490 lines
22 KiB
JavaScript
490 lines
22 KiB
JavaScript
import { useState, useCallback } from 'react'
|
|
import { motion, AnimatePresence } from 'framer-motion'
|
|
import {
|
|
Sparkles, Zap, Loader2, Settings, Image as ImageIcon, FileText,
|
|
Layers, ChevronLeft, CheckCircle2, Database,
|
|
} from 'lucide-react'
|
|
import ImageUpload from './components/ImageUpload'
|
|
import ModeSelector from './components/ModeSelector'
|
|
import ResultPanel from './components/ResultPanel'
|
|
import AdvancedSettings from './components/AdvancedSettings'
|
|
import PDFProcessor from './components/PDFProcessor'
|
|
import MetadataForm from './components/MetadataForm'
|
|
import JobsPanel from './components/JobsPanel'
|
|
import axios from 'axios'
|
|
|
|
const API_BASE = import.meta.env.VITE_API_URL || '/api'
|
|
|
|
const INPUT_CLASS =
|
|
'w-full bg-white/5 border border-white/10 rounded-lg px-3 py-2 text-sm text-gray-200 ' +
|
|
'placeholder-gray-600 focus:outline-none focus:border-purple-500/50 transition-colors'
|
|
|
|
function App() {
|
|
const [view, setView] = useState('new_job')
|
|
|
|
// OCR state
|
|
const [mode, setMode] = useState('plain_ocr')
|
|
const [fileType, setFileType] = useState('image')
|
|
const [image, setImage] = useState(null)
|
|
const [imagePreview, setImagePreview] = useState(null)
|
|
const [result, setResult] = useState(null)
|
|
const [loading, setLoading] = useState(false)
|
|
const [error, setError] = useState(null)
|
|
const [showAdvanced, setShowAdvanced] = useState(false)
|
|
const [includeCaption, setIncludeCaption] = useState(false)
|
|
|
|
const [prompt, setPrompt] = useState('')
|
|
const [findTerm, setFindTerm] = useState('')
|
|
const [advancedSettings, setAdvancedSettings] = useState({
|
|
base_size: 1024, image_size: 640, crop_mode: true, test_compress: false,
|
|
})
|
|
|
|
const [metadata, setMetadata] = useState({ author: '', book: '', chapter: '', page: '' })
|
|
const [editedOcrText, setEditedOcrText] = useState('')
|
|
const [commitLoading, setCommitLoading] = useState(false)
|
|
const [commitResult, setCommitResult] = useState(null)
|
|
|
|
// Whether to show the full-screen result view
|
|
const showResultView = view === 'new_job' && mode === 'plain_ocr' && !!result
|
|
|
|
const handleFileTypeChange = useCallback((newType) => {
|
|
setImage(null)
|
|
if (imagePreview) URL.revokeObjectURL(imagePreview)
|
|
setImagePreview(null)
|
|
setError(null)
|
|
setResult(null)
|
|
setFileType(newType)
|
|
}, [imagePreview])
|
|
|
|
const handleImageSelect = useCallback((file) => {
|
|
if (file === null) {
|
|
setImage(null)
|
|
if (imagePreview && fileType === 'image') URL.revokeObjectURL(imagePreview)
|
|
setImagePreview(null)
|
|
setError(null)
|
|
setResult(null)
|
|
setEditedOcrText('')
|
|
setCommitResult(null)
|
|
} else {
|
|
setImage(file)
|
|
setImagePreview(fileType === 'image' ? URL.createObjectURL(file) : file)
|
|
setError(null)
|
|
setResult(null)
|
|
setEditedOcrText('')
|
|
setCommitResult(null)
|
|
}
|
|
}, [imagePreview, fileType])
|
|
|
|
const handleSubmit = async () => {
|
|
if (!image) { setError('Please upload an image first'); return }
