refactor(hands): 移除空壳 Hand — Whiteboard/Slideshow/Speech (Phase 5)
Some checks failed
CI / Lint & TypeCheck (push) Has been cancelled
CI / Unit Tests (push) Has been cancelled
CI / Build Frontend (push) Has been cancelled
CI / Rust Check (push) Has been cancelled
CI / Security Scan (push) Has been cancelled
CI / E2E Tests (push) Has been cancelled

删除 3 个仅含 UI 占位的 Hand,清理 Rust 实现与前端引用:
- Rust: whiteboard.rs(422行) + slideshow.rs(797行) + speech.rs(442行)
- 前端: WhiteboardCanvas + SlideshowRenderer + speech-synth + 相关类型/常量
- 配置: 3 个 HAND.toml
- 净减 ~5400 行,Hands 9→6(启用) + Quiz/Browser/Researcher/Collector/Clip/Twitter/Reminder
This commit is contained in:
iven
2026-04-17 19:55:59 +08:00
parent 14fa7e150a
commit cb9e48f11d
21 changed files with 11 additions and 3014 deletions

View File

@@ -1,9 +1,6 @@
//! Educational Hands - Teaching and presentation capabilities
//!
//! This module provides hands for interactive classroom experiences:
//! - Whiteboard: Drawing and annotation
//! - Slideshow: Presentation control
//! - Speech: Text-to-speech synthesis
//! This module provides hands for interactive experiences:
//! - Quiz: Assessment and evaluation
//! - Browser: Web automation
//! - Researcher: Deep research and analysis
@@ -11,9 +8,6 @@
//! - Clip: Video processing
//! - Twitter: Social media automation
mod whiteboard;
mod slideshow;
mod speech;
pub mod quiz;
mod browser;
mod researcher;
@@ -22,9 +16,6 @@ mod clip;
mod twitter;
pub mod reminder;
pub use whiteboard::*;
pub use slideshow::*;
pub use speech::*;
pub use quiz::*;
pub use browser::*;
pub use researcher::*;

View File

@@ -1,797 +0,0 @@
//! Slideshow Hand - Presentation control capabilities
//!
//! Provides slideshow control for teaching:
//! - next_slide/prev_slide: Navigation
//! - goto_slide: Jump to specific slide
//! - spotlight: Highlight elements
//! - laser: Show laser pointer
//! - highlight: Highlight areas
//! - play_animation: Trigger animations
use async_trait::async_trait;
use serde::{Deserialize, Serialize};
use serde_json::Value;
use std::sync::Arc;
use tokio::sync::RwLock;
use zclaw_types::Result;
use crate::{Hand, HandConfig, HandContext, HandResult, HandStatus};
/// Slideshow action types
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(tag = "action", rename_all = "snake_case")]
pub enum SlideshowAction {
/// Go to next slide
NextSlide,
/// Go to previous slide
PrevSlide,
/// Go to specific slide
GotoSlide {
slide_number: usize,
},
/// Spotlight/highlight an element
Spotlight {
element_id: String,
#[serde(default = "default_spotlight_duration")]
duration_ms: u64,
},
/// Show laser pointer at position
Laser {
x: f64,
y: f64,
#[serde(default = "default_laser_duration")]
duration_ms: u64,
},
/// Highlight a rectangular area
Highlight {
x: f64,
y: f64,
width: f64,
height: f64,
#[serde(default)]
color: Option<String>,
#[serde(default = "default_highlight_duration")]
duration_ms: u64,
},
/// Play animation
PlayAnimation {
animation_id: String,
},
/// Pause auto-play
Pause,
/// Resume auto-play
Resume,
/// Start auto-play
AutoPlay {
#[serde(default = "default_interval")]
interval_ms: u64,
},
/// Stop auto-play
StopAutoPlay,
/// Get current state
GetState,
/// Set slide content (for dynamic slides)
SetContent {
slide_number: usize,
content: SlideContent,
},
}
fn default_spotlight_duration() -> u64 { 2000 }
fn default_laser_duration() -> u64 { 3000 }
fn default_highlight_duration() -> u64 { 2000 }
fn default_interval() -> u64 { 5000 }
/// Slide content structure
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct SlideContent {
pub title: String,
#[serde(default)]
pub subtitle: Option<String>,
#[serde(default)]
pub content: Vec<ContentBlock>,
#[serde(default)]
pub notes: Option<String>,
#[serde(default)]
pub background: Option<String>,
}
/// Presentation/slideshow rendering content block. Domain-specific for slide content.
/// Distinct from zclaw_types::ContentBlock (LLM messages) and zclaw_protocols::ContentBlock (MCP).
