Image capture — camera button, image preview, voice annotation, combined upload, client-side resize

This commit is contained in:
2026-04-27 00:31:28 -04:00
parent 337f1fcca7
commit 75cee895ce
+322 -139
View File
@@ -1,12 +1,13 @@
'use client';
import { useState, useRef, useEffect } from 'react';
import { useState, useRef, useEffect, useCallback } from 'react';
const API_URL = process.env.NEXT_PUBLIC_API_URL || '';
const DB_NAME = 'bird-capture-queue';
const STORE_NAME = 'pending';
const MAX_IMAGE_PX = 1568; // Claude vision optimal max dimension
// IndexedDB queue for offline captures
// IndexedDB queue
async function openDB(): Promise<IDBDatabase> {
return new Promise((resolve, reject) => {
const req = indexedDB.open(DB_NAME, 1);
@@ -16,12 +17,19 @@ async function openDB(): Promise<IDBDatabase> {
});
}
async function queueCapture(blob: Blob, mimeType: string) {
async function queueCapture(formData: FormData) {
const db = await openDB();
const id = Date.now().toString();
const audio = formData.get('audio') as Blob | null;
const image = formData.get('image') as Blob | null;
return new Promise<void>((resolve, reject) => {
const tx = db.transaction(STORE_NAME, 'readwrite');
tx.objectStore(STORE_NAME).put({ id, blob, mimeType, timestamp: new Date().toISOString() });
tx.objectStore(STORE_NAME).put({
id,
audio: audio || null,
image: image || null,
timestamp: new Date().toISOString(),
});
tx.oncomplete = () => resolve();
tx.onerror = () => reject(tx.error);
});
@@ -47,10 +55,36 @@ async function removeCapture(id: string) {
});
}
async function submitCapture(blob: Blob, mimeType: string): Promise<boolean> {
// Resize image client-side before upload
async function resizeImage(file: File): Promise<Blob> {
return new Promise((resolve) => {
const img = new Image();
const url = URL.createObjectURL(file);
img.onload = () => {
URL.revokeObjectURL(url);
const { width, height } = img;
const maxDim = MAX_IMAGE_PX;
let w = width, h = height;
if (w > maxDim || h > maxDim) {
if (w > h) { h = Math.round(h * maxDim / w); w = maxDim; }
else { w = Math.round(w * maxDim / h); h = maxDim; }
}
const canvas = document.createElement('canvas');
canvas.width = w;
canvas.height = h;
const ctx = canvas.getContext('2d')!;
ctx.drawImage(img, 0, 0, w, h);
canvas.toBlob((blob) => resolve(blob!), 'image/jpeg', 0.88);
};
img.src = url;
});
}
async function submitCapture(audio: Blob | null, image: Blob | null): Promise<boolean> {
try {
const form = new FormData();
form.append('audio', blob, 'capture.webm');
if (audio) form.append('audio', audio, 'capture.webm');
if (image) form.append('image', image, 'capture.jpg');
const res = await fetch(`${API_URL}/api/capture`, { method: 'POST', body: form });
return res.ok;
} catch {
@@ -58,23 +92,28 @@ async function submitCapture(blob: Blob, mimeType: string): Promise<boolean> {
}
}
type CaptureMode = 'idle' | 'image-selected' | 'recording' | 'submitting' | 'saved' | 'queued' | 'error';
export default function CapturePage() {
const [state, setState] = useState<'idle' | 'recording' | 'transcribing' | 'saved' | 'queued' | 'error'>('idle');
const [mode, setMode] = useState<CaptureMode>('idle');
const [imagePreview, setImagePreview] = useState<string | null>(null);
const [imageBlob, setImageBlob] = useState<Blob | null>(null);
const [countdown, setCountdown] = useState(0);
const [recentCaptures, setRecentCaptures] = useState<{name: string, duration: string}[]>([]);
const [pendingCount, setPendingCount] = useState(0);
const [recentCaptures, setRecentCaptures] = useState<{name: string}[]>([]);
const mediaRecorderRef = useRef<MediaRecorder | null>(null);
