From 6c438f7a4da234028e97dad38d4ab8f2e8514f31 Mon Sep 17 00:00:00 2001
From: Florian Egger <flo.egger@icloud.com>
Date: Tue, 19 May 2026 21:24:38 +0200
Subject: [PATCH] fix: remove artificial MAX_STREAM_CHUNKS limit that cut off
 long responses

The plugin hard-capped streaming responses at 1000 chunks. For large
models like qwen2.5:32b generating detailed answers, this limit was
easily exceeded, causing the response to stop mid-sentence.

The Ollama stream already terminates naturally when the model sends the
final done signal, so the artificial chunk limit served no purpose.

- src/chat-view.ts: Remove chunkCount tracking and MAX_STREAM_CHUNKS
  constant. Let the stream run until Ollama signals completion.
---
 main.js          | 6 ------
 src/chat-view.ts | 7 -------
 2 files changed, 13 deletions(-)

diff --git a/main.js b/main.js
index 33fc725..3f6b802 100644
--- a/main.js
+++ b/main.js
@@ -9337,7 +9337,6 @@ ${userMessage}` : userMessage;
       const stream = this.ollamaClient.streamChat(completeMessages, tools);
       let fullResponse = "";
       let toolCalls = [];
-      let chunkCount = 0;
       for await (const chunk of stream) {
         if (chunk.content) {
           fullResponse += chunk.content;
@@ -9350,10 +9349,6 @@ ${userMessage}` : userMessage;
         if (chunk.tool_calls) {
           toolCalls = [...toolCalls, ...chunk.tool_calls];
         }
-        chunkCount++;
-        if (chunkCount > MAX_STREAM_CHUNKS) {
-          break;
-        }
       }
       if (toolCalls.length > 0) {
         await this.processToolCalls(
@@ -9391,7 +9386,6 @@ ${userMessage}` : userMessage;
     }
   }
 };
-var MAX_STREAM_CHUNKS = 1e3;
 var MAX_TOOL_CALLS = 5;
 
 // src/constants.ts
diff --git a/src/chat-view.ts b/src/chat-view.ts
index 7a2f14f..4bbff71 100755
--- a/src/chat-view.ts
+++ b/src/chat-view.ts
@@ -487,7 +487,6 @@ export class ChatView extends ItemView {
 
       let fullResponse = '';
       let toolCalls: OllamaToolCall[] = [];
-      let chunkCount = 0;
 
       for await (const chunk of stream) {
         if (chunk.content) {
@@ -502,11 +501,6 @@ export class ChatView extends ItemView {
         if (chunk.tool_calls) {
           toolCalls = [...toolCalls, ...chunk.tool_calls];
         }
-
-        chunkCount++;
-        if (chunkCount > MAX_STREAM_CHUNKS) {
-          break;
-        }
       }
 
       // Process tool calls if any
@@ -575,5 +569,4 @@ export class ChatView extends ItemView {
   private conversationStateManager: ConversationStateManager;
 }
 
-const MAX_STREAM_CHUNKS = 1000;
 const MAX_TOOL_CALLS = 5;