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum ContentBlock {
Text { text: String, style: Option<TextStyle> },
Image { url: String, alt: Option<String> },
List { items: Vec<String>, ordered: bool },
Code { code: String, language: Option<String> },
Math { latex: String },
Table { headers: Vec<String>, rows: Vec<Vec<String>> },
Chart { chart_type: String, data: serde_json::Value },
}
/// Text style options
#[derive(Debug, Clone, Serialize, Deserialize, Default)]
pub struct TextStyle {
#[serde(default)]
pub bold: bool,
#[serde(default)]
pub italic: bool,
#[serde(default)]
pub size: Option<u32>,
#[serde(default)]
pub color: Option<String>,
}
/// Slideshow state
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct SlideshowState {
pub current_slide: usize,
pub total_slides: usize,
pub is_playing: bool,
pub auto_play_interval_ms: u64,
pub slides: Vec<SlideContent>,
}
impl Default for SlideshowState {
fn default() -> Self {
Self {
current_slide: 0,
total_slides: 0,
is_playing: false,
auto_play_interval_ms: 5000,
slides: Vec::new(),
}
}
}
/// Slideshow Hand implementation
pub struct SlideshowHand {
config: HandConfig,
state: Arc<RwLock<SlideshowState>>,
}
impl SlideshowHand {
/// Create a new slideshow hand
pub fn new() -> Self {
Self {
config: HandConfig {
id: "slideshow".to_string(),
name: "幻灯片".to_string(),
description: "控制演示文稿的播放、导航和标注".to_string(),
needs_approval: false,
dependencies: vec![],
input_schema: Some(serde_json::json!({
"type": "object",
"properties": {
"action": { "type": "string" },
"slide_number": { "type": "integer" },
"element_id": { "type": "string" },
}
})),
tags: vec!["presentation".to_string(), "education".to_string()],
enabled: true,
max_concurrent: 0,
timeout_secs: 0,
},
state: Arc::new(RwLock::new(SlideshowState::default())),
}
}
/// Create with slides (async version)
pub async fn with_slides_async(slides: Vec<SlideContent>) -> Self {
let hand = Self::new();
let mut state = hand.state.write().await;
state.total_slides = slides.len();
state.slides = slides;
drop(state);
hand
}
/// Execute a slideshow action
pub async fn execute_action(&self, action: SlideshowAction) -> Result<HandResult> {
let mut state = self.state.write().await;
match action {
SlideshowAction::NextSlide => {
if state.current_slide < state.total_slides.saturating_sub(1) {
state.current_slide += 1;
}
Ok(HandResult::success(serde_json::json!({
"status": "next",
"current_slide": state.current_slide,
"total_slides": state.total_slides,
})))
}
SlideshowAction::PrevSlide => {
if state.current_slide > 0 {
state.current_slide -= 1;
}
Ok(HandResult::success(serde_json::json!({
"status": "prev",
"current_slide": state.current_slide,
"total_slides": state.total_slides,
})))
}
SlideshowAction::GotoSlide { slide_number } => {
if slide_number < state.total_slides {
state.current_slide = slide_number;
Ok(HandResult::success(serde_json::json!({
"status": "goto",
"current_slide": state.current_slide,
"slide_content": state.slides.get(slide_number),
})))
} else {
Ok(HandResult::error(format!("Slide {} out of range", slide_number)))
}
}
SlideshowAction::Spotlight { element_id, duration_ms } => {
Ok(HandResult::success(serde_json::json!({
"status": "spotlight",
"element_id": element_id,
"duration_ms": duration_ms,
})))
}
SlideshowAction::Laser { x, y, duration_ms } => {
Ok(HandResult::success(serde_json::json!({
"status": "laser",
"x": x,
"y": y,
"duration_ms": duration_ms,
})))
}
SlideshowAction::Highlight { x, y, width, height, color, duration_ms } => {
Ok(HandResult::success(serde_json::json!({
"status": "highlight",
"x": x, "y": y,
"width": width, "height": height,
"color": color.unwrap_or_else(|| "#ffcc00".to_string()),
"duration_ms": duration_ms,
})))
}
SlideshowAction::PlayAnimation { animation_id } => {
Ok(HandResult::success(serde_json::json!({
"status": "animation",
"animation_id": animation_id,
})))
}
SlideshowAction::Pause => {
state.is_playing = false;
Ok(HandResult::success(serde_json::json!({
"status": "paused",
})))
}
SlideshowAction::Resume => {
state.is_playing = true;
Ok(HandResult::success(serde_json::json!({
"status": "resumed",
})))
}
SlideshowAction::AutoPlay { interval_ms } => {
state.is_playing = true;
state.auto_play_interval_ms = interval_ms;
Ok(HandResult::success(serde_json::json!({
"status": "autoplay",
"interval_ms": interval_ms,
})))
}
SlideshowAction::StopAutoPlay => {
state.is_playing = false;
Ok(HandResult::success(serde_json::json!({
"status": "stopped",
})))
}
SlideshowAction::GetState => {
Ok(HandResult::success(serde_json::to_value(&*state).unwrap_or(Value::Null)))
}
SlideshowAction::SetContent { slide_number, content } => {
if slide_number < state.slides.len() {
state.slides[slide_number] = content.clone();
Ok(HandResult::success(serde_json::json!({
"status": "content_set",
"slide_number": slide_number,
})))
} else if slide_number == state.slides.len() {
state.slides.push(content);
state.total_slides = state.slides.len();
Ok(HandResult::success(serde_json::json!({
"status": "slide_added",
"slide_number": slide_number,
})))
} else {
Ok(HandResult::error(format!("Invalid slide number: {}", slide_number)))
}
}
}
}
/// Get current state
pub async fn get_state(&self) -> SlideshowState {
self.state.read().await.clone()
}
/// Add a slide
pub async fn add_slide(&self, content: SlideContent) {
let mut state = self.state.write().await;
state.slides.push(content);
state.total_slides = state.slides.len();
}
}
impl Default for SlideshowHand {
fn default() -> Self {
Self::new()
}
}
#[async_trait]
impl Hand for SlideshowHand {
fn config(&self) -> &HandConfig {
&self.config
}
async fn execute(&self, _context: &HandContext, input: Value) -> Result<HandResult> {
let action: SlideshowAction = match serde_json::from_value(input) {
Ok(a) => a,
Err(e) => {
return Ok(HandResult::error(format!("Invalid slideshow action: {}", e)));
}
};
self.execute_action(action).await
}
fn status(&self) -> HandStatus {
HandStatus::Idle
}
}
#[cfg(test)]
mod tests {
use super::*;
use serde_json::json;
// === Config & Defaults ===
#[tokio::test]
async fn test_slideshow_creation() {
let hand = SlideshowHand::new();