const chunksRef = useRef<Blob[]>([]);
const streamRef = useRef<MediaStream | null>(null);
const wakeLockRef = useRef<any>(null);
const countdownRef = useRef<any>(null);
const mimeTypeRef = useRef('audio/webm');
const MAX_SECONDS = 600;
const fileInputRef = useRef<HTMLInputElement>(null);
const MAX_SECONDS = 120; // shorter for annotation — 2 min max
useEffect(() => {
loadRecentCaptures();
checkPendingQueue();
// Retry queue on focus (when user returns to app)
checkPending();
window.addEventListener('focus', retryQueue);
window.addEventListener('online', retryQueue);
return () => {
@@ -93,31 +132,54 @@ export default function CapturePage() {
} catch {}
}
async function checkPendingQueue() {
async function checkPending() {
try {
const pending = await getPendingCaptures();
setPendingCount(pending.length);
const p = await getPendingCaptures();
setPendingCount(p.length);
} catch {}
}
async function retryQueue() {
try {
const pending = await getPendingCaptures();
if (pending.length === 0) return;
for (const item of pending) {
const ok = await submitCapture(item.blob, item.mimeType);
if (ok) {
await removeCapture(item.id);
}
const ok = await submitCapture(item.audio, item.image);
if (ok) await removeCapture(item.id);
}
const remaining = await getPendingCaptures();
setPendingCount(remaining.length);
if (remaining.length < pending.length) {
await loadRecentCaptures();
}
if (remaining.length < pending.length) loadRecentCaptures();
} catch {}
}
// ── Image selection ────────────────────────────────────────────────────────
function handleCameraClick() {
fileInputRef.current?.click();
}
async function handleImageSelected(e: React.ChangeEvent<HTMLInputElement>) {
const file = e.target.files?.[0];
if (!file) return;
const resized = await resizeImage(file);
setImageBlob(resized);
const preview = URL.createObjectURL(resized);
setImagePreview(preview);
setMode('image-selected');
// Reset input so same file can be reselected
e.target.value = '';
}
function clearImage() {
if (imagePreview) URL.revokeObjectURL(imagePreview);
setImagePreview(null);
setImageBlob(null);
setMode('idle');
stopRecording();
}
// ── Voice recording ────────────────────────────────────────────────────────
async function acquireWakeLock() {
try {
if ('wakeLock' in navigator) {
@@ -145,29 +207,17 @@ export default function CapturePage() {
mr.onstop = async () => {
streamRef.current?.getTracks().forEach(t => t.stop());
streamRef.current = null;
if (chunksRef.current.length === 0) return;
setState('transcribing');
const blob = new Blob(chunksRef.current, { type: mimeTypeRef.current });
const ok = await submitCapture(blob, mimeTypeRef.current);
if (ok) {
setState('saved');
await loadRecentCaptures();
setTimeout(() => setState('idle'), 3000);
} else {
// Queue for later — don't lose the capture
try {
await queueCapture(blob, mimeTypeRef.current);
await checkPendingQueue();
setState('queued');
} catch {
setState('error');
}
setTimeout(() => setState('idle'), 4000);
if (chunksRef.current.length === 0) {
// No audio recorded — submit image only
await submitFinal(null);
return;
}
const audioBlob = new Blob(chunksRef.current, { type: mimeTypeRef.current });
await submitFinal(audioBlob);
};
mr.start(1000);
mediaRecorderRef.current = mr;
setState('recording');
setMode('recording');
setCountdown(MAX_SECONDS);
await acquireWakeLock();
countdownRef.current = setInterval(() => {
@@ -177,24 +227,64 @@ export default function CapturePage() {
});
}, 1000);
} catch {
setState('error');
setTimeout(() => setState('idle'), 3000);
setMode('error');
setTimeout(() => setMode(imageBlob ? 'image-selected' : 'idle'), 3000);
