LibreChat/packages/data-schemas/src/models/plugins/mongoMeili.ts
Danny Avila 29ef91b4dd
🧠 feat: User Memories for Conversational Context (#7760)
* 🧠 feat: User Memories for Conversational Context

chore: mcp typing, use `t`

WIP: first pass, Memories UI

- Added MemoryViewer component for displaying, editing, and deleting user memories.
- Integrated data provider hooks for fetching, updating, and deleting memories.
- Implemented pagination and loading states for better user experience.
- Created unit tests for MemoryViewer to ensure functionality and interaction with data provider.
- Updated translation files to include new UI strings related to memories.

chore: move mcp-related files to own directory

chore: rename librechat-mcp to librechat-api

WIP: first pass, memory processing and data schemas

chore: linting in fileSearch.js query description

chore: rename librechat-api to @librechat/api across the project

WIP: first pass, functional memory agent

feat: add MemoryEditDialog and MemoryViewer components for managing user memories

- Introduced MemoryEditDialog for editing memory entries with validation and toast notifications.
- Updated MemoryViewer to support editing and deleting memories, including pagination and loading states.
- Enhanced data provider to handle memory updates with optional original key for better management.
- Added new localization strings for memory-related UI elements.

feat: add memory permissions management

- Implemented memory permissions in the backend, allowing roles to have specific permissions for using, creating, updating, and reading memories.
- Added new API endpoints for updating memory permissions associated with roles.
- Created a new AdminSettings component for managing memory permissions in the frontend.
- Integrated memory permissions into the existing roles and permissions schemas.
- Updated the interface to include memory settings and permissions.
- Enhanced the MemoryViewer component to conditionally render admin settings based on user roles.
- Added localization support for memory permissions in the translation files.

feat: move AdminSettings component to a new position in MemoryViewer for better visibility

refactor: clean up commented code in MemoryViewer component

feat: enhance MemoryViewer with search functionality and improve MemoryEditDialog integration

- Added a search input to filter memories in the MemoryViewer component.
- Refactored MemoryEditDialog to accept children for better customization.
- Updated MemoryViewer to utilize the new EditMemoryButton and DeleteMemoryButton components for editing and deleting memories.
- Improved localization support by adding new strings for memory filtering and deletion confirmation.

refactor: optimize memory filtering in MemoryViewer using match-sorter

- Replaced manual filtering logic with match-sorter for improved search functionality.
- Enhanced performance and readability of the filteredMemories computation.

feat: enhance MemoryEditDialog with triggerRef and improve updateMemory mutation handling

feat: implement access control for MemoryEditDialog and MemoryViewer components

refactor: remove commented out code and create runMemory method

refactor: rename role based files

feat: implement access control for memory usage in AgentClient

refactor: simplify checkVisionRequest method in AgentClient by removing commented-out code

refactor: make `agents` dir in api package

refactor: migrate Azure utilities to TypeScript and consolidate imports

refactor: move sanitizeFilename function to a new file and update imports, add related tests

refactor: update LLM configuration types and consolidate Azure options in the API package

chore: linting

chore: import order

refactor: replace getLLMConfig with getOpenAIConfig and remove unused LLM configuration file

chore: update winston-daily-rotate-file to version 5.0.0 and add object-hash dependency in package-lock.json

refactor: move primeResources and optionalChainWithEmptyCheck functions to resources.ts and update imports

refactor: move createRun function to a new run.ts file and update related imports

fix: ensure safeAttachments is correctly typed as an array of TFile

chore: add node-fetch dependency and refactor fetch-related functions into packages/api/utils, removing the old generators file

refactor: enhance TEndpointOption type by using Pick to streamline endpoint fields and add new properties for model parameters and client options

feat: implement initializeOpenAIOptions function and update OpenAI types for enhanced configuration handling

fix: update types due to new TEndpointOption typing

fix: ensure safe access to group parameters in initializeOpenAIOptions function

fix: remove redundant API key validation comment in initializeOpenAIOptions function

