Files
galaxy/app/src/ai/runtime/event_translator.rs
T

933 lines
33 KiB
Rust

use std::collections::HashMap;
use galaxy_agent_core::{
AgentEvent, ProviderRunOutcome, RuntimeActivity, RuntimeActivityStatus, RuntimeCapabilities,
StopReason, Usage,
};
use uuid::Uuid;
use warp_multi_agent_api::response_event::stream_finished;
use warp_multi_agent_api::{self as api, ClientAction, ResponseEvent};
use super::provider_run_coordinator::ProviderRunProjection;
use crate::ai::agent::runtime_activity;
use crate::ai::openai::response_translator::{build_stream_finished, StreamUsage};
use crate::ai::provider::response_translator::{
build_add_agent_output_message, build_append_text, build_create_task, build_stream_init,
build_user_query_message,
};
#[derive(Clone, Debug, serde::Serialize, serde::Deserialize)]
pub(crate) struct RuntimeResponseConfig {
pub(crate) task_id: String,
pub(crate) conversation_id: String,
pub(crate) needs_create_task: bool,
pub(crate) user_query: Option<String>,
pub(crate) model_id: String,
pub(crate) max_context_tokens: Option<u32>,
pub(crate) capabilities: RuntimeCapabilities,
pub(crate) empty_output_message: Option<String>,
/// Todo items supplied by the existing task transcript, when one is available.
#[serde(skip)]
pub(crate) todo_items: Option<Vec<api::TodoItem>>,
}
/// Converts the provider-neutral runtime lifecycle into Galaxy's existing
/// transcript protocol. Runtime implementations never need to know about the
/// protobuf messages consumed by the UI.
pub(crate) struct RuntimeResponseTranslator {
config: RuntimeResponseConfig,
request_id: String,
initialized: bool,
text_message_id: Option<String>,
reasoning_message_id: Option<String>,
activity_message_ids: HashMap<String, String>,
activities: HashMap<String, RuntimeActivity>,
has_visible_output: bool,
/// Usage for the most recent model call.
usage: Usage,
context_usage: Option<(u64, u64)>,
}
/// Projects a multi-turn provider run into one existing Galaxy response stream.
/// Intermediate model stops remain coordinator-internal; only the run outcome
/// emits the UI's terminal `Finished` event.
pub(crate) struct ProviderRunResponseProjector {
translator: RuntimeResponseTranslator,
has_started_model_turn: bool,
todo_phase: usize,
todo_started: bool,
finished: bool,
}
impl ProviderRunResponseProjector {
pub(crate) fn new(config: RuntimeResponseConfig) -> Self {
let todo_started = config.todo_items.is_some();
Self {
translator: RuntimeResponseTranslator::new(config),
has_started_model_turn: false,
todo_phase: 0,
todo_started,
finished: false,
}
}
pub(crate) fn restored(
config: RuntimeResponseConfig,
projection_was_initialized: bool,
) -> Self {
Self {
translator: RuntimeResponseTranslator::restored(config, projection_was_initialized),
has_started_model_turn: false,
// Task-list events are part of the already persisted projection.