}
}
function stopRecording() {
if (countdownRef.current) { clearInterval(countdownRef.current); countdownRef.current = null; }
releaseWakeLock();
mediaRecorderRef.current?.stop();
if (mediaRecorderRef.current?.state === 'recording') {
mediaRecorderRef.current.stop();
}
}
function handleTap() {
if (state === 'recording') stopRecording();
else if (state === 'idle') startRecording();
// ── Submission ─────────────────────────────────────────────────────────────
async function submitFinal(audioBlob: Blob | null) {
setMode('submitting');
const ok = await submitCapture(audioBlob, imageBlob);
if (ok) {
setMode('saved');
if (imagePreview) URL.revokeObjectURL(imagePreview);
setImagePreview(null);
setImageBlob(null);
await loadRecentCaptures();
setTimeout(() => setMode('idle'), 3000);
} else {
try {
const form = new FormData();
if (audioBlob) form.append('audio', audioBlob, 'capture.webm');
if (imageBlob) form.append('image', imageBlob, 'capture.jpg');
await queueCapture(form);
await checkPending();
setMode('queued');
} catch {
setMode('error');
}
if (imagePreview) URL.revokeObjectURL(imagePreview);
setImagePreview(null);
setImageBlob(null);
setTimeout(() => setMode('idle'), 4000);
}
}
const isRecording = state === 'recording';
const isBusy = state === 'transcribing';
async function submitImageOnly() {
await submitFinal(null);
}
// ── Tap handler for voice button ───────────────────────────────────────────
function handleVoiceTap() {
if (mode === 'recording') stopRecording();
else if (mode === 'image-selected') startRecording();
else if (mode === 'idle') startRecording(); // voice-only capture
}
const isRecording = mode === 'recording';
const isBusy = mode === 'submitting';
const hasImage = imageBlob !== null;
return (
<div style={{
@@ -205,7 +295,7 @@ export default function CapturePage() {
alignItems: 'center',
justifyContent: 'space-between',
paddingTop: 'max(48px, env(safe-area-inset-top))',
paddingBottom: 'max(48px, env(safe-area-inset-bottom))',
paddingBottom: 'max(32px, env(safe-area-inset-bottom))',
paddingLeft: 'max(24px, env(safe-area-inset-left))',
paddingRight: 'max(24px, env(safe-area-inset-right))',
fontFamily: 'var(--font-mono)',
@@ -218,110 +308,190 @@ export default function CapturePage() {
</p>
{pendingCount > 0 && (
<p style={{ fontSize: '9px', color: '#c8821a', letterSpacing: '0.1em', margin: '4px 0 0', textTransform: 'uppercase' }}>
{pendingCount} capture{pendingCount > 1 ? 's' : ''} queued will sync when online
{pendingCount} capture{pendingCount > 1 ? 's' : ''} queued
</p>
)}
</div>
{/* Main button area */}
<div style={{ display: 'flex', flexDirection: 'column', alignItems: 'center', gap: '40px' }}>
{/* Main area */}
<div style={{ display: 'flex', flexDirection: 'column', alignItems: 'center', gap: '28px', width: '100%', maxWidth: '320px' }}>
{/* Waveform */}
<div style={{
display: 'flex', alignItems: 'center', gap: '3px', height: '40px',
opacity: isRecording ? 1 : 0.2, transition: 'opacity 0.4s ease',
}}>
{[4,8,16,6,24,10,32,14,20,8,12,28,6,18,4,22,10,16,8,4].map((h, i) => (
<div key={i} style={{
width: '2px', height: `${h}px`, background: 'var(--accent)', borderRadius: '1px',
animation: isRecording ? `wave ${0.8 + (i % 4) * 0.2}s ease-in-out infinite alternate` : 'none',
}} />
))}
{/* Image preview */}
{imagePreview && (
<div style={{ position: 'relative', width: '100%' }}>
<img
src={imagePreview}
alt="capture preview"
style={{
width: '100%',
maxHeight: '240px',
objectFit: 'cover',
borderRadius: '12px',
border: '1px solid var(--border)',
display: 'block',
}}
/>
<button
onClick={clearImage}
style={{
