Spaces:
Sleeping
Sleeping
Upload 6 files
Browse files- Dockerfile.dockerfile +27 -0
- main.py +187 -0
- requirements.txt +13 -0
- static/app.js +123 -0
- static/index.html +67 -0
- static/style.css +52 -0
Dockerfile.dockerfile
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM python:3.10
|
| 2 |
+
|
| 3 |
+
WORKDIR /app
|
| 4 |
+
|
| 5 |
+
# Instalar tesseract en el sistema
|
| 6 |
+
RUN apt-get update && apt-get install -y \
|
| 7 |
+
tesseract-ocr \
|
| 8 |
+
&& rm -rf /var/lib/apt/lists/*
|
| 9 |
+
|
| 10 |
+
# Instalar dependencias para OpenCV
|
| 11 |
+
RUN apt-get update && apt-get install -y \
|
| 12 |
+
libgl1 \
|
| 13 |
+
libglib2.0-0 \
|
| 14 |
+
&& rm -rf /var/lib/apt/lists/*
|
| 15 |
+
|
| 16 |
+
# Copiar e instalar las dependencias de Python
|
| 17 |
+
COPY requirements.txt .
|
| 18 |
+
RUN pip install --no-cache-dir -r requirements.txt
|
| 19 |
+
|
| 20 |
+
# Copiar el código del proyecto
|
| 21 |
+
COPY . .
|
| 22 |
+
|
| 23 |
+
# Exponer el puerto que usa Hugging Face Spaces
|
| 24 |
+
EXPOSE 7860
|
| 25 |
+
|
| 26 |
+
# Comando para arrancar la aplicación con Uvicorn
|
| 27 |
+
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]
|
main.py
ADDED
|
@@ -0,0 +1,187 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from unittest import result
|
| 2 |
+
|
| 3 |
+
from fastapi import FastAPI, File, UploadFile, HTTPException
|
| 4 |
+
from google import genai
|
| 5 |
+
from httpcore import request
|
| 6 |
+
from google.genai import types
|
| 7 |
+
from dotenv import load_dotenv
|
| 8 |
+
from pydantic import BaseModel
|
| 9 |
+
from fastapi.staticfiles import StaticFiles
|
| 10 |
+
from fastapi.responses import FileResponse
|
| 11 |
+
from transformers import pipeline
|
| 12 |
+
import cv2
|
| 13 |
+
from paddleocr import PaddleOCR
|
| 14 |
+
import numpy as np
|
| 15 |
+
import time
|
| 16 |
+
import os
|
| 17 |
+
|
| 18 |
+
MODEL_ID = "models/gemini-2.5-flash"
|
| 19 |
+
MODEL_FALLBACK = "models/gemini-3.1-flash-lite-preview"
|
| 20 |
+
modelo_local = pipeline("text2text-generation", model="google/flan-t5-small")
|
| 21 |
+
|
| 22 |
+
load_dotenv()
|
| 23 |
+
|
| 24 |
+
api_key = os.environ.get("GEMINI_API_KEY")
|
| 25 |
+
|
| 26 |
+
client = genai.Client(api_key=api_key)
|
| 27 |
+
|
| 28 |
+
|
| 29 |
+
config = types.GenerateContentConfig(
|
| 30 |
+
temperature=0,
|
| 31 |
+
)
|
| 32 |
+
|
| 33 |
+
class ChatRequest(BaseModel):
|
| 34 |
+
pregunta: str
|
| 35 |
+
contexto: str = ""
|
| 36 |
+
historial: list[str] = []
|
| 37 |
+
|
| 38 |
+
class ChatResponse(BaseModel):
|
| 39 |
+
respuesta: str
|
| 40 |
+
historial: list[str] = []
|
| 41 |
+
|
| 42 |
+
class OCRResponse(BaseModel):
|
| 43 |
+
texto: str
|
| 44 |
+
|
| 45 |
+
app = FastAPI(title="OCR API", version="1.0.0")
|
| 46 |
+
ocr = PaddleOCR(use_angle_cls=True, lang="es")
|
| 47 |
+
|
| 48 |
+
try:
|
| 49 |
+
local_model = pipeline("text2text-generation", model="google/flan-t5-small")
|
| 50 |
+
except Exception as e:
|
| 51 |
+
print(f"Aviso: No se pudo cargar el modelo local: {e}")
|
| 52 |
+
local_model = None
|
| 53 |
+
|
| 54 |
+
#@app.get("/")
|
| 55 |
+
#def home():
|
| 56 |
+
#return FileResponse("static/index.html")
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
@app.post("/api/chat", response_model=ChatResponse)
|
| 61 |
+
def prompt(request: ChatRequest):
|
| 62 |
+
max_reintentos = 2
|
| 63 |
+
segundos_espera = 10
|
| 64 |
+
e = "Error desconocido"
|
| 65 |
+
|
| 66 |
+
try:
|
| 67 |
+
pregunta = request.pregunta
|
| 68 |
+
historial = request.historial
|
| 69 |
+
texto_ocr = request.contexto # <--- Tomamos el texto enviado por el frontend
|
| 70 |
+
|
| 71 |
+
texto_prompt = f"""
|
| 72 |
+
Eres un asistente que responde preguntas usando SOLO información del documento.