todo_phase: usize::MAX,
todo_started: true,
finished: false,
}
}
pub(crate) fn project(
&mut self,
projection: ProviderRunProjection,
) -> Result<Vec<ResponseEvent>, String> {
if self.finished {
return Err("provider run projection is already finished".to_string());
}
match projection {
ProviderRunProjection::ModelTurnStarted { .. } => {
if self.has_started_model_turn {
self.translator.begin_followup_turn();
}
self.has_started_model_turn = true;
let mut events = self.translator.translate(AgentEvent::TurnStarted {
runtime_request_id: String::new(),
})?;
events.extend(self.todo_phase_events());
Ok(events)
}
ProviderRunProjection::ModelEvent { event, .. } => self.translator.translate(event),
ProviderRunProjection::ModelRetry { .. } => {
Ok(self.translator.discard_failed_turn_output())
}
ProviderRunProjection::ModelTurnRequested { .. }
| ProviderRunProjection::ModelTurnFinished { .. } => Ok(Vec::new()),
ProviderRunProjection::ToolBatchReady { batch } => {
if !self.todo_started {
if let Some(todos) = todos_from_plan_batch(&batch) {
self.todo_started = true;
self.todo_phase = 1;
return Ok(vec![build_todo_update(
&self.translator.config.task_id,
api::message::update_todos::Operation::CreateTodoList(
api::CreateTodoList {
initial_todos: todos,
},
),
)]);
}
return Ok(Vec::new());
}
let todo_index = self.todo_phase.saturating_sub(1);
let Some(todo) = self.todo_items().get(todo_index).cloned() else {
return Ok(Vec::new());
};
self.todo_phase += 1;
let mut events = vec![build_todo_update(
&self.translator.config.task_id,
api::message::update_todos::Operation::MarkTodosCompleted(
api::MarkTodosCompleted {
todo_ids: vec![todo.id],
},
),
)];
events.push(build_todo_update(
&self.translator.config.task_id,
api::message::update_todos::Operation::UpdatePendingTodos(
api::UpdatePendingTodos {
updated_pending_todos: self
.todo_items()
.into_iter()
.skip(self.todo_phase)
.collect(),
},
),
));
Ok(events)
}
}
}
pub(crate) fn set_task_id(&mut self, task_id: impl Into<String>) {
self.translator.set_task_id(task_id);
}
pub(crate) fn finish(
&mut self,
outcome: &ProviderRunOutcome,
aggregate_usage: &Usage,
) -> Result<Vec<ResponseEvent>, String> {
if self.finished {
return Err("provider run projection is already finished".to_string());
}
self.finished = true;
match outcome {
ProviderRunOutcome::Completed(completion) => {
let mut events = self.todo_completion_events();
events.extend(
self.translator
.finish_provider_run(completion.stop_reason.clone(), aggregate_usage),
);
Ok(events)
}
ProviderRunOutcome::Failed(failure) => Ok(self.translator.provider_failure(
&failure.message,
failure.source.as_ref(),
aggregate_usage,
)),
ProviderRunOutcome::Cancelled { .. } => Ok(self
.translator
.finish_provider_run(StopReason::Cancelled, aggregate_usage)),
}
}
// Keep the direct-provider workflow visible in the existing task list protocol. These are
// response events, so the normal history model remains the sole owner of task-list state.
fn todo_phase_events(&mut self) -> Vec<ResponseEvent> {
Vec::new()
}
fn todo_completion_events(&self) -> Vec<ResponseEvent> {
if self.todo_phase == 0 || self.todo_phase == usize::MAX {
return Vec::new();
}
let todos = self.todo_items();
if todos.is_empty() {
return Vec::new();
}
vec![
build_todo_update(
&self.translator.config.task_id,
api::message::update_todos::Operation::MarkTodosCompleted(
api::MarkTodosCompleted {
todo_ids: todos.iter().map(|todo| todo.id.clone()).collect(),
},
),
),
build_todo_update(
&self.translator.config.task_id,
api::message::update_todos::Operation::UpdatePendingTodos(
api::UpdatePendingTodos {
updated_pending_todos: Vec::new(),
},
),
),
]
}
fn todo_items(&self) -> Vec<api::TodoItem> {
self.translator
.config
.todo_items
.clone()
.unwrap_or_default()
}
}
fn todos_from_plan_batch(
batch: &galaxy_agent_core::PendingToolBatch,
) -> Option<Vec<api::TodoItem>> {
let plan_call = batch.calls.iter().find(|pending| {
matches!(
pending.call.name.as_str(),
"create_plan" | "create_documents"
)
})?;
let documents = plan_call.call.arguments.get("documents")?.as_array()?;
let content = documents.first()?.get("content")?.as_str()?;
let section = content
.split_once("## Tasks")
.or_else(|| content.split_once("## Implementation Tasks"))
.map(|(_, section)| section)
.unwrap_or(content);
let todos = section
.lines()
.filter_map(|line| {
let item = line
.trim()
.strip_prefix("- [ ]")
.or_else(|| line.trim().strip_prefix("-"))?