position: 'absolute', top: '8px', right: '8px',
background: 'rgba(0,0,0,0.6)', border: 'none',
borderRadius: '50%', width: '28px', height: '28px',
color: '#fff', cursor: 'pointer', fontSize: '16px',
display: 'flex', alignItems: 'center', justifyContent: 'center',
}}
>×</button>
</div>
)}
{/* Waveform — shows during recording */}
{isRecording && (
<div style={{ display: 'flex', alignItems: 'center', gap: '3px', height: '32px' }}>
{[4,8,16,6,24,10,32,14,20,8,12,28,6,18,4].map((h, i) => (
<div key={i} style={{
width: '2px', height: `${h}px`, background: 'var(--accent)', borderRadius: '1px',
animation: `wave ${0.8 + (i % 4) * 0.2}s ease-in-out infinite alternate`,
}} />
))}
</div>
)}
{/* Capture buttons row */}
<div style={{ display: 'flex', gap: '24px', alignItems: 'center', justifyContent: 'center' }}>
{/* Camera button — always visible, opens image picker */}
<button
onClick={handleCameraClick}
disabled={isBusy || isRecording}
style={{
width: '72px', height: '72px', borderRadius: '50%',
border: `1.5px solid ${hasImage ? 'var(--accent)' : 'var(--border2)'}`,
background: hasImage ? 'var(--accent-light)' : 'var(--bg2)',
cursor: isBusy || isRecording ? 'not-allowed' : 'pointer',
display: 'flex', alignItems: 'center', justifyContent: 'center',
transition: 'all 0.2s', touchAction: 'manipulation',
opacity: isBusy || isRecording ? 0.5 : 1,
}}
aria-label="Take photo or choose image"
>
<svg width="24" height="24" viewBox="0 0 24 24" fill="none"
stroke={hasImage ? 'var(--accent)' : 'var(--text3)'} strokeWidth="1.5">
<path d="M23 19a2 2 0 0 1-2 2H3a2 2 0 0 1-2-2V8a2 2 0 0 1 2-2h4l2-3h6l2 3h4a2 2 0 0 1 2 2z"/>
<circle cx="12" cy="13" r="4"/>
</svg>
</button>
{/* Mic / submit button — large center */}
<button
onPointerUp={handleVoiceTap}
disabled={isBusy}
style={{
width: '120px', height: '120px', borderRadius: '50%',
border: `1.5px solid ${isRecording ? '#a32d2d' : 'var(--accent)'}`,
background: isRecording ? 'rgba(163,45,45,0.08)' : 'var(--bg2)',
cursor: isBusy ? 'not-allowed' : 'pointer',
display: 'flex', alignItems: 'center', justifyContent: 'center',
transition: 'all 0.3s', touchAction: 'manipulation',
animation: isBusy ? 'none' : isRecording ? 'recordPulse 1.5s ease-in-out infinite' : 'idlePulse 3s ease-in-out infinite',
}}
aria-label={isRecording ? 'Stop recording' : 'Record voice annotation'}
>
{isBusy ? (
<svg width="24" height="24" viewBox="0 0 24 24" fill="currentColor" style={{ color: 'var(--accent)' }}>
<circle cx="12" cy="12" r="3" opacity="0.6">
<animate attributeName="opacity" values="0.6;1;0.6" dur="1s" repeatCount="indefinite"/>
</circle>
</svg>
) : isRecording ? (
<div style={{ width: '20px', height: '20px', background: '#a32d2d', borderRadius: '4px' }} />
) : (
<svg width="28" height="28" viewBox="0 0 24 24" fill="none" stroke="var(--accent)" strokeWidth="1.5">
<path d="M12 1a4 4 0 0 1 4 4v6a4 4 0 0 1-8 0V5a4 4 0 0 1 4-4z"/>
<path d="M5 11a7 7 0 0 0 14 0"/>
<line x1="12" y1="18" x2="12" y2="22"/>
<line x1="8" y1="22" x2="16" y2="22"/>
</svg>
)}
</button>
{/* Submit image-only button — visible only when image selected and not recording */}
{hasImage && !isRecording && !isBusy ? (
<button
onClick={submitImageOnly}
style={{
width: '72px', height: '72px', borderRadius: '50%',
border: '1.5px solid var(--accent)',
background: 'var(--accent-light)',
cursor: 'pointer',
display: 'flex', alignItems: 'center', justifyContent: 'center',
transition: 'all 0.2s', touchAction: 'manipulation',
}}
aria-label="Save image without annotation"
>
<svg width="22" height="22" viewBox="0 0 24 24" fill="none" stroke="var(--accent)" strokeWidth="2">