refactor: rename initializeOpenAIOptions to initializeOpenAI for consistency and update related documentation

refactor: decouple req.body fields and tool loading from initializeAgentOptions

chore: linting

refactor: adjust column widths in MemoryViewer for improved layout

refactor: simplify agent initialization by creating loadAgent function and removing unused code

feat: add memory configuration loading and validation functions

WIP: first pass, memory processing with config

feat: implement memory callback and artifact handling

feat: implement memory artifacts display and processing updates

feat: add memory configuration options and schema validation for validKeys

fix: update MemoryEditDialog and MemoryViewer to handle memory state and display improvements

refactor: remove padding from BookmarkTable and MemoryViewer headers for consistent styling

WIP: initial tokenLimit config and move Tokenizer to @librechat/api

refactor: update mongoMeili plugin methods to use callback for better error handling

feat: enhance memory management with token tracking and usage metrics

- Added token counting for memory entries to enforce limits and provide usage statistics.
- Updated memory retrieval and update routes to include total token usage and limit.
- Enhanced MemoryEditDialog and MemoryViewer components to display memory usage and token information.
- Refactored memory processing functions to handle token limits and provide feedback on memory capacity.

feat: implement memory artifact handling in attachment handler

- Enhanced useAttachmentHandler to process memory artifacts when receiving updates.
- Introduced handleMemoryArtifact utility to manage memory updates and deletions.
- Updated query client to reflect changes in memory state based on incoming data.

refactor: restructure web search key extraction logic

- Moved the logic for extracting API keys from the webSearchAuth configuration into a dedicated function, getWebSearchKeys.
- Updated webSearchKeys to utilize the new function for improved clarity and maintainability.
- Prevents build time errors

feat: add personalization settings and memory preferences management

- Introduced a new Personalization tab in settings to manage user memory preferences.
- Implemented API endpoints and client-side logic for updating memory preferences.
- Enhanced user interface components to reflect personalization options and memory usage.
- Updated permissions to allow users to opt out of memory features.
- Added localization support for new settings and messages related to personalization.

style: personalization switch class

feat: add PersonalizationIcon and align Side Panel UI

feat: implement memory creation functionality

- Added a new API endpoint for creating memory entries, including validation for key and value.
- Introduced MemoryCreateDialog component for user interface to facilitate memory creation.
- Integrated token limit checks to prevent exceeding user memory capacity.
- Updated MemoryViewer to include a button for opening the memory creation dialog.
- Enhanced localization support for new messages related to memory creation.

feat: enhance message processing with configurable window size

- Updated AgentClient to use a configurable message window size for processing messages.
- Introduced messageWindowSize option in memory configuration schema with a default value of 5.
- Improved logic for selecting messages to process based on the configured window size.

chore: update librechat-data-provider version to 0.7.87 in package.json and package-lock.json

chore: remove OpenAPIPlugin and its associated tests

chore: remove MIGRATION_README.md as migration tasks are completed

ci: fix backend tests

chore: remove unused translation keys from localization file

chore: remove problematic test file and unused var in AgentClient

chore: remove unused import and import directly for JSDoc

* feat: add api package build stage in Dockerfile for improved modularity

* docs: reorder build steps in contributing guide for clarity
2025-06-07 18:52:22 -04:00