|
| 73 |
+
|
| 74 |
+
REGLAS IMPORTANTES:
|
| 75 |
+
- No copies el documento completo.
|
| 76 |
+
- No repitas texto largo del documento.
|
| 77 |
+
- Extrae SOLO la información necesaria.
|
| 78 |
+
- Si la respuesta no está en el documento, di: "No aparece en el documento".
|
| 79 |
+
- Responde de forma breve y directa.
|
| 80 |
+
|
| 81 |
+
DOCUMENTO:
|
| 82 |
+
\"\"\"{texto_ocr}\"\"\" <--- Cambiado: ahora usa el OCR
|
| 83 |
+
|
| 84 |
+
HISTORIAL:
|
| 85 |
+
{historial}
|
| 86 |
+
|
| 87 |
+
PREGUNTA:
|
| 88 |
+
{pregunta}
|
| 89 |
+
|
| 90 |
+
RESPUESTA:
|
| 91 |
+
"""
|
| 92 |
+
|
| 93 |
+
for intento in range(max_reintentos):
|
| 94 |
+
try:
|
| 95 |
+
response = client.models.generate_content(
|
| 96 |
+
model=MODEL_ID,
|
| 97 |
+
contents=texto_prompt,
|
| 98 |
+
config=config
|
| 99 |
+
)
|
| 100 |
+
return ChatResponse(respuesta=response.text)
|
| 101 |
+
except Exception as ex:
|
| 102 |
+
print(f"{intento+1} - Error Gemini 2.5: {ex}")
|
| 103 |
+
e = ex
|
| 104 |
+
time.sleep(segundos_espera)
|
| 105 |
+
|
| 106 |
+
try:
|
| 107 |
+
response = client.models.generate_content(
|
| 108 |
+
model=MODEL_FALLBACK,
|
| 109 |
+
contents=texto_prompt,
|
| 110 |
+
config=config
|
| 111 |
+
)
|
| 112 |
+
return ChatResponse(respuesta=response.text)
|
| 113 |
+
except Exception as e_fallback:
|
| 114 |
+
print(f"Error en Gemini 1.5 Fallback: {e_fallback}")
|
| 115 |
+
try:
|
| 116 |
+
# Simplificamos el prompt para el modelo local pequeño
|
| 117 |
+
res_local = modelo_local(f"question: {pregunta} context: {texto_ocr}", max_new_tokens=50)
|
| 118 |
+
return ChatResponse(respuesta=res_local[0]['generated_text'])
|
| 119 |
+
except Exception as e_final:
|
| 120 |
+
raise HTTPException(status_code=500, detail="Error en todos los modelos (incluyendo local)")
|
| 121 |
+
|
| 122 |
+
except Exception as e:
|
| 123 |
+
if isinstance(e, HTTPException): raise e
|
| 124 |
+
raise HTTPException(status_code=500, detail=str(e))
|
| 125 |
+
|
| 126 |
+
except Exception as e:
|
| 127 |
+
if isinstance(e, HTTPException): raise e
|
| 128 |
+
raise HTTPException(status_code=500, detail=str(e))
|
| 129 |
+
|
| 130 |
+
|
| 131 |
+
def extraer_texto(resultado_ocr):
|
| 132 |
+
textos = []