<polyline points="20 6 9 17 4 12"/>
</svg>
</button>
) : (
<div style={{ width: '72px', height: '72px' }} />
)}
</div>
{/* Big button */}
<button
onPointerUp={handleTap}
disabled={isBusy}
style={{
width: '160px', height: '160px', borderRadius: '50%',
border: `1.5px solid ${isRecording ? '#a32d2d' : 'var(--accent)'}`,
background: isRecording ? 'rgba(163,45,45,0.08)' : 'var(--bg2)',
cursor: isBusy ? 'not-allowed' : 'pointer',
display: 'flex', alignItems: 'center', justifyContent: 'center',
transition: 'all 0.3s ease', touchAction: 'manipulation',
animation: isBusy ? 'none' : isRecording ? 'recordPulse 1.5s ease-in-out infinite' : 'idlePulse 3s ease-in-out infinite',
}}
aria-label={isRecording ? 'Stop recording' : 'Start recording'}
>
{isBusy ? (
<svg width="28" height="28" viewBox="0 0 24 24" fill="currentColor" style={{ color: 'var(--accent)' }}>
<circle cx="12" cy="12" r="3" opacity="0.6">
<animate attributeName="opacity" values="0.6;1;0.6" dur="1s" repeatCount="indefinite"/>
</circle>
</svg>
) : isRecording ? (
<div style={{ width: '24px', height: '24px', background: '#a32d2d', borderRadius: '4px' }} />
) : (
<svg width="32" height="32" viewBox="0 0 24 24" fill="none" stroke="var(--accent)" strokeWidth="1.5">
<path d="M12 1a4 4 0 0 1 4 4v6a4 4 0 0 1-8 0V5a4 4 0 0 1 4-4z"/>
<path d="M5 11a7 7 0 0 0 14 0"/>
<line x1="12" y1="18" x2="12" y2="22"/>
<line x1="8" y1="22" x2="16" y2="22"/>
</svg>
)}
</button>
{/* Recording progress */}
{isRecording && (
<div style={{ width: '100%' }}>
<div style={{ width: '100%', height: '2px', background: 'var(--border)', borderRadius: '1px', overflow: 'hidden', marginBottom: '6px' }}>
<div style={{
height: '100%',
width: `${((MAX_SECONDS - countdown) / MAX_SECONDS) * 100}%`,
background: countdown <= 15 ? '#a32d2d' : 'var(--accent)',
borderRadius: '1px', transition: 'width 1s linear',
}} />
</div>
<p style={{ fontSize: '10px', color: 'var(--text3)', margin: 0, textAlign: 'center' }}>
{(() => {
const rm = Math.floor(countdown / 60);
const rs = countdown % 60;
return `${rm}:${rs.toString().padStart(2,'0')} left`;
})()}
</p>
</div>
)}
{/* Status */}
<div style={{ textAlign: 'center', minHeight: '36px' }}>
{state === 'idle' && (
<p style={{ fontSize: '11px', color: 'var(--text3)', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
tap to capture
<div style={{ textAlign: 'center', minHeight: '20px' }}>
{mode === 'idle' && (
<p style={{ fontSize: '10px', color: 'var(--text3)', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
tap camera or mic
</p>
)}
{state === 'recording' && (
<>
<p style={{ fontSize: '11px', color: '#a32d2d', letterSpacing: '0.12em', textTransform: 'uppercase', margin: '0 0 10px' }}>
recording
</p>
<div style={{ width: '160px', height: '2px', background: 'var(--border)', borderRadius: '1px', overflow: 'hidden', marginBottom: '8px' }}>
<div style={{
height: '100%',
width: `${((MAX_SECONDS - countdown) / MAX_SECONDS) * 100}%`,
background: countdown <= 30 ? '#a32d2d' : 'var(--accent)',
borderRadius: '1px',
transition: 'width 1s linear, background 0.5s ease',
}} />
</div>
<p style={{ fontSize: '11px', color: 'var(--text3)', letterSpacing: '0.08em', margin: 0 }}>
{(() => {
const elapsed = MAX_SECONDS - countdown;
const em = Math.floor(elapsed / 60);
const es = elapsed % 60;
const rm = Math.floor(countdown / 60);
const rs = countdown % 60;
return `${em}:${es.toString().padStart(2,'0')} / ${rm}:${rs.toString().padStart(2,'0')} left`;
})()}
</p>
</>
)}
{state === 'transcribing' && (
<p style={{ fontSize: '11px', color: 'var(--text3)', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
transcribing...