.trim();
if item.is_empty() {
return None;
}
let title = item
.split_once(" - ")
.map_or(item, |(title, _)| title)
.trim();
let id = format!(
"plan-{}",
title
.chars()
.filter_map(|character| character
.is_ascii_alphanumeric()
.then_some(character.to_ascii_lowercase()))
.collect::<String>()
);
Some(api::TodoItem {
id,
title: title.to_owned(),
description: item.to_owned(),
})
})
.take(50)
.collect::<Vec<_>>();
(!todos.is_empty()).then_some(todos)
}
fn build_todo_update(
task_id: &str,
operation: api::message::update_todos::Operation,
) -> ResponseEvent {
let message = api::Message {
id: Uuid::new_v4().to_string(),
task_id: task_id.to_owned(),
request_id: String::new(),
timestamp: None,
server_message_data: String::new(),
citations: Vec::new(),
fetched_memories: Vec::new(),
message: Some(api::message::Message::UpdateTodos(
api::message::UpdateTodos {
operation: Some(operation),
},
)),
};
ResponseEvent {
r#type: Some(api::response_event::Type::ClientActions(
api::response_event::ClientActions {
actions: vec![api::ClientAction {
action: Some(api::client_action::Action::AddMessagesToTask(
api::client_action::AddMessagesToTask {
task_id: task_id.to_owned(),
messages: vec![message],
},
)),
}],
},
)),
}
}
impl RuntimeResponseTranslator {
pub(crate) fn new(config: RuntimeResponseConfig) -> Self {
Self::with_initialization(config, false)
}
pub(crate) fn restored(
mut config: RuntimeResponseConfig,
projection_was_initialized: bool,
) -> Self {
// The task and exchange already exist in restored history. If its output was never
// initialized, replay only the stream Init rather than duplicating task/input messages.
config.needs_create_task = false;
config.user_query = None;
Self::with_initialization(config, projection_was_initialized)
}
fn with_initialization(config: RuntimeResponseConfig, initialized: bool) -> Self {
Self {
config,
request_id: Uuid::new_v4().to_string(),
initialized,
text_message_id: None,
reasoning_message_id: None,
activity_message_ids: HashMap::new(),
activities: HashMap::new(),
has_visible_output: false,
usage: Usage::default(),
context_usage: None,
}
}
pub(crate) fn translate(&mut self, event: AgentEvent) -> Result<Vec<ResponseEvent>, String> {
let mut events = Vec::new();
match event {
AgentEvent::TurnStarted { .. } => self.initialize(&mut events),
AgentEvent::KeepAlive | AgentEvent::ToolCallProgress { .. } => {}
AgentEvent::TextDelta { text } => {
self.initialize(&mut events);
self.add_or_append_text(&text, &mut events);
}
AgentEvent::ReasoningDelta { text } => {
self.initialize(&mut events);
self.add_or_append_reasoning(&text, &mut events);
}
AgentEvent::ReasoningCompleted { text, .. } => {
self.initialize(&mut events);
self.complete_reasoning(&text, &mut events);
}
AgentEvent::RuntimeActivityUpdated { activity } => {
if self.config.capabilities.host_tool_execution {
return Err(
"a host-tool runtime emitted runtime-owned tool activity".to_owned()
);
}
self.initialize(&mut events);
self.text_message_id = None;
self.upsert_runtime_activity(activity, &mut events)?;
}
AgentEvent::ContextUsageUpdated {
used_tokens,
context_size,
} => {
if self.config.capabilities.host_managed_history {
return Err(
"a host-history runtime reported session context occupancy".to_owned()
);
}
self.context_usage = Some((used_tokens, context_size));
}
AgentEvent::UserInputAccepted { text } => {
if !self.config.capabilities.steering {
return Err("a non-steerable runtime accepted live user input".to_owned());
}
self.initialize(&mut events);
