feat(capture): dictation, personalities, 0.5.0

Ships the Capture release end to end. Global-hotkey dictation with synthetic paste into the focused app on macOS and Windows, an on-screen pill across recording / transcribing / refining, customizable push-to- talk and toggle chords, and an accessibility-permission prompt scoped to Settings → Captures with inline re-check feedback. Voice profiles gain optional personalities that power compose / rewrite / respond actions via a local Qwen3 LLM — shared with refinement, so there is one local LLM in the app, not two. Refinement hardened with deterministic Whisper-loop collapse before the LLM sees the transcript, per-capture flag snapshots for re-runs, and a ten-transcript evaluation harness across every bundled refinement size. Version bump 0.4.5 → 0.5.0. Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>
2026-09-18 14:20:42 -07:00 · 2026-04-22 18:49:16 -07:00
parent ed2eec591a
commit 87c582ad54
84 changed files with 11043 additions and 512 deletions
@@ -0,0 +1,127 @@
+import { invoke } from '@tauri-apps/api/core';
+import { listen, type UnlistenFn } from '@tauri-apps/api/event';
+import { AlertTriangle, ExternalLink } from 'lucide-react';
+import { useCallback, useEffect, useState } from 'react';
+import { Button } from '@/components/ui/button';
+import { usePlatform } from '@/platform/PlatformContext';
+
+/**
+ * Tracks macOS Accessibility permission state. Without this permission the
+ * global chord can still record, but the synthetic-⌘V paste silently drops —
+ * so callers can surface an inline prompt instead of relying on the
+ * system-level permission dialog (which only fires once, the first time the
+ * app tries to post a keystroke).
+ *
+ * Triggered on three signals:
+ * - app mount in Tauri
+ * - `system:accessibility-missing` event from the dictate window's paste
+ *   failure handler
+ * - window focus (cheap way to re-check after the user flips the toggle in
+ *   System Settings and alt-tabs back)
+ */
+export function useAccessibilityPermission() {
+  const platform = usePlatform();
+  const [needsPermission, setNeedsPermission] = useState(false);
+  const [checking, setChecking] = useState(false);
+
+  const recheck = useCallback(async (): Promise<boolean> => {
+    if (!platform.metadata.isTauri) return true;
+    setChecking(true);
+    try {
+      const trusted = await invoke<boolean>('check_accessibility_permission');
+      setNeedsPermission(!trusted);
+      return trusted;
+    } catch (err) {
+      console.warn('[accessibility] check failed:', err);
+      return false;
+    } finally {
+      setChecking(false);
+    }
+  }, [platform.metadata.isTauri]);
+
+  useEffect(() => {
+    if (!platform.metadata.isTauri) return;
+    recheck();
+    const onFocus = () => {
+      recheck();
+    };
+    window.addEventListener('focus', onFocus);
+    return () => window.removeEventListener('focus', onFocus);
+  }, [platform.metadata.isTauri, recheck]);
+
+  useEffect(() => {
+    if (!platform.metadata.isTauri) return;
+    let unlisten: UnlistenFn | null = null;
+    listen('system:accessibility-missing', () => {
+      setNeedsPermission(true);
+    })
+      .then((fn) => {
+        unlisten = fn;
+      })
+      .catch(() => {});
+    return () => {
+      if (unlisten) unlisten();
+    };
+  }, [platform.metadata.isTauri]);
+
+  const openSettings = useCallback(async () => {
+    try {
+      await invoke('open_accessibility_settings');
+    } catch (err) {
+      console.warn('[accessibility] open settings failed:', err);
+    }
+  }, []);
+
+  return { needsPermission, checking, recheck, openSettings };
+}
+
+/**
+ * Inline notice rendered next to the auto-paste setting when macOS
+ * Accessibility permission is missing. Returns null when the permission is
+ * already granted.
+ */
+export function AccessibilityNotice() {
+  const { needsPermission, checking, recheck, openSettings } = useAccessibilityPermission();
+  const [stillMissing, setStillMissing] = useState(false);
+
+  const handleRecheck = useCallback(async () => {
+    setStillMissing(false);
+    const trusted = await recheck();
+    if (!trusted) setStillMissing(true);
+  }, [recheck]);
+
+  if (!needsPermission) return null;
+
+  return (
+    <div className="mt-3 rounded-lg border border-amber-500/30 bg-amber-500/10 px-3.5 py-3">
+      <div className="flex items-start gap-3">
+        <AlertTriangle className="h-4 w-4 shrink-0 mt-0.5 text-amber-500" />
+        <div className="flex-1 min-w-0 space-y-1">
+          <p className="text-sm font-medium text-foreground">
+            Grant Accessibility permission to enable auto-paste
+          </p>
+          <p className="text-sm text-muted-foreground leading-relaxed">
+            Voicebox needs System Settings → Privacy &amp; Security → Accessibility
+            to paste transcriptions into other apps. Your dictation still lands
+            in the Captures tab without it.
+          </p>
+          <div className="flex items-center gap-2 pt-1.5">
+            <Button size="sm" onClick={openSettings} className="gap-1.5">
+              <ExternalLink className="h-3.5 w-3.5" />
+              Open Settings
+            </Button>
+            <Button variant="outline" size="sm" onClick={handleRecheck} disabled={checking}>
+              {checking ? 'Checking…' : "I've enabled it"}
+            </Button>
+          </div>
+          {stillMissing && !checking && (
+            <p className="text-xs text-amber-600 dark:text-amber-400 pt-1">
+              Still not detected. macOS usually requires quitting and reopening
+              Voicebox after toggling the permission.
+            </p>
+          )}
+        </div>
+      </div>
+    </div>
+  );
+}