assert_eq!(hand.config().id, "slideshow");
assert_eq!(hand.config().name, "幻灯片");
assert!(!hand.config().needs_approval);
assert!(hand.config().enabled);
assert!(hand.config().tags.contains(&"presentation".to_string()));
}
#[test]
fn test_default_impl() {
let hand = SlideshowHand::default();
assert_eq!(hand.config().id, "slideshow");
}
#[test]
fn test_needs_approval() {
let hand = SlideshowHand::new();
assert!(!hand.needs_approval());
}
#[test]
fn test_status() {
let hand = SlideshowHand::new();
assert_eq!(hand.status(), HandStatus::Idle);
}
#[test]
fn test_default_state() {
let state = SlideshowState::default();
assert_eq!(state.current_slide, 0);
assert_eq!(state.total_slides, 0);
assert!(!state.is_playing);
assert_eq!(state.auto_play_interval_ms, 5000);
assert!(state.slides.is_empty());
}
// === Navigation ===
#[tokio::test]
async fn test_navigation() {
let hand = SlideshowHand::with_slides_async(vec![
SlideContent { title: "Slide 1".to_string(), subtitle: None, content: vec![], notes: None, background: None },
SlideContent { title: "Slide 2".to_string(), subtitle: None, content: vec![], notes: None, background: None },
SlideContent { title: "Slide 3".to_string(), subtitle: None, content: vec![], notes: None, background: None },
]).await;
// Next
hand.execute_action(SlideshowAction::NextSlide).await.unwrap();
assert_eq!(hand.get_state().await.current_slide, 1);
// Goto
hand.execute_action(SlideshowAction::GotoSlide { slide_number: 2 }).await.unwrap();
assert_eq!(hand.get_state().await.current_slide, 2);
// Prev
hand.execute_action(SlideshowAction::PrevSlide).await.unwrap();
assert_eq!(hand.get_state().await.current_slide, 1);
}
#[tokio::test]
async fn test_next_slide_at_end() {
let hand = SlideshowHand::with_slides_async(vec![
SlideContent { title: "Only Slide".to_string(), subtitle: None, content: vec![], notes: None, background: None },
]).await;
// At slide 0, should not advance past last slide
hand.execute_action(SlideshowAction::NextSlide).await.unwrap();
assert_eq!(hand.get_state().await.current_slide, 0);
}
#[tokio::test]
async fn test_prev_slide_at_beginning() {
let hand = SlideshowHand::with_slides_async(vec![
SlideContent { title: "Slide 1".to_string(), subtitle: None, content: vec![], notes: None, background: None },
SlideContent { title: "Slide 2".to_string(), subtitle: None, content: vec![], notes: None, background: None },
]).await;
// At slide 0, should not go below 0
hand.execute_action(SlideshowAction::PrevSlide).await.unwrap();
assert_eq!(hand.get_state().await.current_slide, 0);
}
#[tokio::test]
async fn test_goto_slide_out_of_range() {
let hand = SlideshowHand::with_slides_async(vec![
SlideContent { title: "Slide 1".to_string(), subtitle: None, content: vec![], notes: None, background: None },
]).await;
let result = hand.execute_action(SlideshowAction::GotoSlide { slide_number: 5 }).await.unwrap();
assert!(!result.success);
}
#[tokio::test]
async fn test_goto_slide_returns_content() {
let hand = SlideshowHand::with_slides_async(vec![
SlideContent { title: "First".to_string(), subtitle: None, content: vec![], notes: None, background: None },
SlideContent { title: "Second".to_string(), subtitle: None, content: vec![], notes: None, background: None },
]).await;
let result = hand.execute_action(SlideshowAction::GotoSlide { slide_number: 1 }).await.unwrap();
assert!(result.success);
assert_eq!(result.output["slide_content"]["title"], "Second");
}
// === Spotlight & Laser & Highlight ===
#[tokio::test]
async fn test_spotlight() {
let hand = SlideshowHand::new();
let action = SlideshowAction::Spotlight {
element_id: "title".to_string(),
duration_ms: 2000,
};
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
assert_eq!(result.output["element_id"], "title");
assert_eq!(result.output["duration_ms"], 2000);
}
#[tokio::test]
async fn test_spotlight_default_duration() {
let hand = SlideshowHand::new();
let action = SlideshowAction::Spotlight {
element_id: "elem".to_string(),
duration_ms: default_spotlight_duration(),
};
let result = hand.execute_action(action).await.unwrap();
assert_eq!(result.output["duration_ms"], 2000);
}
#[tokio::test]
async fn test_laser() {
let hand = SlideshowHand::new();
let action = SlideshowAction::Laser {
x: 100.0,
y: 200.0,
duration_ms: 3000,
};
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
assert_eq!(result.output["x"], 100.0);
assert_eq!(result.output["y"], 200.0);
}
#[tokio::test]
async fn test_highlight_default_color() {
let hand = SlideshowHand::new();
let action = SlideshowAction::Highlight {
x: 10.0, y: 20.0, width: 100.0, height: 50.0,
color: None, duration_ms: 2000,
};
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
assert_eq!(result.output["color"], "#ffcc00");
}
#[tokio::test]
async fn test_highlight_custom_color() {
let hand = SlideshowHand::new();
let action = SlideshowAction::Highlight {
x: 0.0, y: 0.0, width: 50.0, height: 50.0,
color: Some("#ff0000".to_string()), duration_ms: 1000,
};
let result = hand.execute_action(action).await.unwrap();
assert_eq!(result.output["color"], "#ff0000");
}
// === AutoPlay / Pause / Resume ===
#[tokio::test]
async fn test_autoplay_pause_resume() {
let hand = SlideshowHand::new();
// AutoPlay
let result = hand.execute_action(SlideshowAction::AutoPlay { interval_ms: 3000 }).await.unwrap();
assert!(result.success);
assert!(hand.get_state().await.is_playing);
assert_eq!(hand.get_state().await.auto_play_interval_ms, 3000);
// Pause
hand.execute_action(SlideshowAction::Pause).await.unwrap();
assert!(!hand.get_state().await.is_playing);
// Resume
hand.execute_action(SlideshowAction::Resume).await.unwrap();
assert!(hand.get_state().await.is_playing);
// Stop
hand.execute_action(SlideshowAction::StopAutoPlay).await.unwrap();
assert!(!hand.get_state().await.is_playing);
}
#[tokio::test]
async fn test_autoplay_default_interval() {
let hand = SlideshowHand::new();
hand.execute_action(SlideshowAction::AutoPlay { interval_ms: default_interval() }).await.unwrap();
assert_eq!(hand.get_state().await.auto_play_interval_ms, 5000);
}
// === PlayAnimation ===
#[tokio::test]
async fn test_play_animation() {
