@goodandready/dsh-voice 0.9.1 → 0.9.3

package/lib/client.js

@@ -547,15 +547,15 @@ window.__ModuleLoader__.load({
       const trimmed = String(text || '').trim().toLowerCase()
       const { buttons, input } = promptInfo
       if (buttons && buttons.length > 0) {
-        if (/^(1|первый|first)$/i.test(trimmed) && buttons[0]) { buttons[0].click(); return true }
-        if (/^(2|второй|second)$/i.test(trimmed) && buttons[1]) { buttons[1].click(); return true }
-        if (/^(3|третий|third)$/i.test(trimmed) && buttons[2]) { buttons[2].click(); return true }
-        if (/^(да|yes|ok|confirm|согласен)$/i.test(trimmed)) {
-          const yesBtn = buttons.find((b) => /да|yes|ok|confirm/i.test(b.textContent || ''))
+        if (/^(1|первый|first|第一)$/i.test(trimmed) && buttons[0]) { buttons[0].click(); return true }
+        if (/^(2|второй|second|第二)$/i.test(trimmed) && buttons[1]) { buttons[1].click(); return true }
+        if (/^(3|третий|third|第三)$/i.test(trimmed) && buttons[2]) { buttons[2].click(); return true }
+        if (/^(да|yes|ok|confirm|согласен|是|好|确认|同意)$/i.test(trimmed)) {
+          const yesBtn = buttons.find((b) => /да|yes|ok|confirm|是|好|确认|同意/i.test(b.textContent || ''))
           if (yesBtn) { yesBtn.click(); return true }
         }
-        if (/^(нет|no|отмена|cancel)$/i.test(trimmed)) {
-          const noBtn = buttons.find((b) => /нет|no|cancel|отмена/i.test(b.textContent || ''))
+        if (/^(нет|no|отмена|cancel|不|否|取消)$/i.test(trimmed)) {
+          const noBtn = buttons.find((b) => /нет|no|cancel|отмена|不|否|取消/i.test(b.textContent || ''))
           if (noBtn) { noBtn.click(); return true }
         }
       }
@@ -697,8 +697,8 @@ window.__ModuleLoader__.load({
       if (!res.ok || !parsed || !parsed.ok) {
         throw new Error((parsed && parsed.error && parsed.error.message) || `HTTP ${res.status}`)
       }
-      if (parsed.command) return { command: parsed.command, text: '' }
-      return { text: String(parsed.text || '').trim(), command: null }
+      if (parsed.command) return { command: parsed.command, text: '', action: null }
+      return { text: String(parsed.text || '').trim(), command: null, action: parsed.action || null }
     }
 
     function tidyPhrase(text) {
@@ -773,6 +773,9 @@ window.__ModuleLoader__.load({
 
     function keyLabel(name) {
       if (!name) return t('keyUnset')
+      if (typeof name === 'string' && name.includes('+')) {
+        return name.split('+').map((p) => keyLabel(p)).join(' + ')
+      }
       if (name === 'Space') return t('keySpace')
       if (KEY_LABELS[name]) return KEY_LABELS[name]
       // Drop the Key/Digit prefix from codes like KeyR and Digit5.
@@ -781,10 +784,23 @@ window.__ModuleLoader__.load({
 
     // What to store on key press. Pure modifiers are remembered by name:
     // left and right have different codes but the user means "either".
+    // Supports combos like Control+Space or Alt+KeyV.
     function keyFromEvent(event) {
-      if (['Control', 'Alt', 'Shift', 'Meta'].includes(event.key)) return event.key
-      if (event.code) return event.code
-      return event.key || ''
+      if (!event) return ''
+      const modifiers = []
+      if (event.ctrlKey && event.key !== 'Control') modifiers.push('Control')
+      if (event.altKey && event.key !== 'Alt') modifiers.push('Alt')
+      if (event.shiftKey && event.key !== 'Shift') modifiers.push('Shift')
+      if (event.metaKey && event.key !== 'Meta') modifiers.push('Meta')
+
+      const main = ['Control', 'Alt', 'Shift', 'Meta'].includes(event.key)
+        ? event.key
+        : (event.code || event.key || '')
+
+      if (modifiers.length > 0 && !['Control', 'Alt', 'Shift', 'Meta'].includes(main)) {
+        return modifiers.concat(main).join('+')
+      }
+      return main
     }
 