|
|
setLoading(true)
|
|
setError(null)
|
|
setCommitResult(null)
|
|
try {
|
|
const formData = new FormData()
|
|
formData.append('image', image)
|
|
formData.append('mode', mode)
|
|
formData.append('prompt', prompt)
|
|
formData.append('grounding', mode === 'find_ref')
|
|
formData.append('include_caption', includeCaption)
|
|
formData.append('find_term', findTerm)
|
|
formData.append('schema', '')
|
|
formData.append('base_size', advancedSettings.base_size)
|
|
formData.append('image_size', advancedSettings.image_size)
|
|
formData.append('crop_mode', advancedSettings.crop_mode)
|
|
formData.append('test_compress', advancedSettings.test_compress)
|
|
|
|
const response = await axios.post(`${API_BASE}/ocr`, formData, {
|
|
headers: { 'Content-Type': 'multipart/form-data' },
|
|
})
|
|
setResult(response.data)
|
|
setEditedOcrText(response.data.text || '')
|
|
setCommitResult(null)
|
|
} catch (err) {
|
|
setError(err.response?.data?.detail || err.message || 'An error occurred')
|
|
} finally {
|
|
setLoading(false)
|
|
}
|
|
}
|
|
|
|
const handleNewAnalysis = () => {
|
|
setResult(null)
|
|
setEditedOcrText('')
|
|
setCommitResult(null)
|
|
}
|
|
|
|
const handleCommitJob = useCallback(async () => {
|
|
if (!image) return
|
|
setCommitLoading(true)
|
|
setCommitResult(null)
|
|
try {
|
|
const formData = new FormData()
|
|
formData.append('image', image)
|
|
formData.append('author', metadata.author)
|
|
formData.append('book', metadata.book)
|
|
formData.append('chapter', metadata.chapter)
|
|
formData.append('page', metadata.page)
|
|
formData.append('ocr_text', editedOcrText)
|
|
formData.append('mode', mode)
|
|
|
|
const response = await axios.post(`${API_BASE}/jobs`, formData, {
|
|
headers: { 'Content-Type': 'multipart/form-data' },
|
|
})
|
|
setCommitResult({ success: true, job: response.data })
|
|
} catch (err) {
|
|
setCommitResult({ success: false, error: err.response?.data?.detail || err.message })
|
|
} finally {
|
|
setCommitLoading(false)
|
|
}
|
|
}, [image, editedOcrText, metadata, mode])
|
|
|
|
const handleCopy = useCallback(() => {
|
|
const text = editedOcrText || result?.text
|
|
if (text) navigator.clipboard.writeText(text)
|
|
}, [editedOcrText, result])
|
|
|
|
const handleDownload = useCallback(() => {
|
|
const text = editedOcrText || result?.text
|
|
if (!text) return
|
|
const ext = { plain_ocr: 'txt', describe: 'txt', find_ref: 'txt', freeform: 'txt' }[mode] || 'txt'
|
|
const blob = new Blob([text], { type: 'text/plain' })
|
|
const url = URL.createObjectURL(blob)
|
|
const a = document.createElement('a')
|
|
a.href = url
|
|
a.download = `deepseek-ocr-result.${ext}`
|
|
a.click()
|
|
URL.revokeObjectURL(url)
|
|
}, [editedOcrText, result, mode])
|
|
|
|
const metaField = (key) => (e) => setMetadata(m => ({ ...m, [key]: e.target.value }))
|
|
|
|
return (
|
|
<div className="min-h-screen relative overflow-hidden">
|
|
{/* Animated background */}
|
|
<div className="fixed inset-0 -z-10">
|
|
<div className="absolute inset-0 bg-gradient-to-br from-purple-900/20 via-pink-900/20 to-cyan-900/20" />
|
|