events.push(build_user_query_message(&self.config.task_id, &text));
self.text_message_id = None;
self.reasoning_message_id = None;
}
AgentEvent::RuntimeNotice { message } => {
self.initialize(&mut events);
self.text_message_id = None;
self.add_or_append_text(&message, &mut events);
self.text_message_id = None;
}
AgentEvent::UsageUpdated { usage } => {
if !self.config.capabilities.host_managed_history {
return Err("a session runtime reported provider request usage".to_owned());
}
self.usage = usage;
}
AgentEvent::TurnStopped { reason } => {
self.initialize(&mut events);
if !self.has_visible_output && reason != StopReason::Cancelled {
if let Some(message) = self.config.empty_output_message.clone() {
self.add_or_append_text(&message, &mut events);
}
}
events.push(self.finished(reason));
}
AgentEvent::Tool { .. } => {
let owner = if self.config.capabilities.host_tool_execution {
"host-tool runtime emitted an unhandled tool lifecycle event"
} else {
"session runtime handed runtime-owned tool execution to Galaxy"
};
return Err(owner.to_owned());
}
}
Ok(events)
}
pub(crate) fn startup_error(&mut self, message: &str) -> Vec<ResponseEvent> {
let mut events = Vec::new();
self.initialize(&mut events);
self.text_message_id = None;
self.add_or_append_text(message, &mut events);
events.push(self.finished(StopReason::Refusal));
events
}
fn finish_provider_run(
&mut self,
reason: StopReason,
aggregate_usage: &Usage,
) -> Vec<ResponseEvent> {
let mut events = Vec::new();
self.initialize(&mut events);
if !self.has_visible_output && reason != StopReason::Cancelled {
if let Some(message) = self.config.empty_output_message.clone() {
self.add_or_append_text(&message, &mut events);
}
}
events.push(self.finished_with_usage(map_stop_reason(reason), aggregate_usage));
events
}
pub(crate) fn begin_followup_turn(&mut self) {
self.text_message_id = None;
self.reasoning_message_id = None;
self.usage = Usage::default();
}
fn discard_failed_turn_output(&mut self) -> Vec<ResponseEvent> {
let mut events = Vec::new();
self.usage = Usage::default();
if let Some(message_id) = self.text_message_id.take() {
events.push(build_replace_text_message(
&self.config.task_id,
&message_id,
"",
));
}
if let Some(message_id) = self.reasoning_message_id.take() {
events.push(build_replace_reasoning_message(
&self.config.task_id,
&message_id,
"",
));
}
events
}
pub(crate) fn set_task_id(&mut self, task_id: impl Into<String>) {
let task_id = task_id.into();
if self.config.task_id == task_id {
return;
}
self.config.task_id = task_id;
self.text_message_id = None;
self.reasoning_message_id = None;
self.activity_message_ids.clear();
self.activities.clear();
}
fn initialize(&mut self, events: &mut Vec<ResponseEvent>) {
if self.initialized {
return;
}
events.push(build_stream_init(
&self.request_id,
&self.config.conversation_id,
));
if self.config.needs_create_task {
events.push(build_create_task(&self.config.task_id));
}
if let Some(user_query) = &self.config.user_query {
events.push(build_user_query_message(&self.config.task_id, user_query));
}
self.initialized = true;
}
fn add_or_append_text(&mut self, text: &str, events: &mut Vec<ResponseEvent>) {
if text.is_empty() {
return;
}
self.has_visible_output = true;
if let Some(message_id) = &self.text_message_id {
events.push(build_append_text(&self.config.task_id, message_id, text));
} else {
let message_id = Uuid::new_v4().to_string();
events.push(build_add_agent_output_message(
&self.config.task_id,
&message_id,
text,
));
self.text_message_id = Some(message_id);