     // Announce that the user started speaking.
@@ -893,7 +909,10 @@ window.__ModuleLoader__.load({
         chainsPromise = fetch('/dsh-voice/status', { cache: 'no-store' })
           .then((res) => res.json())
           .then((data) => (data && data.modes) || {})
-          .catch(() => ({}))
+          .catch(() => {
+            chainsPromise = null
+            return {}
+          })
       }
       return chainsPromise.then((modes) => {
         const row = modes[mode] || {}
@@ -1212,18 +1231,56 @@ window.__ModuleLoader__.load({
     // Hotkey: holding a key is easier than aiming the mouse. While the key is
     // down we record; Escape cancels.
     function hotkeyMatches(event, name) {
+      if (!event || !name) return false
+      if (typeof name === 'string' && name.includes('+')) {
+        const parts = name.split('+')
+        const main = parts[parts.length - 1]
+        const needCtrl = parts.includes('Control')
+        const needAlt = parts.includes('Alt')
+        const needShift = parts.includes('Shift')
+        const needMeta = parts.includes('Meta')
+        if (needCtrl && !event.ctrlKey) return false
+        if (needAlt && !event.altKey) return false
+        if (needShift && !event.shiftKey) return false
+        if (needMeta && !event.metaKey) return false
+        return event.code === main || event.key === main
+      }
       if (name === 'Control') return event.key === 'Control'
       if (name === 'Alt') return event.key === 'Alt'
       if (name === 'Shift') return event.key === 'Shift'
+      if (name === 'Meta') return event.key === 'Meta'
       return event.code === name || event.key === name
     }
 
+    function isAllowedInEditable(name) {
+      if (!name) return false
+      if (['Control', 'Alt', 'Shift', 'Meta'].includes(name)) return true
+      if (/^F\d{1,2}$/.test(name)) return true
+      if (typeof name === 'string' && name.includes('+')) return true
+      if (['Tab', 'Pause', 'ScrollLock', 'Insert'].includes(name)) return true
+      return false
+    }
+
+    let activeHotkeyCleanup = null
+
+    function clearGlobalHotkey() {
+      if (typeof activeHotkeyCleanup === 'function') {
+        try { activeHotkeyCleanup() } catch (_) { /* ignore */ }
+        activeHotkeyCleanup = null
+      }
+      if (typeof window !== 'undefined' && typeof window.__dsh_voice_hotkey_cleanup === 'function') {
+        try { window.__dsh_voice_hotkey_cleanup() } catch (_) { /* ignore */ }
+        window.__dsh_voice_hotkey_cleanup = null
+      }
+    }
+
     function installHotkey(ctx, keyName, mode) {
+      clearGlobalHotkey()
       if (typeof document === 'undefined' || !keyName) return () => {}
       const down = (event) => {
         if (event.repeat) return
-        // A modifier hotkey does not fight the text field: modifiers do not
-        // type. A plain letter key must not be hijacked while typing.
+        // A modifier, function key or combo does not fight the text field:
+        // modifiers/combos do not type plain text. A plain letter key must not be hijacked while typing.
         const target = event.target
         const isEditable = target && (
           target.tagName === 'INPUT' ||
@@ -1231,23 +1288,41 @@ window.__ModuleLoader__.load({
           target.isContentEditable ||
           (typeof target.getAttribute === 'function' && target.getAttribute('contenteditable') === 'true')
         )
-        if (isEditable && !['Control', 'Alt', 'Shift'].includes(keyName)) return
+        if (isEditable && !isAllowedInEditable(keyName)) return
 
