/pagination/PaginationEngine.d.ts",
-      "default": "./dist/pagination/PaginationEngine.js"
+      "types": "./dist/pagination/PaginationEngine.d.mts",
+      "default": "./dist/pagination/PaginationEngine.mjs"
     },
     "./plugins": {
-      "types": "./dist/plugins/index.d.ts",
-      "default": "./dist/plugins/index.js"
+      "types": "./dist/plugins/index.d.mts",
+      "default": "./dist/plugins/index.mjs"
     },
     "./utils": {
-      "types": "./dist/utils/index.d.ts",
-      "default": "./dist/utils/index.js"
+      "types": "./dist/utils/index.d.mts",
+      "default": "./dist/utils/index.mjs"
     },
     "./actions": {
-      "types": "./dist/actions/index.d.ts",
-      "default": "./dist/actions/index.js"
+      "types": "./dist/actions/index.d.mts",
+      "default": "./dist/actions/index.mjs"
     },
     "./ai": {
-      "types": "./dist/ai/index.d.ts",
-      "default": "./dist/ai/index.js"
+      "types": "./dist/ai/index.d.mts",
+      "default": "./dist/ai/index.mjs"
     }
   },
   "files": [
@@ -87,16 +87,16 @@
     "node": ">=18"
   },
   "scripts": {
-    "build": "tsup",
-    "dev": "tsup --watch",
+    "build": "tsdown",
+    "dev": "tsdown --watch",
     "test": "vitest run",
     "test:watch": "vitest",
     "test:coverage": "vitest run --coverage",
     "typecheck": "tsc --noEmit",
-    "prepublishOnly": "npm run build && npm run typecheck",
+    "prepublishOnly": "npm run build && npm run typecheck && npm test",
     "publish:dry": "npm publish --dry-run --access public",
     "publish:npm": "npm publish --access public",
-    "release": "npm run build && npm run typecheck && npm publish --access public",
+    "release": "npm run build && npm run typecheck && npm test && npm publish --access public",
     "release:patch": "npm version patch && npm run release",
     "release:minor": "npm version minor && npm run release",
     "release:major": "npm version major && npm run release"
@@ -106,7 +106,7 @@
     "@vitest/coverage-v8": "^3.2.4",
     "mongodb-memory-server": "^10.2.3",
     "mongoose": "^9.1.3",
-    "tsup": "^8.0.0",
+    "tsdown": "^0.20.3",
     "typescript": "^5.7.0",
     "vitest": "^3.0.0"
   }

package/dist/actions/index.d.ts

@@ -1,3 +0,0 @@
-export { b as aggregate, c as create, _ as deleteActions, r as read, u as update } from '../index-BDn5fSTE.js';
-import 'mongoose';
-import '../types-Jni1KgkP.js';

package/dist/actions/index.js

@@ -1,5 +0,0 @@
-export { aggregate_exports as aggregate, delete_exports as deleteActions, read_exports as read, update_exports as update } from '../chunks/chunk-VWKIKZYF.js';
-export { create_exports as create } from '../chunks/chunk-I7CWNAJB.js';
-import '../chunks/chunk-URLJFIR7.js';
-import '../chunks/chunk-JWUAVZ3L.js';
-import '../chunks/chunk-WSFCRVEQ.js';