}
}
fn add_or_append_reasoning(&mut self, text: &str, events: &mut Vec<ResponseEvent>) {
if text.is_empty() {
return;
}
self.has_visible_output = true;
if let Some(message_id) = &self.reasoning_message_id {
events.push(build_reasoning_message(
&self.config.task_id,
message_id,
text,
true,
));
} else {
let message_id = Uuid::new_v4().to_string();
events.push(build_reasoning_message(
&self.config.task_id,
&message_id,
text,
false,
));
self.reasoning_message_id = Some(message_id);
}
}
fn complete_reasoning(&mut self, text: &str, events: &mut Vec<ResponseEvent>) {
if let Some(message_id) = &self.reasoning_message_id {
events.push(build_replace_reasoning_message(
&self.config.task_id,
message_id,
text,
));
} else if !text.is_empty() {
self.add_or_append_reasoning(text, events);
}
}
fn upsert_runtime_activity(
&mut self,
activity: RuntimeActivity,
events: &mut Vec<ResponseEvent>,
) -> Result<(), String> {
let activity_id = activity.id.clone();
let merged_activity = self
.activities
.entry(activity_id.clone())
.or_insert_with(|| activity.clone());
if !activity.title.trim().is_empty() {
merged_activity.title = activity.title;
}
if activity.status.is_some() {
merged_activity.status = activity.status;
}
if activity.output.is_some() {
merged_activity.output = activity.output;
}
let server_message_data = runtime_activity::encode(merged_activity)
.map_err(|error| format!("failed to encode runtime activity: {error}"))?;
let fallback_text = runtime_activity_fallback_text(merged_activity);
if let Some(message_id) = self.activity_message_ids.get(&activity_id) {
events.push(build_update_runtime_activity_message(
&self.config.task_id,
message_id,
&fallback_text,
&server_message_data,
));
} else {
let message_id = Uuid::new_v4().to_string();
events.push(build_add_runtime_activity_message(
&self.config.task_id,
&message_id,
&fallback_text,
&server_message_data,
));
self.activity_message_ids.insert(activity_id, message_id);
}
self.has_visible_output = true;
Ok(())
}
fn finished(&self, reason: StopReason) -> ResponseEvent {
self.finished_with_reason(map_stop_reason(reason))
}
fn provider_failure(
&mut self,
message: &str,
source: Option<&galaxy_agent_core::AgentError>,
aggregate_usage: &Usage,
) -> Vec<ResponseEvent> {
let mut events = Vec::new();
self.initialize(&mut events);
let reason = if source
.is_some_and(|error| error.kind == galaxy_agent_core::AgentErrorKind::Authentication)
{
stream_finished::Reason::InvalidApiKey(stream_finished::InvalidApiKey {
provider: warp_multi_agent_api::LlmProvider::AwsBedrock as i32,
model_name: self.config.model_id.clone(),
})
} else {
stream_finished::Reason::InternalError(stream_finished::InternalError {
message: message.to_owned(),
})
};
events.push(self.finished_with_usage(reason, aggregate_usage));
events
}
fn finished_with_reason(&self, reason: stream_finished::Reason) -> ResponseEvent {
self.finished_with_usage(reason, &self.usage)
}
fn finished_with_usage(
&self,
reason: stream_finished::Reason,
aggregate_usage: &Usage,
) -> ResponseEvent {
if !self.config.capabilities.host_managed_history {
let (used_tokens, context_size) = self.context_usage.unwrap_or_default();
return build_context_finished(
reason,
&self.config.model_id,
used_tokens,
context_size,
);
}
build_stream_finished(
reason,
StreamUsage {
input_tokens: saturating_i32(aggregate_usage.input_tokens),
output_tokens: saturating_i32(aggregate_usage.output_tokens),
cache_read_tokens: saturating_i32(aggregate_usage.cached_input_tokens),