-        if (hotkeyMatches(event, keyName)) beginHold(mode)
+        if (hotkeyMatches(event, keyName)) {
+          event.preventDefault()
+          event.stopPropagation()
+          beginHold(mode)
+        }
       }
       const up = (event) => {
-        if (hotkeyMatches(event, keyName)) endHold(false)
-        else if (event.key === 'Escape' && hold.armed) endHold(true)
+        if (hotkeyMatches(event, keyName)) {
+          event.preventDefault()
+          endHold(false)
+        } else if (event.key === 'Escape' && hold.armed) {
+          event.preventDefault()
+          endHold(true)
+        }
       }
       const blur = () => { if (hold.armed) endHold(true) }
       document.addEventListener('keydown', down, true)
       document.addEventListener('keyup', up, true)
       window.addEventListener('blur', blur)
-      return () => {
+      const cleanup = () => {
         document.removeEventListener('keydown', down, true)
         document.removeEventListener('keyup', up, true)
         window.removeEventListener('blur', blur)
+        if (activeHotkeyCleanup === cleanup) activeHotkeyCleanup = null
+        if (typeof window !== 'undefined' && window.__dsh_voice_hotkey_cleanup === cleanup) {
+          window.__dsh_voice_hotkey_cleanup = null
+        }
+      }
+      activeHotkeyCleanup = cleanup
+      if (typeof window !== 'undefined') {
+        window.__dsh_voice_hotkey_cleanup = cleanup
       }
+      return cleanup
     }
 
     function cancelCurrent() {
@@ -1938,6 +2013,7 @@ window.__ModuleLoader__.load({
           alive = false
           window.removeEventListener('dsh-voice:settings-saved', reload)
           dispose()
+          clearGlobalHotkey()
         }
       }, 'dsh-voice: hold hotkey')
 
@@ -2187,6 +2263,7 @@ window.__ModuleLoader__.load({
       const [err, setErr] = React.useState('')
       const [serverConfig, setServerConfig] = React.useState(null)
       const [configLoading, setConfigLoading] = React.useState(true)
+      const [jargonInput, setJargonInput] = React.useState(null)
 
       React.useEffect(() => {
         let alive = true
@@ -2215,11 +2292,17 @@ window.__ModuleLoader__.load({
           if (event.key === 'Escape') { setCatching(false); return }
           const chosen = keyFromEvent(event)
           setDraft((d) => Object.assign({}, d || {}, { hotkey: chosen }))
+          voice.hotkey = chosen || ''
+          if (chosen) {
+            installHotkey(ctx, chosen, 'message')
+          } else {
+            clearGlobalHotkey()
+          }
           setCatching(false)
         }
         document.addEventListener('keydown', onKey, true)
         return () => document.removeEventListener('keydown', onKey, true)
-      }, [catching])
+      }, [catching, ctx])
 
       React.useEffect(() => {
         let alive = true
@@ -2505,6 +2588,21 @@ window.__ModuleLoader__.load({
       const save = async () => {
         setErr(''); setSaved(false)
         if (!draft) return
+        if (jargonInput !== null) {
+          const lines = jargonInput.split('\n')
+          const parsed = lines
+            .map((line) => {
+              const parts = line.split(/->|=|:/)
+              if (parts.length >= 2) {
+                const from = parts[0].trim()
+                const to = parts.slice(1).join(':').trim()
+                if (from && to) return { from, to }
+              }
+              return null
+            })
+            .filter(Boolean)
+          draft.jargonDictionary = parsed
+        }
 
         let httpOk = false
         let httpError = null
@@ -2543,7 +2641,14 @@ window.__ModuleLoader__.load({
         }
 