package/dist/ai/index.d.ts

@@ -1,175 +0,0 @@
-import { ClientSession, PipelineStage } from 'mongoose';
-import { g as Plugin } from '../types-Jni1KgkP.js';
-
-/**
- * AI/Vector Search Type Definitions
- *
- * Types for vector embedding storage, search, and similarity operations.
- * Requires MongoDB Atlas for `$vectorSearch` aggregation.
- */
-
-/** Supported similarity metrics for vector search */
-type SimilarityMetric = 'cosine' | 'euclidean' | 'dotProduct';
-/** A single piece of content to embed — text, image, or any media */
-interface EmbeddingInput {
-    /** Text content to embed */
-    text?: string;
-    /** Image URL or base64 data (for multimodal models like CLIP, Jina v3) */
-    image?: string;
-    /** Audio URL or base64 data */
-    audio?: string;
-    /** Arbitrary media — for custom model inputs (video frames, PDFs, etc.) */
-    media?: Record<string, unknown>;
-}
-/**
- * Unified embedding function — receives structured input, returns vector.
- * Works for text-only, multimodal, or any custom model.
- *
- * @example
- * ```typescript
- * // Text-only (OpenAI)
- * const embed: EmbedFn = async ({ text }) =>
- *   openai.embeddings.create({ input: text!, model: 'text-embedding-3-small' })
- *     .then(r => r.data[0].embedding);
- *
- * // Multimodal (Jina CLIP v3)
- * const embed: EmbedFn = async ({ text, image }) =>
- *   jina.embed({ input: [{ text, image }] }).then(r => r.data[0].embedding);
- *
- * // Local model
- * const embed: EmbedFn = async ({ text }) =>
- *   fetch('http://localhost:11434/api/embeddings', {
- *     method: 'POST', body: JSON.stringify({ model: 'nomic-embed-text', prompt: text })
- *   }).then(r => r.json()).then(j => j.embedding);
- * ```
- */
-type EmbedFn = (input: EmbeddingInput) => Promise<number[]>;
-/**
- * Batch embedding function — same contract, multiple inputs at once.
- * Falls back to sequential EmbedFn calls if not provided.
- */
-type BatchEmbedFn = (inputs: EmbeddingInput[]) => Promise<number[][]>;
-/** Vector field configuration for a model */
-interface VectorFieldConfig {
-    /** Field path where the vector is stored (e.g., 'embedding') */
-    path: string;
-    /** Atlas Search index name for this field */
-    index: string;
-    /** Number of dimensions in the embedding */
-    dimensions: number;
-    /** Similarity metric used by the index (informational — the index defines this) */
-    similarity?: SimilarityMetric;
-    /** Text source fields to embed from (e.g., ['title', 'description']) */
-    sourceFields?: string[];
-    /** Image/media source fields (e.g., ['imageUrl', 'thumbnailUrl']) */
-    mediaFields?: string[];
-}
-/** Options for vector search operations */
-interface VectorSearchParams {
-    /** Query — vector, text string, or structured multimodal input */
-    query: number[] | string | EmbeddingInput;
-    /** Maximum number of results */
-    limit?: number;
-    /** Candidates to consider (higher = more accurate, slower). Default: limit * 10 */
-    numCandidates?: number;
-    /** Pre-filter documents before vector search */
-    filter?: Record<string, unknown>;
-    /** Use exact KNN instead of approximate (slower but precise) */
-    exact?: boolean;
-    /** Which vector field config to use (default: first configured) */
-    field?: string;
-    /** MongoDB session for transactions */
-    session?: ClientSession;
-    /** Fields to include/exclude in results */
-    project?: Record<string, 0 | 1>;
-    /** Include similarity score in results */
-    includeScore?: boolean;
-    /** Minimum score threshold (0-1 for cosine) */
-    minScore?: number;
