@retrivora-ai/rag-engine 2.0.4 → 2.0.6

package/src/llm/providers/OllamaProvider.ts

@@ -24,9 +24,11 @@ export class OllamaProvider implements ILLMProvider {
   private readonly embeddingConfig?: EmbeddingConfig;
 
   constructor(llmConfig: LLMConfig, embeddingConfig?: EmbeddingConfig) {
-    const baseURL = llmConfig.baseUrl ?? 'http://localhost:11434';
+    const rawBaseURL = (llmConfig.baseUrl ?? 'http://localhost:11434').replace(/\/+$/, '');
+    const baseURL = rawBaseURL.replace(/\/(?:api\/v1|v1|api)$/i, '');
     const timeout = Number(llmConfig.options?.timeout) || 300_000;
-    this.http = axios.create({ baseURL, timeout });
+    const headers = (llmConfig.options?.headers as Record<string, string>) || {};
+    this.http = axios.create({ baseURL, timeout, headers });
     this.llmConfig = llmConfig;
     this.embeddingConfig = embeddingConfig;
   }

package/src/llm/providers/UniversalLLMAdapter.ts

@@ -94,6 +94,27 @@ export class UniversalLLMAdapter implements ILLMProvider {
       };
     }
 
+    const isSelfHost = this.baseUrl.includes('retrivora.com') || this.baseUrl.includes('localhost');
+    if (isSelfHost || Boolean(process.env.VERCEL)) {
+      try {
+        const _g = globalThis as any;
+        if (typeof _g.__retrivoraDispatchChat === 'function') {
+          const res = await _g.__retrivoraDispatchChat({
+            model: this.model,
+            messages: formattedMessages as any,
+            max_tokens: this.maxTokens,
+            temperature: this.temperature,
+          }, this.apiKey);
+
+          if (res?.response?.choices?.[0]?.message?.content) {
+            return res.response.choices[0].message.content;
+          }
+        }
+      } catch {
+        /* proceed to HTTP fetch */
+      }
+    }
+
     const { data } = await this.http.post(path, payload);
 
     const extractPath = this.opts.responseExtractPath ?? 'choices[0].message.content';
@@ -145,22 +166,52 @@ export class UniversalLLMAdapter implements ILLMProvider {
       };
     }
 
-    const response = await fetch(url, {
-      method: 'POST',
-      headers: this.resolvedHeaders,
-      body: JSON.stringify(payload),
-    });
-
-    if (!response.ok) {
-      const errorText = await response.text().catch(() => response.statusText);
-      throw new Error(`[UniversalLLMAdapter] Streaming request failed (${response.status}): ${errorText}`);
+    const isSelfHost = this.baseUrl.includes('retrivora.com') || this.baseUrl.includes('localhost');
+    let streamBody: ReadableStream<Uint8Array> | null = null;
+
+    if (isSelfHost || Boolean(process.env.VERCEL)) {
+      try {
+        const _g = globalThis as any;
+        if (typeof _g.__retrivoraDispatchChat === 'function') {
+          const res = await _g.__retrivoraDispatchChat({
+            model: this.model,
+            messages: formattedMessages as any,
+            max_tokens: this.maxTokens,
+            temperature: this.temperature,
+            stream: true,
+          }, this.apiKey);
+
+          if (res?.stream) {
+            streamBody = res.stream as ReadableStream<Uint8Array>;
+          } else if (res?.response?.choices?.[0]?.message?.content) {
+            yield res.response.choices[0].message.content;
+            return;
+          }
+        }
+      } catch {
+        /* proceed to HTTP fetch */
+      }
     }
 
-    if (!response.body) {
-      throw new Error('[UniversalLLMAdapter] Response body is null — server did not send a streaming response.');
+    if (!streamBody) {
+      const response = await fetch(url, {
+        method: 'POST',
+        headers: this.resolvedHeaders,
+        body: JSON.stringify(payload),
+      });
+
+      if (!response.ok) {
+        const errorText = await response.text().catch(() => response.statusText);
+        throw new Error(`[UniversalLLMAdapter] Streaming request failed (${response.status}): ${errorText}`);
+      }
+
+      if (!response.body) {
+        throw new Error('[UniversalLLMAdapter] Response body is null — server did not send a streaming response.');
+      }
+      streamBody = response.body;
     }
 
-    const reader = response.body.getReader();
+    const reader = streamBody.getReader();
     const decoder = new TextDecoder('utf-8');
     let buffer = '';
 
@@ -218,6 +269,25 @@ export class UniversalLLMAdapter implements ILLMProvider {
       };
     }
 
+    const isSelfHost = this.baseUrl.includes('retrivora.com') || this.baseUrl.includes('localhost');
+    if (isSelfHost || Boolean(process.env.VERCEL)) {
+      try {
+        const _g = globalThis as any;
+        if (typeof _g.__retrivoraDispatchEmbedding === 'function') {
+          const res = await _g.__retrivoraDispatchEmbedding({
+            model: this.model,
+            input: text,
+          }, this.apiKey);
+
+          if (res?.data?.[0]?.embedding) {
+            return res.data[0].embedding;
+          }
+        }
+      } catch {
+        /* proceed to HTTP fetch */
+      }
+    }
+
     const { data } = await this.http.post(path, payload);
 
     const extractPath = this.opts.embedExtractPath ?? 'data[0].embedding';