         const d = draft || {}
+        voice.hotkey = d.hotkey || ''
+        if (voice.hotkey) {
+          installHotkey(ctx, voice.hotkey, 'message')
+        } else {
+          clearGlobalHotkey()
+        }
         Object.assign(voice.settings, {
+          hotkey: d.hotkey || '',
           vadSilenceMs: Number(d.dictation && d.dictation.vadSilenceMs) || 700,
           autoSendMs: Number(d.message && d.message.autoSendMs) || 4000,
           beep: !!d.beep,
@@ -2592,7 +2697,11 @@ window.__ModuleLoader__.load({
           }, catching ? t('pressKey') : keyLabel(draft && draft.hotkey)),
           React.createElement('button', {
             type: 'button', className: 'cb-btn cb-btn-mini', title: t('clearKey'),
-            disabled: !writable, onClick: () => setDraft((d) => Object.assign({}, d || {}, { hotkey: '' })),
+            disabled: !writable, onClick: () => {
+              setDraft((d) => Object.assign({}, d || {}, { hotkey: '' }))
+              voice.hotkey = ''
+              clearGlobalHotkey()
+            },
           }, '×'),
         ),
         React.createElement('span', { className: 'dvs-sub' }, t('hotkeyHint1') + t('hotkeyHint2')),
@@ -3028,16 +3137,24 @@ window.__ModuleLoader__.load({
               rows: 3,
               disabled: !writable,
               placeholder: t('jargonPlaceholder'),
-              value: Array.isArray(draft && draft.jargonDictionary)
-                ? draft.jargonDictionary.map((item) => item && item.from ? `${item.from} -> ${item.to}` : '').filter(Boolean).join('\n')
-                : '',
+              value: jargonInput !== null
+                ? jargonInput
+                : (Array.isArray(draft && draft.jargonDictionary)
+                    ? draft.jargonDictionary.map((item) => item && item.from ? `${item.from} -> ${item.to}` : '').filter(Boolean).join('\n')
+                    : ''),
               onChange: (e) => {
-                const lines = e.target.value.split('\n')
+                setJargonInput(e.target.value)
+              },
+              onBlur: () => {
+                if (jargonInput === null) return
+                const lines = jargonInput.split('\n')
                 const parsed = lines
                   .map((line) => {
                     const parts = line.split(/->|=|:/)
                     if (parts.length >= 2) {
-                      return { from: parts[0].trim(), to: parts.slice(1).join(':').trim() }
+                      const from = parts[0].trim()
+                      const to = parts.slice(1).join(':').trim()
+                      if (from && to) return { from, to }
                     }
                     return null
                   })
@@ -3110,7 +3227,7 @@ window.__ModuleLoader__.load({
           React.createElement('div', { className: 'cb-section-desc' }, t('versionHint')),
           React.createElement('div', { className: 'cb-row', style: { alignItems: 'center', gap: '12px' } },
             React.createElement('span', { className: 'cb-badge cb-badge-ok' },
-              'v' + ((updaterStatus && updaterStatus.currentVersion) || '0.9.1')
+              'v' + ((updaterStatus && updaterStatus.currentVersion) || '0.9.3')
             ),
             updaterStatus && updaterStatus.updateAvailable
               ? React.createElement('span', { className: 'cb-badge cb-badge-warn' },

package/lib/index.js

@@ -27,14 +27,11 @@ import { createPolishText } from './polish.js'
 import { createUserMessage } from '@deepseek-ai/dsh-llm'
 import { buildProviderOrder, sessionCommand } from './transcribe-core.js'
 import { registerPluginUpdater } from './updater.js'
-import { registerSensevoiceInstaller } from './sensevoice-installer.js'
+import { registerSensevoiceInstaller, findTokensFile } from './sensevoice-installer.js'
 import { createConfigReader } from './config-compat.js'
 import { registerConfigRoutes } from './config-routes.js'
 import { getAllowedAudioRoots, validateAudioPath, validateAudioContent } from './audio-guard.js'
-
-function isAutoLang(lang) {
-  return !lang || lang === 'auto' || String(lang).includes(',')
-}
+import { isAutoLang } from './provider-http.js'
 
 export const name = '@goodandready/dsh-voice'
 export const NS = 'dsh-voice'
@@ -324,9 +321,12 @@ export function apply(ctx, baseConfig) {
     try {
       let port = '6006'
       try { port = new URL(cfg.sensevoiceUrl).port || '6006' } catch { /* invalid URL */ }
+      const tokensPath = findTokensFile(path.dirname(cfg.sensevoiceModel))
+      const tokensArg = tokensPath ? ` --tokens=${JSON.stringify(tokensPath)}` : ''
       const spec = ctx.shell.resolve({
         command: `${JSON.stringify(cfg.sensevoiceBin)}`
           + ` --sense-voice-model=${JSON.stringify(cfg.sensevoiceModel)}`
+          + tokensArg
           + ` --port=${port} --num-threads=4`,
         timeoutMs: 0,
         stdoutMaxBytes: 4 * 1024 * 1024,
@@ -595,7 +595,7 @@ export function apply(ctx, baseConfig) {
     },
   }), 'dsh-voice: /transcribe route')
 