cache_write_tokens: saturating_i32(aggregate_usage.cache_creation_input_tokens),
current_context_tokens: Some(saturating_i32(
self.usage
.input_tokens
.saturating_add(self.usage.cached_input_tokens)
.saturating_add(self.usage.cache_creation_input_tokens),
)),
cost_in_cents: 0.0,
model_id: self.config.model_id.clone(),
max_context_tokens: self.config.max_context_tokens,
},
)
}
}
fn build_reasoning_message(
task_id: &str,
message_id: &str,
text: &str,
append: bool,
) -> ResponseEvent {
let message = api::Message {
id: message_id.to_owned(),
task_id: task_id.to_owned(),
request_id: String::new(),
timestamp: None,
server_message_data: String::new(),
citations: Vec::new(),
fetched_memories: Vec::new(),
message: Some(api::message::Message::AgentReasoning(
api::message::AgentReasoning {
reasoning: text.to_owned(),
finished_duration: None,
},
)),
};
let action = if append {
api::client_action::Action::AppendToMessageContent(
api::client_action::AppendToMessageContent {
task_id: task_id.to_owned(),
message: Some(message),
mask: Some(prost_types::FieldMask {
paths: vec!["agent_reasoning.reasoning".to_owned()],
}),
},
)
} else {
api::client_action::Action::AddMessagesToTask(api::client_action::AddMessagesToTask {
task_id: task_id.to_owned(),
messages: vec![message],
})
};
runtime_client_action(action)
}
fn build_replace_reasoning_message(task_id: &str, message_id: &str, text: &str) -> ResponseEvent {
let message = api::Message {
id: message_id.to_owned(),
task_id: task_id.to_owned(),
request_id: String::new(),
timestamp: None,
server_message_data: String::new(),
citations: Vec::new(),
fetched_memories: Vec::new(),
message: Some(api::message::Message::AgentReasoning(
api::message::AgentReasoning {
reasoning: text.to_owned(),
finished_duration: None,
},
)),
};
runtime_client_action(api::client_action::Action::UpdateTaskMessage(
api::client_action::UpdateTaskMessage {
task_id: task_id.to_owned(),
message: Some(message),
mask: Some(prost_types::FieldMask {
paths: vec!["agent_reasoning.reasoning".to_owned()],
}),
},
))
}
fn build_replace_text_message(task_id: &str, message_id: &str, text: &str) -> ResponseEvent {
runtime_client_action(api::client_action::Action::UpdateTaskMessage(
api::client_action::UpdateTaskMessage {
task_id: task_id.to_owned(),
message: Some(api::Message {
id: message_id.to_owned(),
task_id: task_id.to_owned(),
request_id: String::new(),
timestamp: None,
server_message_data: String::new(),
citations: Vec::new(),
fetched_memories: Vec::new(),
message: Some(api::message::Message::AgentOutput(
api::message::AgentOutput {
text: text.to_owned(),
},
)),
}),
mask: Some(prost_types::FieldMask {
paths: vec!["agent_output.text".to_owned()],
}),
},
))
}
fn runtime_activity_fallback_text(activity: &RuntimeActivity) -> String {
let title = &activity.title;
let status = activity.status.as_ref().map(|status| match status {
RuntimeActivityStatus::Pending => "waiting",
RuntimeActivityStatus::InProgress => "running",
RuntimeActivityStatus::Completed => "completed",
RuntimeActivityStatus::Failed => "failed",
RuntimeActivityStatus::Other(_) => "updated",
});
let mut text = match status {
Some(status) => format!("> **{title}** — {status}"),
None => format!("> **{title}**"),
};
if let Some(output) = &activity.output {
text.push_str("\n\n");
for line in output.lines() {
text.push_str(" ");
text.push_str(line);
text.push('\n');
}
}
text
}
fn build_add_runtime_activity_message(
task_id: &str,
message_id: &str,
fallback_text: &str,
server_message_data: &str,
) -> ResponseEvent {
