Dockery/Trilium
1
0
Fork 0
mirror of https://github.com/zadam/trilium.git synced 2025-10-26 07:46:30 +01:00
Code Issues Packages Projects Releases Activity

refactor(llm): update chat saving logic to prevent race conditions between client and server

Browse Source
This commit is contained in:
perf3ct
2025-06-02 22:30:59 +00:00
parent ed64a5b4f7
commit 7b498cf384
2 changed files with 54 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

48
apps/client/src/widgets/llm_chat/llm_chat_panel.ts
View File

@@ -951,9 +951,9 @@ export default class LlmChatPanel extends BasicWidget {
}
}
// Save the updated data to the note
this.saveCurrentData()
.catch(err => console.error("Failed to save data after streaming completed:", err));
// DON'T save here - let the server handle saving the complete conversation
// to avoid race conditions between client and server saves
console.log("Updated metadata after streaming completion, server should save");
})
.catch(err => console.error("Error fetching session data after streaming:", err));
}
@@ -991,11 +991,9 @@ export default class LlmChatPanel extends BasicWidget {
console.log(`Cached tool execution for ${toolData.tool} to be saved later`);
// Save immediately after receiving a tool execution
// This ensures we don't lose tool execution data if streaming fails
this.saveCurrentData().catch(err => {
console.error("Failed to save tool execution data:", err);
});
// DON'T save immediately during streaming - let the server handle saving
// to avoid race conditions between client and server saves
console.log(`Tool execution cached, will be saved by server`);
}
},
// Complete handler
@@ -1078,10 +1076,36 @@ export default class LlmChatPanel extends BasicWidget {
// Hide loading indicator
hideLoadingIndicator(this.loadingIndicator);
// Save the final state to the Chat Note
this.saveCurrentData().catch(err => {
console.error("Failed to save assistant response to note:", err);
});
// DON'T save here immediately - let the server save the accumulated response first
// to avoid race conditions. We'll reload the data from the server after a short delay.
console.log("Stream completed, waiting for server to save then reloading data...");
setTimeout(async () => {
try {
console.log("About to reload data from server...");
const currentMessageCount = this.messages.length;
console.log(`Current client message count before reload: ${currentMessageCount}`);
// Reload the data from the server which should have the complete conversation
const reloadSuccess = await this.loadSavedData();
const newMessageCount = this.messages.length;
console.log(`Reload success: ${reloadSuccess}, message count after reload: ${newMessageCount}`);
if (reloadSuccess && newMessageCount > currentMessageCount) {
console.log("Successfully reloaded data with more complete conversation");
} else if (!reloadSuccess) {
console.warn("Reload failed, keeping current client state");
} else {
console.warn("Reload succeeded but message count didn't increase");
}
} catch (error) {
console.error("Failed to reload data after stream completion:", error);
// Fallback: save our current state if reload fails
this.saveCurrentData().catch(err => {
console.error("Failed to save assistant response to note:", err);
});
}
}, 3000); // Wait 3 seconds for server to complete its save
}
// Scroll to bottom

25
apps/server/src/services/llm/chat/rest_chat_service.ts
View File

@@ -150,7 +150,7 @@ class RestChatService {
// Import WebSocket service for streaming
const wsService = await import('../../ws.js');
let accumulatedContent = '';
const accumulatedContentRef = { value: '' };
const pipelineInput: ChatPipelineInput = {
messages: chat.messages.map(msg => ({
@@ -162,8 +162,7 @@ class RestChatService {
showThinking: showThinking,
options: pipelineOptions,
streamCallback: req.method === 'GET' ? (data, done, rawChunk) => {
this.handleStreamCallback(data, done, rawChunk, wsService.default, chatNoteId, res);
if (data) accumulatedContent += data;
this.handleStreamCallback(data, done, rawChunk, wsService.default, chatNoteId, res, accumulatedContentRef);
} : undefined
};
@@ -194,13 +193,15 @@ class RestChatService {
};
} else {
// For streaming, response is already sent via WebSocket/SSE
// Save the accumulated content
if (accumulatedContent) {
// Save the accumulated content - prefer accumulated content over response.text
const finalContent = accumulatedContentRef.value || response.text || '';
if (finalContent) {
chat.messages.push({
role: 'assistant',
content: accumulatedContent
content: finalContent
});
await chatStorageService.updateChat(chat.id, chat.messages, chat.title);
log.info(`Saved accumulated streaming content: ${finalContent.length} characters`);
}
return null;
}
@@ -219,7 +220,8 @@ class RestChatService {
rawChunk: any,
wsService: any,
chatNoteId: string,
res: Response
res: Response,
accumulatedContentRef: { value: string }
) {
const message: LLMStreamMessage = {
type: 'llm-stream',
@@ -229,6 +231,15 @@ class RestChatService {
if (data) {
message.content = data;
// Handle accumulation carefully - if this appears to be a complete response
// (done=true and data is much longer than current accumulated), replace rather than append
if (done && data.length > accumulatedContentRef.value.length && data.includes(accumulatedContentRef.value)) {
// This looks like a complete final response that includes what we've accumulated
accumulatedContentRef.value = data;
} else {
// Normal incremental accumulation
accumulatedContentRef.value += data;
}
}
if (rawChunk && 'thinking' in rawChunk && rawChunk.thinking) {