feat(llm): remove LLM deprecated functions

apps/server/src/routes/api/embeddings.ts

@@ -408,7 +408,7 @@ async function reprocessAllNotes(req: Request, res: Response) {
         try {
             // Wrap the operation in cls.init to ensure proper context
             cls.init(async () => {
-                await vectorStore.reprocessAllNotes();
+                await indexService.reprocessAllNotes();
                 log.info("Embedding reprocessing completed successfully");
             });
         } catch (error: any) {

apps/server/src/services/llm/ai_service_manager.ts

@@ -512,7 +512,7 @@ export class AIServiceManager implements IAIServiceManager {
             if (!contextNotes || contextNotes.length === 0) {
                 try {
                     // Get the default LLM service for context enhancement
-                    const provider = this.getPreferredProvider();
+                    const provider = this.getSelectedProvider();
                     const llmService = await this.getService(provider);
 
                     // Find relevant notes
@@ -596,9 +596,9 @@ export class AIServiceManager implements IAIServiceManager {
     }
 
     /**
-     * Get the preferred provider based on configuration (sync version for compatibility)
+     * Get the selected provider based on configuration (sync version for compatibility)
      */
-    getPreferredProvider(): string {
+    getSelectedProvider(): string {
         this.ensureInitialized();
 
         // Return the first available provider in the order
@@ -803,8 +803,8 @@ export default {
     async getService(provider?: string): Promise<AIService> {
         return getInstance().getService(provider);
     },
-    getPreferredProvider(): string {
+    getSelectedProvider(): string {
-        return getInstance().getPreferredProvider();
+        return getInstance().getSelectedProvider();
     },
     isProviderAvailable(provider: string): boolean {
         return getInstance().isProviderAvailable(provider);

apps/server/src/services/llm/chat/rest_chat_service.ts

@@ -14,7 +14,7 @@ import type { LLMStreamMessage } from "../interfaces/chat_ws_messages.js";
 import chatStorageService from '../chat_storage_service.js';
 import {
     isAIEnabled,
-    getFirstValidModelConfig,
+    getSelectedModelConfig,
 } from '../config/configuration_helpers.js';
 
 /**
@@ -419,7 +419,7 @@ class RestChatService {
      */
     async getPreferredModel(): Promise<string | undefined> {
         try {
-            const validConfig = await getFirstValidModelConfig();
+            const validConfig = await getSelectedModelConfig();
             if (!validConfig) {
                 log.error('No valid AI model configuration found');
                 return undefined;

apps/server/src/services/llm/config/configuration_helpers.ts

@@ -150,47 +150,3 @@ export async function getSelectedModelConfig(): Promise<{ model: string; provide
     return await getValidModelConfig(selectedProvider);
 }
 
-// Legacy support functions - these maintain backwards compatibility but now use single provider logic
-/**
- * @deprecated Use getSelectedProvider() instead
- */
-export async function getProviderPrecedence(): Promise<ProviderType[]> {
-    const selected = await getSelectedProvider();
-    return selected ? [selected] : [];
-}
-
-/**
- * @deprecated Use getSelectedProvider() instead  
- */
-export async function getPreferredProvider(): Promise<ProviderType | null> {
-    return await getSelectedProvider();
-}
-
-/**
- * @deprecated Use getSelectedEmbeddingProvider() instead
- */
-export async function getEmbeddingProviderPrecedence(): Promise<string[]> {
-    const selected = await getSelectedEmbeddingProvider();
-    return selected ? [selected] : [];
-}
-
-/**
- * @deprecated Use getSelectedEmbeddingProvider() instead
- */
-export async function getPreferredEmbeddingProvider(): Promise<string | null> {
-    return await getSelectedEmbeddingProvider();
-}
-
-/**
- * @deprecated Use getAvailableSelectedProvider() instead
- */
-export async function getFirstAvailableProvider(): Promise<ProviderType | null> {
-    return await getAvailableSelectedProvider();
-}
-
-/**
- * @deprecated Use getSelectedModelConfig() instead
- */
-export async function getFirstValidModelConfig(): Promise<{ model: string; provider: ProviderType } | null> {
-    return await getSelectedModelConfig();
-}

apps/server/src/services/llm/context/modules/provider_manager.ts

@@ -1,6 +1,6 @@
 import log from '../../../log.js';
 import { getEmbeddingProvider, getEnabledEmbeddingProviders } from '../../providers/providers.js';
-import { getSelectedEmbeddingProvider } from '../../config/configuration_helpers.js';
+import { getSelectedEmbeddingProvider as getSelectedEmbeddingProviderName } from '../../config/configuration_helpers.js';
 