<div className="absolute inset-0 bg-[url('data:image/svg+xml;base64,PHN2ZyB3aWR0aD0iNjAiIGhlaWdodD0iNjAiIHZpZXdCb3g9IjAgMCA2MCA2MCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj48ZyBmaWxsPSJub25lIiBmaWxsLXJ1bGU9ImV2ZW5vZGQiPjxwYXRoIGQ9Ik0zNiAxOGMzLjMxIDAgNiAyLjY5IDYgNnMtMi42OSA2LTYgNi02LTIuNjktNi02IDIuNjktNiA2LTZ6TTI0IDZjMy4zMSAwIDYgMi42OSA2IDZzLTIuNjkgNi02IDYtNi0yLjY5LTYtNiAyLjY5LTYgNi02ek00OCAzNmMzLjMxIDAgNiAyLjY5IDYgNnMtMi42OSA2LTYgNi02LTIuNjktNi02IDIuNjktNiA2LTZ6IiBzdHJva2U9InJnYmEoMTQ3LCA1MSwgMjM0LCAwLjEpIiBzdHJva2Utd2lkdGg9IjIiLz48L2c+PC9zdmc+')] opacity-30" />
|
|
<motion.div
|
|
className="absolute top-20 left-20 w-96 h-96 bg-purple-500/10 rounded-full blur-3xl"
|
|
animate={{ scale: [1, 1.2, 1], opacity: [0.3, 0.5, 0.3] }}
|
|
transition={{ duration: 8, repeat: Infinity, ease: 'easeInOut' }}
|
|
/>
|
|
<motion.div
|
|
className="absolute bottom-20 right-20 w-96 h-96 bg-cyan-500/10 rounded-full blur-3xl"
|
|
animate={{ scale: [1.2, 1, 1.2], opacity: [0.5, 0.3, 0.5] }}
|
|
transition={{ duration: 8, repeat: Infinity, ease: 'easeInOut' }}
|
|
/>
|
|
</div>
|
|
|
|
{/* Header */}
|
|
<header className="sticky top-0 z-50 glass border-b border-white/10">
|
|
<div className="max-w-7xl mx-auto px-6 py-4">
|
|
<div className="flex items-center justify-between">
|
|
<motion.div className="flex items-center gap-3" initial={{ opacity: 0, x: -20 }} animate={{ opacity: 1, x: 0 }}>
|
|
<div className="relative">
|
|
<div className="absolute inset-0 bg-gradient-to-r from-purple-500 to-cyan-500 rounded-xl blur-lg opacity-75" />
|
|
<div className="relative bg-gradient-to-br from-purple-600 to-cyan-500 p-2 rounded-xl">
|
|
<Sparkles className="w-6 h-6" />
|
|
</div>
|
|
</div>
|
|
<div>
|
|
<h1 className="text-2xl font-bold gradient-text">DeepSeek OCR</h1>
|
|
<p className="text-xs text-gray-400">Next-Gen Vision AI</p>
|
|
</div>
|
|
</motion.div>
|
|
|
|
<nav className="flex gap-2">
|
|
{showResultView && (
|
|
<motion.button
|
|
onClick={handleNewAnalysis}
|
|
className="flex items-center gap-2 px-4 py-2 rounded-xl text-sm font-medium glass text-gray-400 hover:bg-white/5 transition-all"
|
|
whileHover={{ scale: 1.02 }} whileTap={{ scale: 0.98 }}
|
|
>
|
|
<ChevronLeft className="w-4 h-4" />
|
|
New Analysis
|
|
</motion.button>
|
|
)}
|
|
<motion.button
|
|
onClick={() => setView('new_job')}
|
|
className={`flex items-center gap-2 px-4 py-2 rounded-xl text-sm font-medium transition-all ${view === 'new_job' ? 'bg-gradient-to-r from-purple-600 to-cyan-600 text-white' : 'glass text-gray-400 hover:bg-white/5'}`}
|
|
whileHover={{ scale: 1.02 }} whileTap={{ scale: 0.98 }}
|
|
>
|
|
<Zap className="w-4 h-4" />
|
|
New Job
|
|
</motion.button>
|
|
<motion.button
|
|
onClick={() => setView('jobs')}
|
|
className={`flex items-center gap-2 px-4 py-2 rounded-xl text-sm font-medium transition-all ${view === 'jobs' ? 'bg-gradient-to-r from-purple-600 to-cyan-600 text-white' : 'glass text-gray-400 hover:bg-white/5'}`}
|
|
whileHover={{ scale: 1.02 }} whileTap={{ scale: 0.98 }}
|
|
>
|
|
<Layers className="w-4 h-4" />
|
|
Browse Jobs
|
|
</motion.button>
|
|
</nav>
|
|
</div>
|
|
</div>
|
|