564 lines
18 KiB
TypeScript

import _ from 'lodash';
import { MeiliSearch, Index } from 'meilisearch';
import type {
CallbackWithoutResultAndOptionalError,
FilterQuery,
Document,
Schema,
Query,
Types,
Model,
} from 'mongoose';
import logger from '~/config/meiliLogger';
interface MongoMeiliOptions {
host: string;
apiKey: string;
indexName: string;
primaryKey: string;
mongoose: typeof import('mongoose');
}
interface MeiliIndexable {
[key: string]: unknown;
_meiliIndex?: boolean;
}
interface ContentItem {
type: string;
text?: string;
}
interface DocumentWithMeiliIndex extends Document {
_meiliIndex?: boolean;
preprocessObjectForIndex?: () => Record<string, unknown>;
addObjectToMeili?: (next: CallbackWithoutResultAndOptionalError) => Promise<void>;
updateObjectToMeili?: (next: CallbackWithoutResultAndOptionalError) => Promise<void>;
deleteObjectFromMeili?: (next: CallbackWithoutResultAndOptionalError) => Promise<void>;
postSaveHook?: (next: CallbackWithoutResultAndOptionalError) => void;
postUpdateHook?: (next: CallbackWithoutResultAndOptionalError) => void;
postRemoveHook?: (next: CallbackWithoutResultAndOptionalError) => void;
conversationId?: string;
content?: ContentItem[];
messageId?: string;
unfinished?: boolean;
messages?: unknown[];
title?: string;
toJSON(): Record<string, unknown>;
}
interface SchemaWithMeiliMethods extends Model<DocumentWithMeiliIndex> {
syncWithMeili(): Promise<void>;
setMeiliIndexSettings(settings: Record<string, unknown>): Promise<unknown>;
meiliSearch(q: string, params: Record<string, unknown>, populate: boolean): Promise<unknown>;
}
// Environment flags
/**
* Flag to indicate if search is enabled based on environment variables.
*/
const searchEnabled = process.env.SEARCH != null && process.env.SEARCH.toLowerCase() === 'true';
/**
* Flag to indicate if MeiliSearch is enabled based on required environment variables.
*/
const meiliEnabled =
process.env.MEILI_HOST != null && process.env.MEILI_MASTER_KEY != null && searchEnabled;
/**
* Local implementation of parseTextParts to avoid dependency on librechat-data-provider
* Extracts text content from an array of content items
*/
const parseTextParts = (content: ContentItem[]): string => {
if (!Array.isArray(content)) {
return '';
}
return content
.filter((item) => item.type === 'text' && typeof item.text === 'string')
.map((item) => item.text)
.join(' ')
.trim();
};
/**
* Local implementation to handle Bing convoId conversion
*/
const cleanUpPrimaryKeyValue = (value: string): string => {
return value.replace(/--/g, '|');
};
/**
* Validates the required options for configuring the mongoMeili plugin.
*/
const validateOptions = (options: Partial<MongoMeiliOptions>): void => {
const requiredKeys: (keyof MongoMeiliOptions)[] = ['host', 'apiKey', 'indexName'];
requiredKeys.forEach((key) => {
if (!options[key]) {
throw new Error(`Missing mongoMeili Option: ${key}`);
}
});
};
/**
* Factory function to create a MeiliMongooseModel class which extends a Mongoose model.
* This class contains static and instance methods to synchronize and manage the MeiliSearch index
* corresponding to the MongoDB collection.
*
* @param config - Configuration object.
* @param config.index - The MeiliSearch index object.
* @param config.attributesToIndex - List of attributes to index.
* @returns A class definition that will be loaded into the Mongoose schema.
*/
const createMeiliMongooseModel = ({
index,
attributesToIndex,
}: {
index: Index<MeiliIndexable>;
attributesToIndex: string[];
}) => {
const primaryKey = attributesToIndex[0];
class MeiliMongooseModel {
/**
* Synchronizes the data between the MongoDB collection and the MeiliSearch index.
*
* The synchronization process involves:
* 1. Fetching all documents from the MongoDB collection and MeiliSearch index.
* 2. Comparing documents from both sources.