let hand = SlideshowHand::new();
let result = hand.execute_action(SlideshowAction::PlayAnimation {
animation_id: "fade_in".to_string(),
}).await.unwrap();
assert!(result.success);
assert_eq!(result.output["animation_id"], "fade_in");
}
// === GetState ===
#[tokio::test]
async fn test_get_state() {
let hand = SlideshowHand::with_slides_async(vec![
SlideContent { title: "A".to_string(), subtitle: None, content: vec![], notes: None, background: None },
]).await;
let result = hand.execute_action(SlideshowAction::GetState).await.unwrap();
assert!(result.success);
assert_eq!(result.output["total_slides"], 1);
assert_eq!(result.output["current_slide"], 0);
}
// === SetContent ===
#[tokio::test]
async fn test_set_content() {
let hand = SlideshowHand::new();
let content = SlideContent {
title: "Test Slide".to_string(),
subtitle: Some("Subtitle".to_string()),
content: vec![ContentBlock::Text {
text: "Hello".to_string(),
style: None,
}],
notes: Some("Speaker notes".to_string()),
background: None,
};
let result = hand.execute_action(SlideshowAction::SetContent {
slide_number: 0,
content,
}).await.unwrap();
assert!(result.success);
assert_eq!(hand.get_state().await.total_slides, 1);
assert_eq!(hand.get_state().await.slides[0].title, "Test Slide");
}
#[tokio::test]
async fn test_set_content_append() {
let hand = SlideshowHand::with_slides_async(vec![
SlideContent { title: "First".to_string(), subtitle: None, content: vec![], notes: None, background: None },
]).await;
let content = SlideContent {
title: "Appended".to_string(), subtitle: None, content: vec![], notes: None, background: None,
};
let result = hand.execute_action(SlideshowAction::SetContent {
slide_number: 1,
content,
}).await.unwrap();
assert!(result.success);
assert_eq!(result.output["status"], "slide_added");
assert_eq!(hand.get_state().await.total_slides, 2);
}
#[tokio::test]
async fn test_set_content_invalid_index() {
let hand = SlideshowHand::new();
let content = SlideContent {
title: "Gap".to_string(), subtitle: None, content: vec![], notes: None, background: None,
};
let result = hand.execute_action(SlideshowAction::SetContent {
slide_number: 5,
content,
}).await.unwrap();
assert!(!result.success);
}
// === Action Deserialization ===
#[test]
fn test_deserialize_next_slide() {
let action: SlideshowAction = serde_json::from_value(json!({"action": "next_slide"})).unwrap();
assert!(matches!(action, SlideshowAction::NextSlide));
}
#[test]
fn test_deserialize_goto_slide() {
let action: SlideshowAction = serde_json::from_value(json!({"action": "goto_slide", "slide_number": 3})).unwrap();
match action {
SlideshowAction::GotoSlide { slide_number } => assert_eq!(slide_number, 3),
_ => panic!("Expected GotoSlide"),
}
}
#[test]
fn test_deserialize_laser() {
let action: SlideshowAction = serde_json::from_value(json!({
"action": "laser", "x": 50.0, "y": 75.0
})).unwrap();
match action {
SlideshowAction::Laser { x, y, .. } => {
assert_eq!(x, 50.0);
assert_eq!(y, 75.0);
}
_ => panic!("Expected Laser"),
}
}
#[test]
fn test_deserialize_autoplay() {
let action: SlideshowAction = serde_json::from_value(json!({"action": "auto_play"})).unwrap();
match action {
SlideshowAction::AutoPlay { interval_ms } => assert_eq!(interval_ms, 5000),
_ => panic!("Expected AutoPlay"),
}
}
#[test]
fn test_deserialize_invalid_action() {
let result = serde_json::from_value::<SlideshowAction>(json!({"action": "nonexistent"}));
assert!(result.is_err());
}
// === ContentBlock Deserialization ===
#[test]
fn test_content_block_text() {
let block: ContentBlock = serde_json::from_value(json!({
"type": "text", "text": "Hello"
})).unwrap();
match block {
ContentBlock::Text { text, style } => {
assert_eq!(text, "Hello");
assert!(style.is_none());
}
_ => panic!("Expected Text"),
}
}
#[test]
fn test_content_block_list() {
let block: ContentBlock = serde_json::from_value(json!({
"type": "list", "items": ["A", "B"], "ordered": true
})).unwrap();
match block {
ContentBlock::List { items, ordered } => {
assert_eq!(items, vec!["A", "B"]);
assert!(ordered);
}
_ => panic!("Expected List"),
}
}
#[test]
fn test_content_block_code() {
let block: ContentBlock = serde_json::from_value(json!({
"type": "code", "code": "fn main() {}", "language": "rust"
})).unwrap();
match block {
ContentBlock::Code { code, language } => {
assert_eq!(code, "fn main() {}");
assert_eq!(language, Some("rust".to_string()));
}
_ => panic!("Expected Code"),
}
}
#[test]
fn test_content_block_table() {
let block: ContentBlock = serde_json::from_value(json!({
"type": "table",
"headers": ["Name", "Age"],
"rows": [["Alice", "30"]]
})).unwrap();
match block {
ContentBlock::Table { headers, rows } => {
assert_eq!(headers, vec!["Name", "Age"]);
assert_eq!(rows, vec![vec!["Alice", "30"]]);
}
_ => panic!("Expected Table"),
}
}
// === Hand trait via execute ===
#[tokio::test]
async fn test_hand_execute_dispatch() {
let hand = SlideshowHand::with_slides_async(vec![
SlideContent { title: "S1".to_string(), subtitle: None, content: vec![], notes: None, background: None },
SlideContent { title: "S2".to_string(), subtitle: None, content: vec![], notes: None, background: None },
]).await;
let ctx = HandContext::default();
let result = hand.execute(&ctx, json!({"action": "next_slide"})).await.unwrap();
assert!(result.success);
assert_eq!(result.output["current_slide"], 1);
}
#[tokio::test]
async fn test_hand_execute_invalid_action() {
let hand = SlideshowHand::new();
let ctx = HandContext::default();
let result = hand.execute(&ctx, json!({"action": "invalid"})).await.unwrap();
assert!(!result.success);
}
// === add_slide helper ===
#[tokio::test]
async fn test_add_slide() {
let hand = SlideshowHand::new();
hand.add_slide(SlideContent {
title: "Dynamic".to_string(), subtitle: None, content: vec![], notes: None, background: None,
}).await;
hand.add_slide(SlideContent {
title: "Dynamic 2".to_string(), subtitle: None, content: vec![], notes: None, background: None,
}).await;
let state = hand.get_state().await;
assert_eq!(state.total_slides, 2);
assert_eq!(state.slides.len(), 2);
}
}

View File

@@ -1,442 +0,0 @@
//! Speech Hand - Text-to-Speech synthesis capabilities
//!