|
| 133 |
+
|
| 134 |
+
# PaddleOCR devuelve una lista de páginas.
|
| 135 |
+
# Validamos que el resultado no sea None y que la primera página tenga contenido.
|
| 136 |
+
if resultado_ocr and resultado_ocr[0] is not None:
|
| 137 |
+
for linea in resultado_ocr[0]:
|
| 138 |
+
# linea[1][0] es donde reside el texto detectado
|
| 139 |
+
textos.append(linea[1][0])
|
| 140 |
+
|
| 141 |
+
return " ".join(textos)
|
| 142 |
+
|
| 143 |
+
|
| 144 |
+
def preprocesar_imagen(image_bytes):
|
| 145 |
+
# Convertir bytes a imagen
|
| 146 |
+
nparr = np.frombuffer(image_bytes, np.uint8)
|
| 147 |
+
img = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
|
| 148 |
+
|
| 149 |
+
if img is None:
|
| 150 |
+
raise ValueError("No se pudo decodificar la imagen")
|
| 151 |
+
|
| 152 |
+
# PaddleOCR maneja internamente el binarizado,
|
| 153 |
+
# es mejor enviarle la imagen limpia o solo en gris.
|
| 154 |
+
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
|
| 155 |
+
return gray
|
| 156 |
+
|
| 157 |
+
@app.post("/api/ocr", response_model=OCRResponse)
|
| 158 |
+
async def ocr_image(file: UploadFile = File(...)):
|
| 159 |
+
if file.content_type not in {"image/png", "image/jpeg", "image/jpg", "image/webp"}:
|
| 160 |
+
raise HTTPException(status_code=400, detail="Formato no soportado")
|
| 161 |
+
|
| 162 |
+
try:
|
| 163 |
+
# Leemos los bytes del archivo
|
| 164 |
+
contenido = await file.read()
|
| 165 |
+
|
| 166 |
+
# Preprocesamos y ejecutamos OCR
|
| 167 |
+
img = preprocesar_imagen(contenido)
|
| 168 |
+
|
| 169 |
+
# cls=True activa la clasificación de ángulo si se configuró en la instancia
|
| 170 |
+
result = ocr.ocr(img, cls=True)
|
| 171 |
+
|
| 172 |
+
texto_extraido = extraer_texto(result)
|
| 173 |
+
|
| 174 |
+
if not texto_extraido.strip():
|
| 175 |
+
return OCRResponse(texto="No se detectó texto en la imagen.")
|
| 176 |
+
|
| 177 |
+
return OCRResponse(texto=texto_extraido)
|
| 178 |
+
|
| 179 |
+
except ValueError as e:
|
| 180 |
+
raise HTTPException(status_code=400, detail=str(e))
|
| 181 |
+
except Exception as e:
|
| 182 |
+
raise HTTPException(status_code=500, detail=f"Error OCR: {str(e)}")
|
| 183 |
+
|
| 184 |
+
app.mount("/", StaticFiles(directory="static", html=True), name="static")
|
| 185 |
+
|
| 186 |
+
|
| 187 |
+
|
requirements.txt
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
fastapi==0.136.1
|
| 2 |
+
uvicorn==0.46.0
|
| 3 |
+
python-multipart==0.0.27
|
| 4 |
+
opencv-python-headless==4.13.0.92
|
| 5 |
+
numpy==2.2.6
|
| 6 |
+
paddleocr==2.10.0
|
| 7 |
+
paddlepaddle==2.6.2
|
| 8 |
+
pytesseract==0.3.13
|
| 9 |
+
ultralytics==8.4.45
|
| 10 |
+
google-genai==1.73.1
|
| 11 |
+
huggingface_hub==1.12.2
|
| 12 |
+
transformers==5.7.0
|
| 13 |
+
requests==2.33.1
|
static/app.js
ADDED
|
@@ -0,0 +1,123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
let historial = [];
|
| 2 |
+
let textoOCR = "";
|
| 3 |
+
|
| 4 |
+
// Elementos
|
| 5 |
+
const imageInput = document.getElementById('imageInput');
|
| 6 |
+
const imagePreview = document.getElementById('imagePreview');
|
| 7 |
+
const extractedTextDiv = document.getElementById('extractedText');
|
| 8 |
+
const chatLog = document.getElementById('chatLog');
|
| 9 |
+
const inputPregunta = document.getElementById('pregunta');
|
| 10 |
+
const btnEnviar = document.getElementById('btnEnviar');
|
| 11 |
+
const ocrBadge = document.getElementById('ocrBadge');
|
| 12 |
+
const previewWrapper = document.getElementById('previewWrapper');
|
| 13 |
+
|
| 14 |
+
// --- 1. Manejo de OCR ---
|
| 15 |