* 3. Deleting documents from MeiliSearch that no longer exist in MongoDB.
* 4. Adding documents to MeiliSearch that exist in MongoDB but not in the index.
* 5. Updating documents in MeiliSearch if key fields (such as `text` or `title`) differ.
* 6. Updating the `_meiliIndex` field in MongoDB to indicate the indexing status.
*
* Note: The function processes documents in batches because MeiliSearch's
* `index.getDocuments` requires an exact limit and `index.addDocuments` does not handle
* partial failures in a batch.
*
* @returns {Promise<void>} Resolves when the synchronization is complete.
*/
static async syncWithMeili(this: SchemaWithMeiliMethods): Promise<void> {
try {
let moreDocuments = true;
const mongoDocuments = await this.find().lean();
const format = (doc: Record<string, unknown>) =>
_.omitBy(_.pick(doc, attributesToIndex), (v, k) => k.startsWith('$'));
const mongoMap = new Map(
mongoDocuments.map((doc) => {
const typedDoc = doc as Record<string, unknown>;
return [typedDoc[primaryKey], format(typedDoc)];
}),
);
const indexMap = new Map<unknown, Record<string, unknown>>();
let offset = 0;
const batchSize = 1000;
while (moreDocuments) {
const batch = await index.getDocuments({ limit: batchSize, offset });
if (batch.results.length === 0) {
moreDocuments = false;
}
for (const doc of batch.results) {
indexMap.set(doc[primaryKey], format(doc));
}
offset += batchSize;
}
logger.debug('[syncWithMeili]', { indexMap: indexMap.size, mongoMap: mongoMap.size });
const updateOps: Array<{
updateOne: {
filter: Record<string, unknown>;
update: { $set: { _meiliIndex: boolean } };
};
}> = [];
// Process documents present in the MeiliSearch index
for (const [id, doc] of indexMap) {
const update: Record<string, unknown> = {};
update[primaryKey] = id;
if (mongoMap.has(id)) {
const mongoDoc = mongoMap.get(id);
if (
(doc.text && doc.text !== mongoDoc?.text) ||
(doc.title && doc.title !== mongoDoc?.title)
) {
logger.debug(
`[syncWithMeili] ${id} had document discrepancy in ${
doc.text ? 'text' : 'title'
} field`,
);
updateOps.push({
updateOne: { filter: update, update: { $set: { _meiliIndex: true } } },
});
await index.addDocuments([doc]);
}
} else {
await index.deleteDocument(id as string);
updateOps.push({
updateOne: { filter: update, update: { $set: { _meiliIndex: false } } },
});
}
}
// Process documents present in MongoDB
for (const [id, doc] of mongoMap) {
const update: Record<string, unknown> = {};
update[primaryKey] = id;
if (!indexMap.has(id)) {
await index.addDocuments([doc]);
updateOps.push({
updateOne: { filter: update, update: { $set: { _meiliIndex: true } } },
});
} else if (doc._meiliIndex === false) {
updateOps.push({
updateOne: { filter: update, update: { $set: { _meiliIndex: true } } },
});
}
}
if (updateOps.length > 0) {
await this.collection.bulkWrite(updateOps);
logger.debug(
`[syncWithMeili] Finished indexing ${
primaryKey === 'messageId' ? 'messages' : 'conversations'
}`,
);
}
} catch (error) {
logger.error('[syncWithMeili] Error adding document to Meili:', error);
}
}
/**
* Updates settings for the MeiliSearch index
*/
static async setMeiliIndexSettings(settings: Record<string, unknown>): Promise<unknown> {
return await index.updateSettings(settings);
}
/**
* Searches the MeiliSearch index and optionally populates results
*/
static async meiliSearch(
this: SchemaWithMeiliMethods,
q: string,
params: Record<string, unknown>,
populate: boolean,
): Promise<unknown> {
const data = await index.search(q, params);
if (populate) {
const query: Record<string, unknown> = {};
query[primaryKey] = _.map(data.hits, (hit) =>
cleanUpPrimaryKeyValue(hit[primaryKey] as string),
);
const projection = Object.keys(this.schema.obj).reduce<Record<string, number>>(
(results, key) => {
if (!key.startsWith('$')) {
results[key] = 1;
}
return results;
},
{ _id: 1, __v: 1 },
);
const hitsFromMongoose = await this.find(query, projection).lean();