//! Provides speech synthesis for teaching:
//! - speak: Convert text to speech
//! - speak_ssml: Advanced speech with SSML markup
//! - pause/resume/stop: Playback control
//! - list_voices: Get available voices
//! - set_voice: Configure voice settings
use async_trait::async_trait;
use serde::{Deserialize, Serialize};
use serde_json::Value;
use std::sync::Arc;
use tokio::sync::RwLock;
use zclaw_types::Result;
use crate::{Hand, HandConfig, HandContext, HandResult, HandStatus};
/// TTS Provider types
#[derive(Debug, Clone, Serialize, Deserialize, Default)]
#[serde(rename_all = "lowercase")]
pub enum TtsProvider {
#[default]
Browser,
Azure,
OpenAI,
ElevenLabs,
Local,
}
/// Speech action types
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(tag = "action", rename_all = "snake_case")]
pub enum SpeechAction {
/// Speak text
Speak {
text: String,
#[serde(default)]
voice: Option<String>,
#[serde(default = "default_rate")]
rate: f32,
#[serde(default = "default_pitch")]
pitch: f32,
#[serde(default = "default_volume")]
volume: f32,
#[serde(default)]
language: Option<String>,
},
/// Speak with SSML markup
SpeakSsml {
ssml: String,
#[serde(default)]
voice: Option<String>,
},
/// Pause playback
Pause,
/// Resume playback
Resume,
/// Stop playback
Stop,
/// List available voices
ListVoices {
#[serde(default)]
language: Option<String>,
},
/// Set default voice
SetVoice {
voice: String,
#[serde(default)]
language: Option<String>,
},
/// Set provider
SetProvider {
provider: TtsProvider,
#[serde(default)]
api_key: Option<String>,
#[serde(default)]
region: Option<String>,
},
}
fn default_rate() -> f32 { 1.0 }
fn default_pitch() -> f32 { 1.0 }
fn default_volume() -> f32 { 1.0 }
/// Voice information
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct VoiceInfo {
pub id: String,
pub name: String,
pub language: String,
pub gender: String,
#[serde(default)]
pub preview_url: Option<String>,
}
/// Playback state
#[derive(Debug, Clone, Serialize, Deserialize, Default)]
pub enum PlaybackState {
#[default]
Idle,
Playing,
Paused,
}
/// Speech configuration
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct SpeechConfig {
pub provider: TtsProvider,
pub default_voice: Option<String>,
pub default_language: String,
pub default_rate: f32,
pub default_pitch: f32,
pub default_volume: f32,
}
impl Default for SpeechConfig {
fn default() -> Self {
Self {
provider: TtsProvider::Browser,
default_voice: None,
default_language: "zh-CN".to_string(),
default_rate: 1.0,
default_pitch: 1.0,
default_volume: 1.0,
}
}
}
/// Speech state
#[derive(Debug, Clone, Default)]
pub struct SpeechState {
pub config: SpeechConfig,
pub playback: PlaybackState,
pub current_text: Option<String>,
pub position_ms: u64,
pub available_voices: Vec<VoiceInfo>,
}
/// Speech Hand implementation
pub struct SpeechHand {
config: HandConfig,
state: Arc<RwLock<SpeechState>>,
}
impl SpeechHand {
/// Create a new speech hand
pub fn new() -> Self {
Self {
config: HandConfig {
id: "speech".to_string(),
name: "语音合成".to_string(),
description: "文本转语音合成输出".to_string(),
needs_approval: false,
dependencies: vec![],
input_schema: Some(serde_json::json!({
"type": "object",
"properties": {
"action": { "type": "string" },
"text": { "type": "string" },
"voice": { "type": "string" },
"rate": { "type": "number" },
}
})),
tags: vec!["audio".to_string(), "tts".to_string(), "education".to_string(), "demo".to_string()],
enabled: true,
max_concurrent: 0,
timeout_secs: 0,
},
state: Arc::new(RwLock::new(SpeechState {
config: SpeechConfig::default(),
playback: PlaybackState::Idle,
available_voices: Self::get_default_voices(),
..Default::default()
})),
}
}
/// Create with custom provider
pub fn with_provider(provider: TtsProvider) -> Self {
let hand = Self::new();
let mut state = hand.state.blocking_write();
state.config.provider = provider;
drop(state);
hand
}
/// Get default voices
fn get_default_voices() -> Vec<VoiceInfo> {
vec![
VoiceInfo {
id: "zh-CN-XiaoxiaoNeural".to_string(),
name: "Xiaoxiao".to_string(),
language: "zh-CN".to_string(),
gender: "female".to_string(),
preview_url: None,
},
VoiceInfo {
id: "zh-CN-YunxiNeural".to_string(),
name: "Yunxi".to_string(),
language: "zh-CN".to_string(),
gender: "male".to_string(),
preview_url: None,
},
VoiceInfo {
id: "en-US-JennyNeural".to_string(),
name: "Jenny".to_string(),
language: "en-US".to_string(),
gender: "female".to_string(),
preview_url: None,
},
VoiceInfo {
id: "en-US-GuyNeural".to_string(),
name: "Guy".to_string(),
language: "en-US".to_string(),
gender: "male".to_string(),
preview_url: None,
},
]
}
/// Execute a speech action
pub async fn execute_action(&self, action: SpeechAction) -> Result<HandResult> {
let mut state = self.state.write().await;
match action {
SpeechAction::Speak { text, voice, rate, pitch, volume, language } => {
let voice_id = voice.or(state.config.default_voice.clone())
.unwrap_or_else(|| "default".to_string());
let lang = language.unwrap_or_else(|| state.config.default_language.clone());
let actual_rate = if rate == 1.0 { state.config.default_rate } else { rate };