-    /** Additional pipeline stages to append after search */
-    postPipeline?: PipelineStage[];
-}
-/** Vector search result with score */
-interface ScoredResult<T = Record<string, unknown>> {
-    /** The matched document */
-    doc: T;
-    /** Similarity score from vector search */
-    score: number;
-}
-/** Options for the vector search plugin */
-interface VectorPluginOptions {
-    /** Vector field configurations */
-    fields: VectorFieldConfig[];
-    /** Unified embedding function (text, image, multimodal) */
-    embedFn?: EmbedFn;
-    /** Batch embedding function for bulk operations */
-    batchEmbedFn?: BatchEmbedFn;
-    /** Auto-generate embeddings on create/update (requires embedFn) */
-    autoEmbed?: boolean;
-    /**
-     * Called when auto-embed fails (e.g., embedding service down).
-     * If provided, the write operation continues without an embedding.
-     * If not provided, the error propagates and blocks the write.
-     */
-    onEmbedError?: (error: Error, doc: unknown) => void;
-}
-
-/**
- * Vector Search Plugin
- *
- * Adds semantic similarity search to any repository using MongoDB Atlas Vector Search.
- * Supports auto-embedding on write, text-to-vector search, and scored results.
- *
- * **Requires MongoDB Atlas** — `$vectorSearch` is an Atlas-only aggregation stage.
- * Running on standalone or self-hosted MongoDB will throw an
- * `Unrecognized pipeline stage name: '$vectorSearch'` error.
- *
- * @example
- * ```typescript
- * import { vectorPlugin } from '@classytic/mongokit/ai';
- *
- * // Text-only (OpenAI)
- * const repo = new Repository(Product, [
- *   methodRegistryPlugin(),
- *   vectorPlugin({
- *     fields: [{ path: 'embedding', index: 'vec_idx', dimensions: 1536, similarity: 'cosine', sourceFields: ['title', 'description'] }],
- *     embedFn: async ({ text }) => openai.embeddings.create({ input: text!, model: 'text-embedding-3-small' }).then(r => r.data[0].embedding),
- *     autoEmbed: true,
- *   }),
- * ]);
- *
- * // Multimodal (Jina CLIP v3 — text + images in one call)
- * const repo = new Repository(Product, [
- *   methodRegistryPlugin(),
- *   vectorPlugin({
- *     fields: [{ path: 'embedding', index: 'vec_idx', dimensions: 1024, similarity: 'cosine', sourceFields: ['title'], mediaFields: ['imageUrl'] }],
- *     embedFn: async ({ text, image }) => jina.embed({ input: [{ text, image }] }).then(r => r.data[0].embedding),
- *     autoEmbed: true,
- *   }),
- * ]);
- *
- * // Search by text
- * const results = await repo.searchSimilar({ query: 'running shoes', limit: 10 });
- *
- * // Search by image + text (multimodal)
- * const results = await repo.searchSimilar({ query: { text: 'red sneakers', image: 'https://...' }, limit: 10 });
- *
- * // Search by vector directly
- * const results = await repo.searchSimilar({ query: [0.1, 0.2, ...], limit: 5 });
- * ```
- */
-
-interface VectorMethods {
-    searchSimilar<T = Record<string, unknown>>(params: VectorSearchParams): Promise<ScoredResult<T>[]>;
-    embed(input: EmbeddingInput | string): Promise<number[]>;
-}
-/**
- * Builds the $vectorSearch pipeline stage
- */
-declare function buildVectorSearchPipeline(field: VectorFieldConfig, queryVector: number[], params: VectorSearchParams): PipelineStage[];
-/**
- * Creates the vector search plugin
- */
-declare function vectorPlugin(options: VectorPluginOptions): Plugin;
-
-export { type BatchEmbedFn, type EmbedFn, type EmbeddingInput, type ScoredResult, type SimilarityMetric, type VectorFieldConfig, type VectorMethods, type VectorPluginOptions, type VectorSearchParams, buildVectorSearchPipeline, vectorPlugin };