let message = runtime_activity_message(task_id, message_id, fallback_text, server_message_data);
runtime_client_action(api::client_action::Action::AddMessagesToTask(
api::client_action::AddMessagesToTask {
task_id: task_id.to_owned(),
messages: vec![message],
},
))
}
fn build_update_runtime_activity_message(
task_id: &str,
message_id: &str,
fallback_text: &str,
server_message_data: &str,
) -> ResponseEvent {
let message = runtime_activity_message(task_id, message_id, fallback_text, server_message_data);
runtime_client_action(api::client_action::Action::UpdateTaskMessage(
api::client_action::UpdateTaskMessage {
task_id: task_id.to_owned(),
message: Some(message),
mask: Some(prost_types::FieldMask {
paths: vec![
"agent_output.text".to_owned(),
"server_message_data".to_owned(),
],
}),
},
))
}
fn runtime_activity_message(
task_id: &str,
message_id: &str,
fallback_text: &str,
server_message_data: &str,
) -> api::Message {
api::Message {
id: message_id.to_owned(),
task_id: task_id.to_owned(),
request_id: String::new(),
timestamp: None,
server_message_data: server_message_data.to_owned(),
citations: Vec::new(),
fetched_memories: Vec::new(),
message: Some(api::message::Message::AgentOutput(
api::message::AgentOutput {
text: fallback_text.to_owned(),
},
)),
}
}
fn runtime_client_action(action: api::client_action::Action) -> ResponseEvent {
ResponseEvent {
r#type: Some(api::response_event::Type::ClientActions(
api::response_event::ClientActions {
actions: vec![ClientAction {
action: Some(action),
}],
},
)),
}
}
fn build_context_finished(
reason: stream_finished::Reason,
model_id: &str,
used_tokens: u64,
context_size: u64,
) -> ResponseEvent {
let total_input_tokens = u32::try_from(used_tokens).unwrap_or(u32::MAX);
let context_window_usage = if context_size == 0 {
0.0
} else {
(used_tokens as f32 / context_size as f32).clamp(0.0, 1.0)
};
#[allow(deprecated)]
let usage_metadata = stream_finished::ConversationUsageMetadata {
context_window_usage,
summarized: false,
credits_spent: 0.0,
platform_credits_spent: 0.0,
total_input_tokens,
token_usage: Vec::new(),
tool_usage_metadata: None,
warp_token_usage: HashMap::new(),
byok_token_usage: HashMap::new(),
custom_endpoint_token_usage: HashMap::new(),
context_window_segments: Vec::new(),
};
ResponseEvent {
r#type: Some(api::response_event::Type::Finished(
api::response_event::StreamFinished {
reason: Some(reason),
token_usage: vec![stream_finished::TokenUsage {
model_id: model_id.to_owned(),
// Session runtimes report current occupancy, not tokens
// consumed by this individual request.
total_input: 0,
output: 0,
input_cache_read: 0,
input_cache_write: 0,
cost_in_cents: 0.0,
}],
should_refresh_model_config: false,
request_cost: None,
conversation_usage_metadata: Some(usage_metadata),
},
)),
}
}
pub(crate) fn map_stop_reason(reason: StopReason) -> stream_finished::Reason {
match reason {
StopReason::Completed | StopReason::Cancelled => {
stream_finished::Reason::Done(stream_finished::Done {})
}
StopReason::MaxTokens => {
stream_finished::Reason::MaxTokenLimit(stream_finished::ReachedMaxTokenLimit {})
}
StopReason::ContextWindowExceeded => stream_finished::Reason::ContextWindowExceeded(
stream_finished::ContextWindowExceeded {},
),
StopReason::Refusal | StopReason::ToolLoopLimit | StopReason::Other(_) => {
stream_finished::Reason::Other(stream_finished::Other {})
}
}
}
pub(crate) fn saturating_i32(value: u64) -> i32 {
i32::try_from(value).unwrap_or(i32::MAX)
}
#[cfg(test)]
#[path = "event_translator_tests.rs"]
mod tests;