const populatedHits = data.hits.map((hit) => {
const queryObj: Record<string, unknown> = {};
queryObj[primaryKey] = hit[primaryKey];
const originalHit = _.find(hitsFromMongoose, (item) => {
const typedItem = item as Record<string, unknown>;
return typedItem[primaryKey] === hit[primaryKey];
});
return {
...(originalHit && typeof originalHit === 'object' ? originalHit : {}),
...hit,
};
});
data.hits = populatedHits;
}
return data;
}
/**
* Preprocesses the current document for indexing
*/
preprocessObjectForIndex(this: DocumentWithMeiliIndex): Record<string, unknown> {
const object = _.omitBy(_.pick(this.toJSON(), attributesToIndex), (v, k) =>
k.startsWith('$'),
);
if (
object.conversationId &&
typeof object.conversationId === 'string' &&
object.conversationId.includes('|')
) {
object.conversationId = object.conversationId.replace(/\|/g, '--');
}
if (object.content && Array.isArray(object.content)) {
object.text = parseTextParts(object.content);
delete object.content;
}
return object;
}
/**
* Adds the current document to the MeiliSearch index
*/
async addObjectToMeili(
this: DocumentWithMeiliIndex,
next: CallbackWithoutResultAndOptionalError,
): Promise<void> {
const object = this.preprocessObjectForIndex!();
try {
await index.addDocuments([object]);
} catch (error) {
logger.error('[addObjectToMeili] Error adding document to Meili:', error);
return next();
}
try {
await this.collection.updateMany(
{ _id: this._id as Types.ObjectId },
{ $set: { _meiliIndex: true } },
);
} catch (error) {
logger.error('[addObjectToMeili] Error updating _meiliIndex field:', error);
return next();
}
next();
}
/**
* Updates the current document in the MeiliSearch index
*/
async updateObjectToMeili(
this: DocumentWithMeiliIndex,
next: CallbackWithoutResultAndOptionalError,
): Promise<void> {
try {
const object = _.omitBy(_.pick(this.toJSON(), attributesToIndex), (v, k) =>
k.startsWith('$'),
);
await index.updateDocuments([object]);
next();
} catch (error) {
logger.error('[updateObjectToMeili] Error updating document in Meili:', error);
return next();
}
}
/**
* Deletes the current document from the MeiliSearch index.
*
* @returns {Promise<void>}
*/
async deleteObjectFromMeili(
this: DocumentWithMeiliIndex,
next: CallbackWithoutResultAndOptionalError,
): Promise<void> {
try {
await index.deleteDocument(this._id as string);
next();
} catch (error) {
logger.error('[deleteObjectFromMeili] Error deleting document from Meili:', error);
return next();
}
}
/**
* Post-save hook to synchronize the document with MeiliSearch.
*
* If the document is already indexed (i.e. `_meiliIndex` is true), it updates it;
* otherwise, it adds the document to the index.
*/
postSaveHook(this: DocumentWithMeiliIndex, next: CallbackWithoutResultAndOptionalError): void {
if (this._meiliIndex) {
this.updateObjectToMeili!(next);
} else {
this.addObjectToMeili!(next);
}
}
/**
* Post-update hook to update the document in MeiliSearch.
*
* This hook is triggered after a document update, ensuring that changes are
* propagated to the MeiliSearch index if the document is indexed.
*/
postUpdateHook(
this: DocumentWithMeiliIndex,
next: CallbackWithoutResultAndOptionalError,
): void {
if (this._meiliIndex) {
this.updateObjectToMeili!(next);
} else {
next();
}
}
/**
* Post-remove hook to delete the document from MeiliSearch.
*
* This hook is triggered after a document is removed, ensuring that the document
* is also removed from the MeiliSearch index if it was previously indexed.
*/
postRemoveHook(
this: DocumentWithMeiliIndex,
next: CallbackWithoutResultAndOptionalError,
): void {
if (this._meiliIndex) {
this.deleteObjectFromMeili!(next);
} else {
next();
}
}
}
return MeiliMongooseModel;
};
/**
* Mongoose plugin to synchronize MongoDB collections with a MeiliSearch index.
*
* This plugin:
* - Validates the provided options.