package/dist/ai/index.js

@@ -1,206 +0,0 @@
-import '../chunks/chunk-WSFCRVEQ.js';
-
-// src/ai/vector.plugin.ts
-var MAX_NUM_CANDIDATES = 1e4;
-function resolveField(fields, fieldPath) {
-  if (fieldPath) {
-    const found = fields.find((f) => f.path === fieldPath);
-    if (!found) throw new Error(`[mongokit] Vector field '${fieldPath}' not configured`);
-    return found;
-  }
-  return fields[0];
-}
-function toEmbeddingInput(query) {
-  return typeof query === "string" ? { text: query } : query;
-}
-function getNestedValue(obj, path) {
-  if (path in obj) return obj[path];
-  return path.split(".").reduce((cur, key) => {
-    if (cur != null && typeof cur === "object") return cur[key];
-    return void 0;
-  }, obj);
-}
-function buildInputFromDoc(data, field) {
-  const input = {};
-  if (field.sourceFields?.length) {
-    const text = field.sourceFields.map((f) => getNestedValue(data, f)).filter(Boolean).join(" ");
-    if (text.trim()) input.text = text;
-  }
-  if (field.mediaFields?.length) {
-    const firstImageField = field.mediaFields[0];
-    const imageValue = getNestedValue(data, firstImageField);
-    if (typeof imageValue === "string") input.image = imageValue;
-    if (field.mediaFields.length > 1) {
-      input.media = {};
-      for (const mf of field.mediaFields) {
-        const val = getNestedValue(data, mf);
-        if (val != null) input.media[mf] = val;
-      }
-    }
-  }
-  return input;
-}
-function hasContent(input) {
-  return !!(input.text?.trim() || input.image || input.audio || input.media && Object.keys(input.media).length);
-}
-function buildVectorSearchPipeline(field, queryVector, params) {
-  const limit = params.limit ?? 10;
-  const stages = [];
-  const rawCandidates = params.numCandidates ?? Math.max(limit * 10, 100);
-  const numCandidates = Math.min(Math.max(rawCandidates, limit), MAX_NUM_CANDIDATES);
-  stages.push({
-    $vectorSearch: {
-      index: field.index,
-      path: field.path,
-      queryVector,
-      numCandidates,
-      limit,
-      ...params.filter && { filter: params.filter },
-      ...params.exact && { exact: true }
-    }
-  });
-  const needsScore = params.includeScore !== false || params.minScore != null;
-  if (needsScore) {
-    stages.push({ $addFields: { _score: { $meta: "vectorSearchScore" } } });
-  }
-  if (params.minScore != null) {
-    stages.push({ $match: { _score: { $gte: params.minScore } } });
-  }
-  if (params.project) {
-    stages.push({ $project: { ...params.project, _score: 1 } });
-  }
-  if (params.postPipeline?.length) {
-    stages.push(...params.postPipeline);
-  }
-  return stages;
-}
-function vectorPlugin(options) {
-  const { fields, autoEmbed = false } = options;
-  if (!fields?.length) {
-    throw new Error("[mongokit] vectorPlugin requires at least one field config");
-  }
-  const { embedFn, batchEmbedFn } = options;
-  return {
-    name: "vector",
-    apply(repo) {
-      if (!repo.registerMethod) {
-        throw new Error("[mongokit] vectorPlugin requires methodRegistryPlugin");
-      }
-      repo.registerMethod("searchSimilar", async function searchSimilar(params) {
-        const field = resolveField(fields, params.field);
-        let queryVector;
-        if (Array.isArray(params.query)) {
-          queryVector = params.query;
-        } else {
-          if (!embedFn) {
-            throw new Error("[mongokit] Non-vector queries require embedFn in vectorPlugin options");
-          }
-          const input = toEmbeddingInput(params.query);
-          queryVector = await embedFn(input);
-        }
-        if (queryVector.length !== field.dimensions) {
-          throw new Error(
-            `[mongokit] Query vector has ${queryVector.length} dimensions, expected ${field.dimensions}`
-          );
-        }
-        const pipeline = buildVectorSearchPipeline(field, queryVector, params);
-        const agg = repo.Model.aggregate(pipeline);
-        if (params.session) agg.session(params.session);
-        const results = await agg.exec();
-        return results.map((doc) => {
-          const score = doc._score ?? 0;
-          const { _score, ...rest } = doc;
-          return { doc: rest, score };
-        });
-      });
-      repo.registerMethod("embed", async function embed(input) {
-        if (!embedFn) {
-          throw new Error("[mongokit] embed requires embedFn in vectorPlugin options");
-        }
-        return embedFn(typeof input === "string" ? { text: input } : input);
-      });
-      if (autoEmbed && embedFn) {
-        const { onEmbedError } = options;
-        const safeEmbed = async (input, doc) => {
-          try {
-            return await embedFn(input);
-          } catch (err) {
-            if (onEmbedError) {
-              onEmbedError(err, doc);
-              return null;
-            }
-            throw err;
-          }
-        };
-        const embedFromSource = async (data, field) => {
-          if (data[field.path] && Array.isArray(data[field.path])) return;
-          const input = buildInputFromDoc(data, field);
-          if (!hasContent(input)) return;
-          const vector = await safeEmbed(input, data);
-          if (vector) data[field.path] = vector;
-        };
-        const embedBatchFromSource = async (dataArray, field) => {
-          const toEmbed = [];
-          for (let i = 0; i < dataArray.length; i++) {
-            const data = dataArray[i];
-            if (data[field.path] && Array.isArray(data[field.path])) continue;
-            const input = buildInputFromDoc(data, field);
-            if (hasContent(input)) toEmbed.push({ idx: i, input });
-          }
-          if (!toEmbed.length) return;
-          if (batchEmbedFn) {
-            try {
-              const vectors = await batchEmbedFn(toEmbed.map((e) => e.input));
-              for (let i = 0; i < toEmbed.length; i++) {
-                dataArray[toEmbed[i].idx][field.path] = vectors[i];
-              }
-            } catch (err) {
-              if (onEmbedError) {
-                onEmbedError(err, dataArray);
-                return;
-              }
-              throw err;
-            }
-          } else {
-            for (const entry of toEmbed) {
-              const vector = await safeEmbed(entry.input, dataArray[entry.idx]);
-              if (vector) dataArray[entry.idx][field.path] = vector;
-            }
-          }
-        };
-        repo.on("before:create", async (context) => {
-          if (!context.data) return;
-          for (const field of fields) {
-            await embedFromSource(context.data, field);
-          }
-        });
-        repo.on("before:createMany", async (context) => {
-          if (!context.dataArray?.length) return;
-          for (const field of fields) {
-            await embedBatchFromSource(context.dataArray, field);
-          }
-        });
-        repo.on("before:update", async (context) => {
-          if (!context.data) return;
-          const fieldsToEmbed = fields.filter((field) => {
-            const allFields = [...field.sourceFields ?? [], ...field.mediaFields ?? []];
-            return allFields.length > 0 && allFields.some((f) => f in context.data);
-          });
-          if (!fieldsToEmbed.length) return;
-          const existing = await repo.Model.findById(context.id).lean().session(context.session ?? null);
-          if (!existing) return;
-          for (const field of fieldsToEmbed) {
-            const merged = { ...existing, ...context.data };
-            delete merged[field.path];
-            const input = buildInputFromDoc(merged, field);
-            if (!hasContent(input)) continue;
-            const vector = await safeEmbed(input, merged);
-            if (vector) context.data[field.path] = vector;
-          }
-        });
-      }
-    }
-  };
-}
-
-export { buildVectorSearchPipeline, vectorPlugin };

