* @
feat(ai): add conversation context management and token usage tracking (Phase 4)
## Background
Phase 4 of AI agent loop upgrade: token usage extraction from LLM responses,
cost estimation display, and automatic context window management (summarization
when approaching limits).
## Changes
### Backend (Rust)
- New token_usage::TokenUsage — cross-turn token accumulation
- New agent_loop::maybe_compact — proactive context compaction: token estimation
→ cut-point selection → LLM summarization → fallback summary
→ tool-pair integrity protection → drain/insert
- New agent_loop::context_window_for_model — auto-detect context window by model
- New agent_loop::is_context_length_error — detect upstream context-length errors
- New agent_loop::compact_tool_result_for_context — JSON/text sampling for large results
- Modified agent_loop::run_agent_loop — pre-turn compaction check, retry on context error
- Modified agent_events::AgentEnd — split total_tokens into input_tokens/output_tokens
- New agent_events::ContextCompacted — compaction event with metrics
- New AiConfig.context_window — user-configurable context window size
- New AiChatMessage.kind — context summary message marker
- Modified stream_with_tools & 3 provider stream functions — extract TokenUsage
- Claude: message_start(input) + message_delta(output)
- OpenAI: stream_options { include_usage } → final usage chunk
- Gemini: usageMetadata per chunk, keep last
### Frontend (TypeScript/Vue)
- AiAssistant.vue: token + cost display row, estimateCost() with 7 model families,
contextCompacted UI step, pendingCompaction deferred apply,
messagesForAgentHistory post-compaction history reconstruction
- EditorSettingsDialog.vue: contextWindow input (min=1000, auto-detect)
- settingsStore.ts: AiConfig.contextWindow field
- tauri.ts: AgentEvent type updates, AiChatMessage.kind
- i18n: 6 new keys in en.ts and zh-CN.ts
## Fixed Issues
- messages[0] dropped from LLM context after compaction
- estimate_text_tokens hard threshold causing 2.8x estimate jump
- context_compacted event field semantics mismatch with frontend insert position
- validate_summary English-only keywords causing fallback on non-English LLMs
- Stale agentTokens after conversation switch
@
* remove cost computation logic
118 lines
2.9 KiB
Rust
118 lines
2.9 KiB
Rust
pub mod agent_catalog;
|
|
pub mod agent_connection;
|
|
pub mod agent_events;
|
|
pub mod agent_kv;
|
|
pub mod agent_loop;
|
|
pub mod agent_manager;
|
|
pub mod agent_runtime;
|
|
pub mod agent_service;
|
|
pub mod agent_tools;
|
|
pub mod ai;
|
|
pub mod cloud_sync;
|
|
pub mod connection;
|
|
pub mod connection_secrets;
|
|
pub mod csv_export;
|
|
pub mod data_compare;
|
|
pub mod data_grid_sql;
|
|
pub mod database_capabilities;
|
|
pub mod database_export;
|
|
pub mod database_search_sql;
|
|
pub mod db;
|
|
pub mod db_admin_sql;
|
|
pub mod driver_runtime;
|
|
pub mod external;
|
|
pub mod history;
|
|
pub mod jdbc;
|
|
pub mod models;
|
|
pub mod mongo_ops;
|
|
pub mod object_source_sql;
|
|
pub mod path_utils;
|
|
pub mod plugins;
|
|
pub mod query;
|
|
pub mod query_cancel;
|
|
pub mod query_execution_sql;
|
|
pub mod query_result_sql;
|
|
pub mod redis_ops;
|
|
pub mod saved_sql;
|
|
pub mod schema;
|
|
pub mod schema_diff;
|
|
pub mod sql;
|
|
pub mod sql_analysis;
|
|
pub mod sql_dialect;
|
|
pub mod sql_editability;
|
|
pub mod sql_file_import;
|
|
pub mod sql_risk;
|
|
pub mod sqlite_backup;
|
|
pub mod storage;
|
|
pub mod table_export;
|
|
pub mod table_import;
|
|
pub mod table_structure_sql;
|
|
pub mod text_export;
|
|
pub mod token_usage;
|
|
pub mod transfer;
|
|
pub mod types;
|
|
pub mod update;
|
|
pub mod xlsx_export;
|
|
|
|
pub const R2_CDN_BASE: &str = "https://dl.dbxio.com/";
|
|
|
|
pub fn download_candidate_urls(github_url: &str, r2_path: &str) -> Vec<String> {
|
|
vec![format!("{R2_CDN_BASE}{r2_path}"), github_url.to_string()]
|
|
}
|
|
|
|
use std::pin::Pin;
|
|
|
|
type ResponseFuture = Pin<Box<dyn std::future::Future<Output = Result<reqwest::Response, String>> + Send>>;
|
|
|
|
pub async fn race_download(
|
|
client: &reqwest::Client,
|
|
github_url: &str,
|
|
r2_path: &str,
|
|
user_agent: &str,
|
|
) -> Result<reqwest::Response, String> {
|
|
use futures::future::select_ok;
|
|
|
|
let urls = download_candidate_urls(github_url, r2_path);
|
|
let mut futs: Vec<ResponseFuture> = Vec::with_capacity(urls.len());
|
|
|
|
for url in urls {
|
|
let client = client.clone();
|
|
let ua = user_agent.to_string();
|
|
futs.push(Box::pin(async move {
|
|
client
|
|
.get(&url)
|
|
.header(reqwest::header::USER_AGENT, ua)
|
|
.send()
|
|
.await
|
|
.and_then(|r| r.error_for_status())
|
|
.map_err(|e| format!("{e}"))
|
|
}) as ResponseFuture);
|
|
}
|
|
|
|
match select_ok(futs).await {
|
|
Ok((resp, _)) => Ok(resp),
|
|
Err(last_err) => Err(last_err),
|
|
}
|
|
}
|
|
|
|
#[cfg(test)]
|
|
mod tests {
|
|
use super::download_candidate_urls;
|
|
|
|
#[test]
|
|
fn download_candidates_exclude_third_party_github_proxy() {
|
|
let urls = download_candidate_urls(
|
|
"https://github.com/t8y2/dbx/releases/latest/download/latest.json",
|
|
"releases/latest/latest.json",
|
|
);
|
|
|
|
assert_eq!(
|
|
urls,
|
|
vec![
|
|
"https://dl.dbxio.com/releases/latest/latest.json",
|
|
"https://github.com/t8y2/dbx/releases/latest/download/latest.json",
|
|
]
|
|
);
|
|
}
|
|
}
|