let actual_pitch = if pitch == 1.0 { state.config.default_pitch } else { pitch };
let actual_volume = if volume == 1.0 { state.config.default_volume } else { volume };
state.playback = PlaybackState::Playing;
state.current_text = Some(text.clone());
// Determine TTS method based on provider:
// - Browser: frontend uses Web Speech API (zero deps, works offline)
// - OpenAI: frontend calls speech_tts command (high-quality, needs API key)
// - Others: future support
let tts_method = match state.config.provider {
TtsProvider::Browser => "browser",
TtsProvider::OpenAI => "openai_api",
TtsProvider::Azure => "azure_api",
TtsProvider::ElevenLabs => "elevenlabs_api",
TtsProvider::Local => "local_engine",
};
let estimated_duration_ms = (text.chars().count() as f64 / 5.0 * 1000.0) as u64;
Ok(HandResult::success(serde_json::json!({
"status": "speaking",
"tts_method": tts_method,
"text": text,
"voice": voice_id,
"language": lang,
"rate": actual_rate,
"pitch": actual_pitch,
"volume": actual_volume,
"provider": format!("{:?}", state.config.provider).to_lowercase(),
"duration_ms": estimated_duration_ms,
"instruction": "Frontend should play this via TTS engine"
})))
}
SpeechAction::SpeakSsml { ssml, voice } => {
let voice_id = voice.or(state.config.default_voice.clone())
.unwrap_or_else(|| "default".to_string());
state.playback = PlaybackState::Playing;
state.current_text = Some(ssml.clone());
Ok(HandResult::success(serde_json::json!({
"status": "speaking_ssml",
"ssml": ssml,
"voice": voice_id,
"provider": state.config.provider,
})))
}
SpeechAction::Pause => {
state.playback = PlaybackState::Paused;
Ok(HandResult::success(serde_json::json!({
"status": "paused",
"position_ms": state.position_ms,
})))
}
SpeechAction::Resume => {
state.playback = PlaybackState::Playing;
Ok(HandResult::success(serde_json::json!({
"status": "resumed",
"position_ms": state.position_ms,
})))
}
SpeechAction::Stop => {
state.playback = PlaybackState::Idle;
state.current_text = None;
state.position_ms = 0;
Ok(HandResult::success(serde_json::json!({
"status": "stopped",
})))
}
SpeechAction::ListVoices { language } => {
let voices: Vec<_> = state.available_voices.iter()
.filter(|v| {
language.as_ref()
.map(|l| v.language.starts_with(l))
.unwrap_or(true)
})
.cloned()
.collect();
Ok(HandResult::success(serde_json::json!({
"voices": voices,
"count": voices.len(),
})))
}
SpeechAction::SetVoice { voice, language } => {
state.config.default_voice = Some(voice.clone());
if let Some(lang) = language {
state.config.default_language = lang;
}
Ok(HandResult::success(serde_json::json!({
"status": "voice_set",
"voice": voice,
"language": state.config.default_language,
})))
}
SpeechAction::SetProvider { provider, api_key, region: _ } => {
state.config.provider = provider.clone();
// In real implementation, would configure provider
Ok(HandResult::success(serde_json::json!({
"status": "provider_set",
"provider": provider,
"configured": api_key.is_some(),
})))
}
}
}
/// Get current state
pub async fn get_state(&self) -> SpeechState {
self.state.read().await.clone()
}
}
impl Default for SpeechHand {
fn default() -> Self {
Self::new()
}
}
#[async_trait]
impl Hand for SpeechHand {
fn config(&self) -> &HandConfig {
&self.config
}
async fn execute(&self, _context: &HandContext, input: Value) -> Result<HandResult> {
let action: SpeechAction = match serde_json::from_value(input) {
Ok(a) => a,
Err(e) => {
return Ok(HandResult::error(format!("Invalid speech action: {}", e)));
}
};
self.execute_action(action).await
}
fn status(&self) -> HandStatus {
HandStatus::Idle
}
}
#[cfg(test)]
mod tests {
use super::*;
#[tokio::test]
async fn test_speech_creation() {
let hand = SpeechHand::new();
assert_eq!(hand.config().id, "speech");
}
#[tokio::test]
async fn test_speak() {
let hand = SpeechHand::new();
let action = SpeechAction::Speak {
text: "Hello, world!".to_string(),
voice: None,
rate: 1.0,
pitch: 1.0,
volume: 1.0,
language: None,
};
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
}
#[tokio::test]
async fn test_pause_resume() {
let hand = SpeechHand::new();
// Speak first
hand.execute_action(SpeechAction::Speak {
text: "Test".to_string(),
voice: None, rate: 1.0, pitch: 1.0, volume: 1.0, language: None,
}).await.unwrap();
// Pause
let result = hand.execute_action(SpeechAction::Pause).await.unwrap();
assert!(result.success);
// Resume
let result = hand.execute_action(SpeechAction::Resume).await.unwrap();
assert!(result.success);
}
#[tokio::test]
async fn test_list_voices() {
let hand = SpeechHand::new();
let action = SpeechAction::ListVoices { language: Some("zh".to_string()) };
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
}
#[tokio::test]
async fn test_set_voice() {
let hand = SpeechHand::new();
let action = SpeechAction::SetVoice {
voice: "zh-CN-XiaoxiaoNeural".to_string(),
language: Some("zh-CN".to_string()),
};
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
let state = hand.get_state().await;
assert_eq!(state.config.default_voice, Some("zh-CN-XiaoxiaoNeural".to_string()));
}
}

View File

@@ -1,422 +0,0 @@
//! Whiteboard Hand - Drawing and annotation capabilities
//!