</header>
|
|
|
|
{/* Main Content */}
|
|
<main className="max-w-7xl mx-auto px-6 py-6">
|
|
<AnimatePresence mode="wait">
|
|
|
|
{/* ── Full-screen OCR result view (plain_ocr + result) ── */}
|
|
{showResultView ? (
|
|
<motion.div
|
|
key="ocr_result"
|
|
initial={{ opacity: 0, y: 20 }}
|
|
animate={{ opacity: 1, y: 0 }}
|
|
exit={{ opacity: 0, y: -20 }}
|
|
className="flex flex-col gap-4"
|
|
>
|
|
{/* Image + Text */}
|
|
<div className="grid gap-6" style={{ gridTemplateColumns: '1fr 1fr', height: '130vh' }}>
|
|
{imagePreview && typeof imagePreview === 'string' ? (
|
|
<div className="glass rounded-2xl overflow-hidden flex items-center justify-center bg-black/20 h-full">
|
|
<img
|
|
src={imagePreview}
|
|
alt="Source"
|
|
className="w-full h-full object-contain"
|
|
/>
|
|
</div>
|
|
) : (
|
|
<div className="glass rounded-2xl flex items-center justify-center h-full">
|
|
<p className="text-gray-500 text-sm">No preview</p>
|
|
</div>
|
|
)}
|
|
<div className="glass rounded-2xl p-4 flex flex-col h-full">
|
|
<p className="text-xs text-gray-400 mb-2 flex-shrink-0">
|
|
OCR Text <span className="text-purple-400">(edit before committing)</span>
|
|
</p>
|
|
<textarea
|
|
value={editedOcrText}
|
|
onChange={e => setEditedOcrText(e.target.value)}
|
|
className="flex-1 w-full bg-transparent text-sm text-gray-200 font-mono resize-none focus:outline-none min-h-0"
|
|
placeholder="OCR text appears here..."
|
|
/>
|
|
</div>
|
|
</div>
|
|
|
|
{/* Metadata row */}
|
|
<div className="glass p-4 rounded-2xl flex-shrink-0">
|
|
<div className="grid grid-cols-4 gap-4">
|
|
{[
|
|
{ key: 'author', label: 'Author', placeholder: 'Author name' },
|
|
{ key: 'book', label: 'Book', placeholder: 'Book title' },
|
|
{ key: 'chapter', label: 'Chapter', placeholder: 'Chapter' },
|
|
{ key: 'page', label: 'Page', placeholder: 'Page number' },
|
|
].map(({ key, label, placeholder }) => (
|
|
<div key={key}>
|
|
<label className="text-xs text-gray-400 mb-1 block">{label}</label>
|
|
<input
|
|
type="text"
|
|
value={metadata[key]}
|
|
onChange={metaField(key)}
|
|
placeholder={placeholder}
|
|
className={INPUT_CLASS}
|
|
/>
|
|
</div>
|
|
))}
|
|
</div>
|
|
</div>
|
|
|
|
{/* Commit row */}
|
|
<div className="flex items-center gap-4 flex-shrink-0">
|
|
<AnimatePresence>
|
|
{commitResult?.success && (
|
|
<motion.div
|
|
initial={{ opacity: 0, x: -10 }} animate={{ opacity: 1, x: 0 }} exit={{ opacity: 0 }}
|
|
className="flex-1 glass p-3 rounded-xl bg-green-500/10 border border-green-500/20"
|
|
>
|
|
<p className="text-xs text-green-400">
|
|
Job saved — ID: <span className="font-mono">{commitResult.job?.id}</span>
|
|
</p>
|
|
</motion.div>
|
|
)}
|
|
{commitResult && !commitResult.success && (
|
|
<motion.div
|
|
initial={{ opacity: 0, x: -10 }} animate={{ opacity: 1, x: 0 }} exit={{ opacity: 0 }}
|
|
className="flex-1 glass p-3 rounded-xl bg-red-500/10 border border-red-500/20"
|
|
>
|
|
<p className="text-xs text-red-400">{commitResult.error}</p>
|
|
</motion.div>
|
|
)}
|
|
</AnimatePresence>
|
|
<motion.button
|
|
onClick={handleCommitJob}
|
|
disabled={commitLoading || commitResult?.success}