-  ctx.tools.register(
+  ctx.effect(() => ctx.tools.register(
     defineTool({
       name: 'transcribe_audio',
       description:
@@ -649,7 +649,7 @@ export function apply(ctx, baseConfig) {
         return out
       },
     }),
-  )
+  ), 'dsh-voice: transcribe_audio tool')
 
   ctx.effect(() => () => {
     if (child) { try { child.kill && child.kill() } catch { /* already dead */ } }

package/lib/normalize.js

@@ -128,28 +128,28 @@ export function extractVoiceActions(text) {
   if (!raw) return { text: '', action: null }
 
   const lower = raw.toLowerCase().replace(/[.!?…]+$/, '').trim()
-  if (/^(отправь|отправить|send|send it|send message)$/i.test(lower)) {
+  if (/^(отправь|отправить|send|send it|send message|发送|发出)$/i.test(lower)) {
     return { text: '', action: 'send' }
   }
-  if (/^(отмена|cancel|discard)$/i.test(lower)) {
+  if (/^(отмена|cancel|discard|取消|算了)$/i.test(lower)) {
     return { text: '', action: 'cancel' }
   }
-  if (/^(очисти|очистить|clear|clear composer)$/i.test(lower)) {
+  if (/^(очисти|очистить|clear|clear composer|清空|清除)$/i.test(lower)) {
     return { text: '', action: 'clear' }
   }
-  if (/^(новая строка|с новой строки|new line|newline)$/i.test(lower)) {
+  if (/^(новая строка|с новой строки|new line|newline|换行|新行)$/i.test(lower)) {
     return { text: '', action: 'newline' }
   }
 
-  const trailingSend = /[\s,]+(отправь|отправить|send|send it)[.!?…]*$/i
+  const trailingSend = /[\s,]+(отправь|отправить|send|send it|发送|发出)[.!?…]*$/i
   if (trailingSend.test(raw)) {
     return { text: raw.replace(trailingSend, '').trim(), action: 'send' }
   }
-  const trailingCancel = /[\s,]+(отмена|cancel)[.!?…]*$/i
+  const trailingCancel = /[\s,]+(отмена|cancel|取消|算了)[.!?…]*$/i
   if (trailingCancel.test(raw)) {
     return { text: raw.replace(trailingCancel, '').trim(), action: 'cancel' }
   }
-  const trailingNewline = /[\s,]+(новая строка|с новой строки|new line)[.!?…]*$/i
+  const trailingNewline = /[\s,]+(новая строка|с новой строки|new line|换行|新行)[.!?…]*$/i
   if (trailingNewline.test(raw)) {
     return { text: raw.replace(trailingNewline, '').trim() + '\n', action: 'newline' }
   }

package/lib/sensevoice-installer.js

@@ -2,7 +2,7 @@
 // Downloads pre-quantized ONNX model and configures autostart on the host.
 
 import { createWriteStream, existsSync, readdirSync } from 'node:fs'
-import { mkdir, stat } from 'node:fs/promises'
+import { mkdir, stat, unlink } from 'node:fs/promises'
 import { homedir } from 'node:os'
 import path from 'node:path'
 import { spawn } from 'node:child_process'
@@ -161,6 +161,7 @@ export async function startSensevoiceInstall(ctx, cfg, updateConfigFn, startSens
     installerState.step = 'extracting'
     installerState.progress = 95
     await extractTarBz2(archivePath, modelDir)
+    await unlink(archivePath).catch(() => {})
 
     const modelFile = findModelFile(modelDir)
     if (!modelFile) throw new Error('model.int8.onnx not found after extraction')
@@ -183,6 +184,7 @@ export async function startSensevoiceInstall(ctx, cfg, updateConfigFn, startSens
 
     return { ok: true, modelPath: modelFile }
   } catch (err) {
+    await unlink(archivePath).catch(() => {})
     installerState.installing = false
     installerState.error = String(err?.message || err)
     installerState.step = 'failed'

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@goodandready/dsh-voice",
-  "version": "0.9.1",
+  "version": "0.9.3",
   "description": "Voice input for DeepSeek Harness: dictation chunked by pauses and voice messages, each with its own provider fallback chain (Deepgram, Groq, HuggingFace, local whisper.cpp, plus any OpenAI-compatible API of your own).",
   "license": "MIT",
   "type": "module",
@@ -43,6 +43,7 @@
   "scripts": {
     "test": "node --test test/*.test.mjs",
     "build:client": "node scripts/build-client.mjs",
+    "build:check": "node scripts/build-client.mjs --check",
     "pretest": "node scripts/build-client.mjs",
     "lint": "node --check lib/*.js scripts/*.mjs",
     "pack:check": "node scripts/pack-check.mjs"