{mode === 'image-selected' && (
<p style={{ fontSize: '10px', color: 'var(--text3)', letterSpacing: '0.10em', textTransform: 'uppercase', margin: 0 }}>
annotate with voice or save now
</p>
)}
{state === 'saved' && (
<p style={{ fontSize: '11px', color: 'var(--accent)', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
{mode === 'recording' && (
<p style={{ fontSize: '10px', color: '#a32d2d', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
recording annotation
</p>
)}
{mode === 'submitting' && (
<p style={{ fontSize: '10px', color: 'var(--text3)', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
{hasImage ? 'processing image...' : 'transcribing...'}
</p>
)}
{mode === 'saved' && (
<p style={{ fontSize: '10px', color: 'var(--accent)', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
saved
</p>
)}
{state === 'queued' && (
<p style={{ fontSize: '11px', color: '#c8821a', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
{mode === 'queued' && (
<p style={{ fontSize: '10px', color: '#c8821a', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
queued will sync when online
</p>
)}
{state === 'error' && (
<p style={{ fontSize: '11px', color: '#a32d2d', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
{mode === 'error' && (
<p style={{ fontSize: '10px', color: '#a32d2d', letterSpacing: '0.12em', textTransform: 'uppercase', margin: 0 }}>
error try again
</p>
)}
@@ -330,19 +500,22 @@ export default function CapturePage() {
{/* Recent captures */}
<div style={{ width: '100%', maxWidth: '320px' }}>
<p style={{ fontSize: '9px', color: 'var(--text3)', letterSpacing: '0.12em', textTransform: 'uppercase', margin: '0 0 10px' }}>
<p style={{ fontSize: '9px', color: 'var(--text3)', letterSpacing: '0.12em', textTransform: 'uppercase', margin: '0 0 8px' }}>
recent captures
</p>
{recentCaptures.length === 0 ? (
<p style={{ fontSize: '10px', color: 'var(--text3)', margin: 0, opacity: 0.5 }}>none yet</p>
) : (
<div style={{ display: 'flex', flexDirection: 'column', gap: '6px' }}>
<div style={{ display: 'flex', flexDirection: 'column', gap: '5px' }}>
{recentCaptures.slice(0, 4).map((c, i) => (
<div key={i} style={{
display: 'flex', alignItems: 'center', gap: '8px', padding: '8px 10px',
display: 'flex', alignItems: 'center', gap: '8px', padding: '6px 10px',
background: 'var(--bg2)', borderRadius: '6px', border: '1px solid var(--border)',
}}>
<div style={{ width: '4px', height: '4px', borderRadius: '50%', background: 'var(--accent)', flexShrink: 0 }} />
<div style={{
width: '6px', height: '6px', borderRadius: '50%', flexShrink: 0,
background: c.name.includes('image') ? '#c8821a' : 'var(--accent)',
}} />
<span style={{ fontSize: '10px', color: 'var(--text2)', flex: 1, overflow: 'hidden', textOverflow: 'ellipsis', whiteSpace: 'nowrap' }}>
{c.name}
</span>
@@ -352,6 +525,16 @@ export default function CapturePage() {
)}
</div>
{/* Hidden file input */}
<input
ref={fileInputRef}
type="file"
accept="image/*"
capture="environment"
onChange={handleImageSelected}
style={{ display: 'none' }}
/>
<style>{`
@keyframes idlePulse {
0%, 100% { box-shadow: 0 0 0 0 rgba(45,158,107,0); }