//! Provides whiteboard drawing actions for teaching:
//! - draw_text: Draw text on the whiteboard
//! - draw_shape: Draw shapes (rectangle, circle, arrow, etc.)
//! - draw_line: Draw lines and curves
//! - draw_chart: Draw charts (bar, line, pie)
//! - draw_latex: Render LaTeX formulas
//! - draw_table: Draw data tables
//! - clear: Clear the whiteboard
//! - export: Export as image
use async_trait::async_trait;
use serde::{Deserialize, Serialize};
use serde_json::Value;
use zclaw_types::Result;
use crate::{Hand, HandConfig, HandContext, HandResult, HandStatus};
/// Whiteboard action types
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(tag = "action", rename_all = "snake_case")]
pub enum WhiteboardAction {
/// Draw text
DrawText {
x: f64,
y: f64,
text: String,
#[serde(default = "default_font_size")]
font_size: u32,
#[serde(default)]
color: Option<String>,
#[serde(default)]
font_family: Option<String>,
},
/// Draw a shape
DrawShape {
shape: ShapeType,
x: f64,
y: f64,
width: f64,
height: f64,
#[serde(default)]
fill: Option<String>,
#[serde(default)]
stroke: Option<String>,
#[serde(default = "default_stroke_width")]
stroke_width: u32,
},
/// Draw a line
DrawLine {
points: Vec<Point>,
#[serde(default)]
color: Option<String>,
#[serde(default = "default_stroke_width")]
stroke_width: u32,
},
/// Draw a chart
DrawChart {
chart_type: ChartType,
data: ChartData,
x: f64,
y: f64,
width: f64,
height: f64,
#[serde(default)]
title: Option<String>,
},
/// Draw LaTeX formula
DrawLatex {
latex: String,
x: f64,
y: f64,
#[serde(default = "default_font_size")]
font_size: u32,
#[serde(default)]
color: Option<String>,
},
/// Draw a table
DrawTable {
headers: Vec<String>,
rows: Vec<Vec<String>>,
x: f64,
y: f64,
#[serde(default)]
column_widths: Option<Vec<f64>>,
},
/// Erase area
Erase {
x: f64,
y: f64,
width: f64,
height: f64,
},
/// Clear whiteboard
Clear,
/// Undo last action
Undo,
/// Redo last undone action
Redo,
/// Export as image
Export {
#[serde(default = "default_export_format")]
format: String,
},
}
fn default_font_size() -> u32 { 16 }
fn default_stroke_width() -> u32 { 2 }
fn default_export_format() -> String { "png".to_string() }
/// Shape types
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(rename_all = "snake_case")]
pub enum ShapeType {
Rectangle,
RoundedRectangle,
Circle,
Ellipse,
Triangle,
Arrow,
Star,
Checkmark,
Cross,
}
/// Point for line drawing
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Point {
pub x: f64,
pub y: f64,
}
/// Chart types
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(rename_all = "snake_case")]
pub enum ChartType {
Bar,
Line,
Pie,
Scatter,
Area,
Radar,
}
/// Chart data
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChartData {
pub labels: Vec<String>,
pub datasets: Vec<Dataset>,
}
/// Dataset for charts
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Dataset {
pub label: String,
pub values: Vec<f64>,
#[serde(default)]
pub color: Option<String>,
}
/// Whiteboard state (for undo/redo)
#[derive(Debug, Clone, Default)]
pub struct WhiteboardState {
pub actions: Vec<WhiteboardAction>,
pub undone: Vec<WhiteboardAction>,
pub canvas_width: f64,
pub canvas_height: f64,
}
/// Whiteboard Hand implementation
pub struct WhiteboardHand {
config: HandConfig,
state: std::sync::Arc<tokio::sync::RwLock<WhiteboardState>>,
}
impl WhiteboardHand {
/// Create a new whiteboard hand
pub fn new() -> Self {
Self {
config: HandConfig {
id: "whiteboard".to_string(),
name: "白板".to_string(),
description: "在虚拟白板上绘制和标注".to_string(),
needs_approval: false,
dependencies: vec![],
input_schema: Some(serde_json::json!({
"type": "object",
"properties": {
"action": { "type": "string" },
"x": { "type": "number" },
"y": { "type": "number" },
"text": { "type": "string" },
}
})),
tags: vec!["presentation".to_string(), "education".to_string()],
enabled: true,
max_concurrent: 0,
timeout_secs: 0,
},
state: std::sync::Arc::new(tokio::sync::RwLock::new(WhiteboardState {
canvas_width: 1920.0,
canvas_height: 1080.0,
..Default::default()
})),
}
}
/// Create with custom canvas size
pub fn with_size(width: f64, height: f64) -> Self {
let hand = Self::new();
let mut state = hand.state.blocking_write();
state.canvas_width = width;
state.canvas_height = height;
drop(state);
hand
}
/// Execute a whiteboard action
pub async fn execute_action(&self, action: WhiteboardAction) -> Result<HandResult> {
let mut state = self.state.write().await;
match &action {
WhiteboardAction::Clear => {
state.actions.clear();
state.undone.clear();
return Ok(HandResult::success(serde_json::json!({
"status": "cleared",
"action_count": 0
})));
}
WhiteboardAction::Undo => {
if let Some(last) = state.actions.pop() {
state.undone.push(last);
return Ok(HandResult::success(serde_json::json!({
"status": "undone",
"remaining_actions": state.actions.len()
})));
}
return Ok(HandResult::success(serde_json::json!({
"status": "no_action_to_undo"
})));
}
WhiteboardAction::Redo => {
if let Some(redone) = state.undone.pop() {
state.actions.push(redone);
return Ok(HandResult::success(serde_json::json!({