 /**
  * Manages embedding providers for context services
@@ -12,10 +12,10 @@ export class ProviderManager {
      *
      * @returns The selected embedding provider or null if none available
      */
-    async getPreferredEmbeddingProvider(): Promise<any> {
+    async getSelectedEmbeddingProvider(): Promise<any> {
         try {
             // Get the selected embedding provider
-            const selectedProvider = await getSelectedEmbeddingProvider();
+            const selectedProvider = await getSelectedEmbeddingProviderName();
             
             if (selectedProvider) {
                 const provider = await getEmbeddingProvider(selectedProvider);
@@ -51,7 +51,7 @@ export class ProviderManager {
     async generateQueryEmbedding(query: string): Promise<Float32Array | null> {
         try {
             // Get the preferred embedding provider
-            const provider = await this.getPreferredEmbeddingProvider();
+            const provider = await this.getSelectedEmbeddingProvider();
             if (!provider) {
                 log.error('No embedding provider available');
                 return null;

apps/server/src/services/llm/context/services/context_service.ts

@@ -58,7 +58,7 @@ export class ContextService {
         this.initPromise = (async () => {
             try {
                 // Initialize provider
-                const provider = await providerManager.getPreferredEmbeddingProvider();
+                const provider = await providerManager.getSelectedEmbeddingProvider();
                 if (!provider) {
                     throw new Error(`No embedding provider available. Could not initialize context service.`);
                 }
@@ -224,7 +224,7 @@ export class ContextService {
             log.info(`Final combined results: ${relevantNotes.length} relevant notes`);
 
             // Step 4: Build context from the notes
-            const provider = await providerManager.getPreferredEmbeddingProvider();
+            const provider = await providerManager.getSelectedEmbeddingProvider();
             const providerId = provider?.name || 'default';
 
             const context = await contextFormatter.buildContextFromNotes(

apps/server/src/services/llm/context/services/vector_search_service.ts

@@ -79,7 +79,7 @@ export class VectorSearchService {
             }
 
             // Get provider information
-            const provider = await providerManager.getPreferredEmbeddingProvider();
+            const provider = await providerManager.getSelectedEmbeddingProvider();
             if (!provider) {
                 log.error('No embedding provider available');
                 return [];
@@ -280,7 +280,7 @@ export class VectorSearchService {
             }
 
             // Get provider information
-            const provider = await providerManager.getPreferredEmbeddingProvider();
+            const provider = await providerManager.getSelectedEmbeddingProvider();
             if (!provider) {
                 log.error('No embedding provider available');
                 return [];

apps/server/src/services/llm/embeddings/index.ts

@@ -64,8 +64,6 @@ export const {
 
 export const {
     getEmbeddingStats,
-    reprocessAllNotes,
-    queueNotesForMissingEmbeddings,
     cleanupEmbeddings
 } = stats;
 
@@ -107,8 +105,6 @@ export default {
 
     // Stats and maintenance
     getEmbeddingStats: stats.getEmbeddingStats,
-    reprocessAllNotes: stats.reprocessAllNotes,
-    queueNotesForMissingEmbeddings: stats.queueNotesForMissingEmbeddings,
     cleanupEmbeddings: stats.cleanupEmbeddings,
 
     // Index operations

apps/server/src/services/llm/embeddings/stats.ts

@@ -1,14 +1,5 @@
 import sql from "../../../services/sql.js";
 import log from "../../../services/log.js";
-import indexService from '../index_service.js';
-
-/**
- * Reprocess all notes to update embeddings
- * @deprecated Use indexService.reprocessAllNotes() directly instead
- */
-export async function reprocessAllNotes() {
-    return indexService.reprocessAllNotes();
-}
 
 /**
  * Get current embedding statistics
@@ -64,14 +55,6 @@ export async function getEmbeddingStats() {
     };
 }
 