|
|
className={`flex items-center gap-2 px-6 py-3 rounded-xl font-medium text-sm transition-all flex-shrink-0 ${
|
|
commitLoading || commitResult?.success
|
|
? 'opacity-50 cursor-not-allowed bg-white/5'
|
|
: 'bg-gradient-to-r from-blue-600 to-indigo-600 hover:from-blue-500 hover:to-indigo-500'
|
|
}`}
|
|
whileHover={!commitLoading && !commitResult?.success ? { scale: 1.02 } : {}}
|
|
whileTap={!commitLoading && !commitResult?.success ? { scale: 0.98 } : {}}
|
|
>
|
|
{commitLoading ? (
|
|
<><Loader2 className="w-4 h-4 animate-spin" /> Committing...</>
|
|
) : commitResult?.success ? (
|
|
<><CheckCircle2 className="w-4 h-4" /> Committed</>
|
|
) : (
|
|
<><Database className="w-4 h-4" /> Commit Job</>
|
|
)}
|
|
</motion.button>
|
|
</div>
|
|
</motion.div>
|
|
|
|
) : view === 'jobs' ? (
|
|
<motion.div
|
|
key="jobs"
|
|
initial={{ opacity: 0, y: 20 }}
|
|
animate={{ opacity: 1, y: 0 }}
|
|
exit={{ opacity: 0, y: -20 }}
|
|
>
|
|
<JobsPanel />
|
|
</motion.div>
|
|
|
|
) : (
|
|
/* ── Upload / Controls layout ── */
|
|
<motion.div
|
|
key="new_job"
|
|
initial={{ opacity: 0, y: 20 }}
|
|
animate={{ opacity: 1, y: 0 }}
|
|
exit={{ opacity: 0, y: -20 }}
|
|
>
|
|
<div className="grid lg:grid-cols-2 gap-6">
|
|
{/* Left Panel */}
|
|
<motion.div
|
|
initial={{ opacity: 0, y: 20 }}
|
|
animate={{ opacity: 1, y: 0 }}
|
|
transition={{ delay: 0.1 }}
|
|
className="space-y-6"
|
|
>
|
|
{/* File Type Toggle */}
|
|
<div className="glass p-4 rounded-2xl">
|
|
<div className="grid grid-cols-2 gap-2">
|
|
<motion.button
|
|
onClick={() => handleFileTypeChange('image')}
|
|
className={`p-3 rounded-xl text-sm font-medium transition-all flex items-center justify-center gap-2 ${fileType === 'image' ? 'bg-gradient-to-r from-purple-600 to-cyan-600 text-white' : 'glass text-gray-400 hover:bg-white/5'}`}
|
|
whileHover={{ scale: 1.02 }} whileTap={{ scale: 0.98 }}
|
|
>
|
|
<ImageIcon className="w-4 h-4" /> Image OCR
|
|
</motion.button>
|
|
<motion.button
|
|
onClick={() => handleFileTypeChange('pdf')}
|
|
className={`p-3 rounded-xl text-sm font-medium transition-all flex items-center justify-center gap-2 ${fileType === 'pdf' ? 'bg-gradient-to-r from-purple-600 to-cyan-600 text-white' : 'glass text-gray-400 hover:bg-white/5'}`}
|
|
whileHover={{ scale: 1.02 }} whileTap={{ scale: 0.98 }}
|
|
>
|
|
<FileText className="w-4 h-4" /> PDF Processing
|
|
</motion.button>
|
|
</div>
|
|
</div>
|
|
|
|
<MetadataForm metadata={metadata} onChange={setMetadata} />
|
|
|
|
<ModeSelector
|
|
mode={mode} onModeChange={setMode}
|
|
prompt={prompt} onPromptChange={setPrompt}
|
|
findTerm={findTerm} onFindTermChange={setFindTerm}
|
|
/>
|
|
|
|
<ImageUpload onImageSelect={handleImageSelect} preview={imagePreview} fileType={fileType} />
|
|
|
|
<motion.button
|
|
onClick={() => setShowAdvanced(!showAdvanced)}
|
|
className="w-full glass px-4 py-3 rounded-2xl flex items-center justify-between hover:bg-white/5 transition-colors"
|
|
whileHover={{ scale: 1.01 }} whileTap={{ scale: 0.99 }}
|
|
>
|
|
<div className="flex items-center gap-2">
|
|
<Settings className="w-4 h-4 text-purple-400" />