"status": "redone",
"total_actions": state.actions.len()
})));
}
return Ok(HandResult::success(serde_json::json!({
"status": "no_action_to_redo"
})));
}
WhiteboardAction::Export { format } => {
// In real implementation, would render to image
return Ok(HandResult::success(serde_json::json!({
"status": "exported",
"format": format,
"data_url": format!("data:image/{};base64,<rendered_data>", format)
})));
}
_ => {
// Regular drawing action
state.actions.push(action.clone());
return Ok(HandResult::success(serde_json::json!({
"status": "drawn",
"action": action,
"total_actions": state.actions.len()
})));
}
}
}
/// Get current state
pub async fn get_state(&self) -> WhiteboardState {
self.state.read().await.clone()
}
/// Get all actions
pub async fn get_actions(&self) -> Vec<WhiteboardAction> {
self.state.read().await.actions.clone()
}
}
impl Default for WhiteboardHand {
fn default() -> Self {
Self::new()
}
}
#[async_trait]
impl Hand for WhiteboardHand {
fn config(&self) -> &HandConfig {
&self.config
}
async fn execute(&self, _context: &HandContext, input: Value) -> Result<HandResult> {
// Parse action from input
let action: WhiteboardAction = match serde_json::from_value(input.clone()) {
Ok(a) => a,
Err(e) => {
return Ok(HandResult::error(format!("Invalid whiteboard action: {}", e)));
}
};
self.execute_action(action).await
}
fn status(&self) -> HandStatus {
// Check if there are any actions
HandStatus::Idle
}
}
#[cfg(test)]
mod tests {
use super::*;
#[tokio::test]
async fn test_whiteboard_creation() {
let hand = WhiteboardHand::new();
assert_eq!(hand.config().id, "whiteboard");
}
#[tokio::test]
async fn test_draw_text() {
let hand = WhiteboardHand::new();
let action = WhiteboardAction::DrawText {
x: 100.0,
y: 100.0,
text: "Hello World".to_string(),
font_size: 24,
color: Some("#333333".to_string()),
font_family: None,
};
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
let state = hand.get_state().await;
assert_eq!(state.actions.len(), 1);
}
#[tokio::test]
async fn test_draw_shape() {
let hand = WhiteboardHand::new();
let action = WhiteboardAction::DrawShape {
shape: ShapeType::Rectangle,
x: 50.0,
y: 50.0,
width: 200.0,
height: 100.0,
fill: Some("#4CAF50".to_string()),
stroke: None,
stroke_width: 2,
};
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
}
#[tokio::test]
async fn test_undo_redo() {
let hand = WhiteboardHand::new();
// Draw something
hand.execute_action(WhiteboardAction::DrawText {
x: 0.0, y: 0.0, text: "Test".to_string(), font_size: 16, color: None, font_family: None,
}).await.unwrap();
// Undo
let result = hand.execute_action(WhiteboardAction::Undo).await.unwrap();
assert!(result.success);
assert_eq!(hand.get_state().await.actions.len(), 0);
// Redo
let result = hand.execute_action(WhiteboardAction::Redo).await.unwrap();
assert!(result.success);
assert_eq!(hand.get_state().await.actions.len(), 1);
}
#[tokio::test]
async fn test_clear() {
let hand = WhiteboardHand::new();
// Draw something
hand.execute_action(WhiteboardAction::DrawText {
x: 0.0, y: 0.0, text: "Test".to_string(), font_size: 16, color: None, font_family: None,
}).await.unwrap();
// Clear
let result = hand.execute_action(WhiteboardAction::Clear).await.unwrap();
assert!(result.success);
assert_eq!(hand.get_state().await.actions.len(), 0);
}
#[tokio::test]
async fn test_chart() {
let hand = WhiteboardHand::new();
let action = WhiteboardAction::DrawChart {
chart_type: ChartType::Bar,
data: ChartData {
labels: vec!["A".to_string(), "B".to_string(), "C".to_string()],
datasets: vec![Dataset {
label: "Values".to_string(),
values: vec![10.0, 20.0, 15.0],
color: Some("#2196F3".to_string()),
}],
},
x: 100.0,
y: 100.0,
width: 400.0,
height: 300.0,
title: Some("Test Chart".to_string()),
};
let result = hand.execute_action(action).await.unwrap();
assert!(result.success);
}
}

View File

@@ -27,7 +27,7 @@ use crate::config::KernelConfig;
use zclaw_memory::MemoryStore;
use zclaw_runtime::{LlmDriver, ToolRegistry, tool::SkillExecutor};
use zclaw_skills::SkillRegistry;
use zclaw_hands::{HandRegistry, hands::{BrowserHand, SlideshowHand, SpeechHand, QuizHand, WhiteboardHand, ResearcherHand, CollectorHand, ClipHand, TwitterHand, ReminderHand, quiz::LlmQuizGenerator}};
use zclaw_hands::{HandRegistry, hands::{BrowserHand, QuizHand, ResearcherHand, CollectorHand, ClipHand, TwitterHand, ReminderHand, quiz::LlmQuizGenerator}};
pub use adapters::KernelSkillExecutor;
pub use messaging::ChatModeConfig;
@@ -93,10 +93,7 @@ impl Kernel {
let quiz_model = config.model().to_string();
let quiz_generator = Arc::new(LlmQuizGenerator::new(driver.clone(), quiz_model));
hands.register(Arc::new(BrowserHand::new())).await;
hands.register(Arc::new(SlideshowHand::new())).await;
hands.register(Arc::new(SpeechHand::new())).await;
hands.register(Arc::new(QuizHand::with_generator(quiz_generator))).await;
hands.register(Arc::new(WhiteboardHand::new())).await;
hands.register(Arc::new(ResearcherHand::new())).await;
hands.register(Arc::new(CollectorHand::new())).await;
hands.register(Arc::new(ClipHand::new())).await;