-/**
- * Queue notes that don't have embeddings for current provider settings
- * @deprecated Use indexService.queueNotesForMissingEmbeddings() directly instead
- */
-export async function queueNotesForMissingEmbeddings() {
-    return indexService.queueNotesForMissingEmbeddings();
-}
-
 /**
  * Cleanup function to remove stale or unused embeddings
  */

apps/server/src/services/llm/embeddings/storage.ts

@@ -11,7 +11,7 @@ import { SEARCH_CONSTANTS } from '../constants/search_constants.js';
 import type { NoteEmbeddingContext } from "./embeddings_interface.js";
 import becca from "../../../becca/becca.js";
 import { isNoteExcludedFromAIById } from "../utils/ai_exclusion_utils.js";
-import { getEmbeddingProviderPrecedence } from '../config/configuration_helpers.js';
+import { getSelectedEmbeddingProvider } from '../config/configuration_helpers.js';
 
 interface Similarity {
     noteId: string;
@@ -277,9 +277,10 @@ export async function findSimilarNotes(
                         log.info('No embeddings found for specified provider, trying fallback providers...');
 
                         // Use the new configuration system - no string parsing!
-                        const preferredProviders = await getEmbeddingProviderPrecedence();
+                        const selectedProvider = await getSelectedEmbeddingProvider();
+                        const preferredProviders = selectedProvider ? [selectedProvider] : [];
 
-                        log.info(`Using provider precedence: ${preferredProviders.join(', ')}`);
+                        log.info(`Using selected provider: ${selectedProvider || 'none'}`);
 
                         // Try providers in precedence order
                         for (const provider of preferredProviders) {

apps/server/src/services/llm/index_service.ts

@@ -266,7 +266,7 @@ export class IndexService {
                 this.indexRebuildTotal = totalNotes;
 
                 log.info("No embeddings found, starting full embedding generation first");
-                await vectorStore.reprocessAllNotes();
+                await this.reprocessAllNotes();
                 log.info("Full embedding generation initiated");
             } else {
                 // For index rebuild, use the number of embeddings as the total
@@ -293,7 +293,7 @@ export class IndexService {
                     // Only start indexing if we're below 90% completion or if embeddings exist but need optimization
                     if (stats.percentComplete < 90) {
                         log.info("Embedding coverage below 90%, starting full embedding generation");
-                        await vectorStore.reprocessAllNotes();
+                        await this.reprocessAllNotes();
                         log.info("Full embedding generation initiated");
                     } else {
                         log.info(`Embedding coverage at ${stats.percentComplete}%, starting index optimization`);

apps/server/src/services/llm/interfaces/ai_service_interfaces.ts

@@ -30,7 +30,7 @@ export interface AIServiceManagerConfig {
 export interface IAIServiceManager {
   getService(provider?: string): Promise<AIService>;
   getAvailableProviders(): string[];
-  getPreferredProvider(): string;
+  getSelectedProvider(): string;
   isProviderAvailable(provider: string): boolean;
   getProviderMetadata(provider: string): ProviderMetadata | null;
   getAIEnabled(): boolean;

apps/server/src/services/llm/pipeline/stages/semantic_context_extraction_stage.ts

@@ -50,7 +50,7 @@ export class SemanticContextExtractionStage extends BasePipelineStage<SemanticCo
             }
 
             // Step 2: Format search results into a context string
-            const provider = await providerManager.getPreferredEmbeddingProvider();
+            const provider = await providerManager.getSelectedEmbeddingProvider();
             const providerId = provider?.name || 'default';
 
             const context = await contextFormatter.buildContextFromNotes(

apps/server/src/services/llm/providers/providers.ts

@@ -328,17 +328,6 @@ export async function getEmbeddingProviderConfigs() {
     return await sql.getRows("SELECT * FROM embedding_providers ORDER BY priority DESC");
 }
 
-/**
- * Initialize the default embedding providers
- * @deprecated - Use on-demand provider creation instead
- */
-export async function initializeDefaultProviders() {
-    // This function is now deprecated in favor of on-demand provider creation
-    // The createProvidersFromCurrentOptions() function should be used instead
-    log.info('initializeDefaultProviders called - using on-demand provider creation instead');
-    return await createProvidersFromCurrentOptions();
-}
-
 export default {
     registerEmbeddingProvider,
     unregisterEmbeddingProvider,
@@ -350,8 +339,7 @@ export default {
     createEmbeddingProviderConfig,
     updateEmbeddingProviderConfig,
     deleteEmbeddingProviderConfig,
-    getEmbeddingProviderConfigs,
+    getEmbeddingProviderConfigs
-    initializeDefaultProviders
 };
 
 /**