package/dist/chunks/chunk-2ZN65ZOP.js

@@ -1,93 +0,0 @@
-// src/utils/field-selection.ts
-function getFieldsForUser(user, preset) {
-  if (!preset) {
-    throw new Error("Field preset is required");
-  }
-  const fields = [...preset.public || []];
-  if (user) {
-    fields.push(...preset.authenticated || []);
-    const roles = Array.isArray(user.roles) ? user.roles : user.roles ? [user.roles] : [];
-    if (roles.includes("admin") || roles.includes("superadmin")) {
-      fields.push(...preset.admin || []);
-    }
-  }
-  return [...new Set(fields)];
-}
-function getMongooseProjection(user, preset) {
-  const fields = getFieldsForUser(user, preset);
-  return fields.join(" ");
-}
-function filterObject(obj, allowedFields) {
-  if (!obj || typeof obj !== "object" || Array.isArray(obj)) {
-    return obj;
-  }
-  const filtered = {};
-  for (const field of allowedFields) {
-    if (field in obj) {
-      filtered[field] = obj[field];
-    }
-  }
-  return filtered;
-}
-function filterResponseData(data, preset, user = null) {
-  const allowedFields = getFieldsForUser(user, preset);
-  if (Array.isArray(data)) {
-    return data.map((item) => filterObject(item, allowedFields));
-  }
-  return filterObject(data, allowedFields);
-}
-function createFieldPreset(config) {
-  return {
-    public: config.public || [],
-    authenticated: config.authenticated || [],
-    admin: config.admin || []
-  };
-}
-
-// src/utils/cache-keys.ts
-function hashString(str) {
-  let hash = 5381;
-  for (let i = 0; i < str.length; i++) {
-    hash = (hash << 5) + hash ^ str.charCodeAt(i);
-  }
-  return (hash >>> 0).toString(16);
-}
-function stableStringify(obj) {
-  if (obj === null || obj === void 0) return "";
-  if (typeof obj !== "object") return String(obj);
-  if (Array.isArray(obj)) {
-    return "[" + obj.map(stableStringify).join(",") + "]";
-  }
-  const sorted = Object.keys(obj).sort().map((key) => `${key}:${stableStringify(obj[key])}`);
-  return "{" + sorted.join(",") + "}";
-}
-function byIdKey(prefix, model, id) {
-  return `${prefix}:id:${model}:${id}`;
-}
-function byQueryKey(prefix, model, query, options) {
-  const hashInput = stableStringify({ q: query, s: options?.select, p: options?.populate });
-  return `${prefix}:one:${model}:${hashString(hashInput)}`;
-}
-function listQueryKey(prefix, model, version, params) {
-  const hashInput = stableStringify({
-    f: params.filters,
-    s: params.sort,
-    pg: params.page,
-    lm: params.limit,
-    af: params.after,
-    sl: params.select,
-    pp: params.populate
-  });
-  return `${prefix}:list:${model}:${version}:${hashString(hashInput)}`;
-}
-function versionKey(prefix, model) {
-  return `${prefix}:ver:${model}`;
-}
-function modelPattern(prefix, model) {
-  return `${prefix}:*:${model}:*`;
-}
-function listPattern(prefix, model) {
-  return `${prefix}:list:${model}:*`;
-}
-
-export { byIdKey, byQueryKey, createFieldPreset, filterResponseData, getFieldsForUser, getMongooseProjection, listPattern, listQueryKey, modelPattern, versionKey };