|
|
<span className="text-sm font-medium text-gray-300">Advanced Settings</span>
|
|
</div>
|
|
<motion.div animate={{ rotate: showAdvanced ? 180 : 0 }} transition={{ duration: 0.3 }}>
|
|
<svg className="w-4 h-4 text-gray-400" fill="none" stroke="currentColor" viewBox="0 0 24 24">
|
|
<path strokeLinecap="round" strokeLinejoin="round" strokeWidth={2} d="M19 9l-7 7-7-7" />
|
|
</svg>
|
|
</motion.div>
|
|
</motion.button>
|
|
|
|
<AnimatePresence>
|
|
{showAdvanced && (
|
|
<AdvancedSettings
|
|
settings={advancedSettings} onSettingsChange={setAdvancedSettings}
|
|
includeCaption={includeCaption} onIncludeCaptionChange={setIncludeCaption}
|
|
/>
|
|
)}
|
|
</AnimatePresence>
|
|
|
|
{fileType === 'pdf' ? (
|
|
<PDFProcessor
|
|
pdfFile={image} mode={mode} prompt={prompt}
|
|
advancedSettings={advancedSettings} includeCaption={includeCaption}
|
|
/>
|
|
) : (
|
|
<>
|
|
<motion.button
|
|
onClick={handleSubmit}
|
|
disabled={!image || loading}
|
|
className={`w-full relative overflow-hidden rounded-2xl p-[2px] ${!image || loading ? 'opacity-50 cursor-not-allowed' : ''}`}
|
|
whileHover={!loading && image ? { scale: 1.02 } : {}}
|
|
whileTap={!loading && image ? { scale: 0.98 } : {}}
|
|
>
|
|
<div className="absolute inset-0 bg-gradient-to-r from-purple-600 via-pink-600 to-cyan-600 animate-gradient" />
|
|
<div className="relative bg-dark-100 px-8 py-4 rounded-2xl flex items-center justify-center gap-3">
|
|
{loading ? (
|
|
<><Loader2 className="w-5 h-5 animate-spin" /><span className="font-semibold">Processing Magic...</span></>
|
|
) : (
|
|
<><Zap className="w-5 h-5" /><span className="font-semibold">Analyze Image</span></>
|
|
)}
|
|
</div>
|
|
</motion.button>
|
|
|
|
{error && (
|
|
<motion.div
|
|
initial={{ opacity: 0, y: -10 }} animate={{ opacity: 1, y: 0 }}
|
|
className="glass p-4 rounded-2xl border-red-500/50 bg-red-500/10"
|
|
>
|
|
<p className="text-sm text-red-400">{error}</p>
|
|
</motion.div>
|
|
)}
|
|
</>
|
|
)}
|
|
</motion.div>
|
|
|
|
{/* Right Panel - Results (non-plain_ocr modes or loading) */}
|
|
<motion.div
|
|
initial={{ opacity: 0, y: 20 }}
|
|
animate={{ opacity: 1, y: 0 }}
|
|
transition={{ delay: 0.2 }}
|
|
>
|
|
<ResultPanel
|
|
result={result}
|
|
loading={loading}
|
|
imagePreview={imagePreview}
|
|
onCopy={handleCopy}
|
|
onDownload={handleDownload}
|
|
/>
|
|
</motion.div>
|
|
</div>
|
|
</motion.div>
|
|
)}
|
|
</AnimatePresence>
|
|
</main>
|
|
|
|
{/* Footer */}
|
|
<footer className="mt-20 border-t border-white/10 glass">
|
|
<div className="max-w-7xl mx-auto px-6 py-8 text-center space-y-2">
|
|
<p className="text-sm text-gray-400">
|
|
Powered by <span className="gradient-text font-semibold">DeepSeek-OCR</span> •
|
|
Built with <span className="text-pink-400">♥</span> using React + FastAPI
|
|
</p>
|
|
<p className="text-xs text-gray-500">
|
|
Thanks to <a href="https://github.com/p-xiexin" target="_blank" rel="noopener noreferrer" className="text-purple-400 hover:text-purple-300 transition-colors">@p-xiexin</a> for the clipboard paste idea!
|
|
</p>
|
|
</div>
|
|
</footer>
|
|
</div>
|
|
)
|
|
}
|
|
|
|
export default App
|