* - Adds a `_meiliIndex` field to the schema to track indexing status.
* - Sets up a MeiliSearch client and creates an index if it doesn't already exist.
* - Loads class methods for syncing, searching, and managing documents in MeiliSearch.
* - Registers Mongoose hooks (post-save, post-update, post-remove, etc.) to maintain index consistency.
*
* @param schema - The Mongoose schema to which the plugin is applied.
* @param options - Configuration options.
* @param options.host - The MeiliSearch host.
* @param options.apiKey - The MeiliSearch API key.
* @param options.indexName - The name of the MeiliSearch index.
* @param options.primaryKey - The primary key field for indexing.
*/
export default function mongoMeili(schema: Schema, options: MongoMeiliOptions): void {
const mongoose = options.mongoose;
validateOptions(options);
// Add _meiliIndex field to the schema to track if a document has been indexed in MeiliSearch.
schema.add({
_meiliIndex: {
type: Boolean,
required: false,
select: false,
default: false,
},
});
const { host, apiKey, indexName, primaryKey } = options;
const client = new MeiliSearch({ host, apiKey });
client.createIndex(indexName, { primaryKey });
const index = client.index<MeiliIndexable>(indexName);
// Collect attributes from the schema that should be indexed
const attributesToIndex: string[] = [
...Object.entries(schema.obj).reduce<string[]>((results, [key, value]) => {
const schemaValue = value as { meiliIndex?: boolean };
return schemaValue.meiliIndex ? [...results, key] : results;
}, []),
];
schema.loadClass(createMeiliMongooseModel({ index, attributesToIndex }));
// Register Mongoose hooks
schema.post('save', function (doc: DocumentWithMeiliIndex, next) {
doc.postSaveHook?.(next);
});
schema.post('updateOne', function (doc: DocumentWithMeiliIndex, next) {
doc.postUpdateHook?.(next);
});
schema.post('deleteOne', function (doc: DocumentWithMeiliIndex, next) {
doc.postRemoveHook?.(next);
});
// Pre-deleteMany hook: remove corresponding documents from MeiliSearch when multiple documents are deleted.
schema.pre('deleteMany', async function (next) {
if (!meiliEnabled) {
return next();
}
try {
const conditions = (this as Query<unknown, unknown>).getQuery();
if (Object.prototype.hasOwnProperty.call(schema.obj, 'messages')) {
const convoIndex = client.index('convos');
const deletedConvos = await mongoose
.model('Conversation')
.find(conditions as FilterQuery<unknown>)
.lean();
const promises = deletedConvos.map((convo: Record<string, unknown>) =>
convoIndex.deleteDocument(convo.conversationId as string),
);
await Promise.all(promises);
}
if (Object.prototype.hasOwnProperty.call(schema.obj, 'messageId')) {
const messageIndex = client.index('messages');
const deletedMessages = await mongoose
.model('Message')
.find(conditions as FilterQuery<unknown>)
.lean();
const promises = deletedMessages.map((message: Record<string, unknown>) =>
messageIndex.deleteDocument(message.messageId as string),
);
await Promise.all(promises);
}
return next();
} catch (error) {
if (meiliEnabled) {
logger.error(
'[MeiliMongooseModel.deleteMany] There was an issue deleting conversation indexes upon deletion. Next startup may be slow due to syncing.',
error,
);
}
return next();
}
});
// Post-findOneAndUpdate hook
schema.post('findOneAndUpdate', async function (doc: DocumentWithMeiliIndex, next) {
if (!meiliEnabled) {
return next();
}
if (doc.unfinished) {
return next();
}
let meiliDoc: Record<string, unknown> | undefined;
if (doc.messages) {
try {
meiliDoc = await client.index('convos').getDocument(doc.conversationId as string);
} catch (error: unknown) {
logger.debug(
'[MeiliMongooseModel.findOneAndUpdate] Convo not found in MeiliSearch and will index ' +
doc.conversationId,
error as Record<string, unknown>,
);
}
}
if (meiliDoc && meiliDoc.title === doc.title) {
return next();
}
doc.postSaveHook?.(next);
});
}