+
imageInput.addEventListener('change', async (e) => {
|
| 16 |
+
const file = e.target.files[0];
|
| 17 |
+
if (!file) return;
|
| 18 |
+
|
| 19 |
+
chatLog.innerHTML = ""; // Limpiar chat
|
| 20 |
+
historial = []; // Limpiar historial
|
| 21 |
+
|
| 22 |
+
// Mostrar preview
|
| 23 |
+
const url = URL.createObjectURL(file);
|
| 24 |
+
imagePreview.src = url;
|
| 25 |
+
previewWrapper.classList.remove('hidden');
|
| 26 |
+
ocrBadge.textContent = "Procesando...";
|
| 27 |
+
ocrBadge.style.background = "#eab308"; // Amarillo
|
| 28 |
+
|
| 29 |
+
const formData = new FormData();
|
| 30 |
+
formData.append('file', file);
|
| 31 |
+
|
| 32 |
+
try {
|
| 33 |
+
const res = await fetch('/api/ocr', {
|
| 34 |
+
method: 'POST',
|
| 35 |
+
body: formData
|
| 36 |
+
});
|
| 37 |
+
const data = await res.json();
|
| 38 |
+
|
| 39 |
+
textoOCR = data.texto;
|
| 40 |
+
extractedTextDiv.textContent = textoOCR;
|
| 41 |
+
ocrBadge.textContent = "OCR Listo";
|
| 42 |
+
ocrBadge.style.background = "#22c55e"; // Verde
|
| 43 |
+
|
| 44 |
+
// Habilitar chat
|
| 45 |
+
inputPregunta.disabled = false;
|
| 46 |
+
btnEnviar.disabled = false;
|
| 47 |
+
inputPregunta.focus();
|
| 48 |
+
|
| 49 |
+
agregarMensaje('assistant', 'He leído la imagen. ¿Qué quieres saber sobre ella?');
|
| 50 |
+
} catch (err) {
|
| 51 |
+
ocrBadge.textContent = "Error OCR";
|
| 52 |
+
ocrBadge.style.background = "#ef4444"; // Rojo
|
| 53 |
+
}
|
| 54 |
+
});
|
| 55 |
+
|
| 56 |
+
// --- 2. Manejo de Chat ---
|
| 57 |
+
async function enviarPregunta() {
|
| 58 |
+
const pregunta = inputPregunta.value.trim();
|
| 59 |
+
if (!pregunta) return;
|
| 60 |
+
|
| 61 |
+
agregarMensaje('user', pregunta);
|
| 62 |
+
inputPregunta.value = "";
|
| 63 |
+
|
| 64 |
+
// Mostramos estado de carga
|
| 65 |
+
const typingId = agregarMensaje('assistant', '<i class="fa-solid fa-ellipsis fa-beat"></i> Pensando...');
|
| 66 |
+
|
| 67 |
+
try {
|
| 68 |
+
const res = await fetch('/api/chat', {
|
| 69 |
+
method: 'POST',
|
| 70 |
+
headers: { 'Content-Type': 'application/json' },
|
| 71 |
+
body: JSON.stringify({
|
| 72 |
+
pregunta: pregunta,
|
| 73 |
+
contexto: textoOCR, // <--- CAMBIO: Usamos la variable global 'textoOCR' definida arriba
|
| 74 |
+
historial: historial
|
| 75 |
+
})
|
| 76 |
+
});
|
| 77 |
+
|
| 78 |
+
const data = await res.json();
|
| 79 |
+
|
| 80 |
+
// Eliminamos indicador de carga y ponemos respuesta
|
| 81 |
+
const loader = document.getElementById(typingId);
|
| 82 |
+
if (loader) loader.remove();
|
| 83 |
+
|
| 84 |
+
if (data.respuesta) {
|
| 85 |
+
agregarMensaje('assistant', data.respuesta);
|
| 86 |
+
|
| 87 |
+
// Actualizamos historial
|
| 88 |
+
historial.push(`USER: ${pregunta}`, `ASSISTANT: ${data.respuesta}`);
|
| 89 |
+
if(historial.length > 10) historial.splice(0, 2);
|
| 90 |
+
} else {
|
| 91 |
+
agregarMensaje('assistant', 'La IA no pudo generar una respuesta clara.');
|
| 92 |
+
}
|
| 93 |
+
|
| 94 |
+
} catch (err) {
|
| 95 |
+
const loader = document.getElementById(typingId);
|
| 96 |
+
if (loader) loader.remove();
|
| 97 |
+
agregarMensaje('assistant', 'Lo siento, ha ocurrido un error al conectar con el servidor.');
|
| 98 |
+
console.error(err);
|
| 99 |
+
}
|
| 100 |
+
}
|
| 101 |
+
|
| 102 |
+
function agregarMensaje(rol, texto) {
|
| 103 |
+
const div = document.createElement('div');
|
| 104 |
+
const id = 'msg-' + Date.now() + '-' + Math.floor(Math.random() * 1000);
|
| 105 |
+
div.id = id;
|
| 106 |
+
div.className = `message ${rol}`;
|
| 107 |
+
div.innerHTML = texto;
|
| 108 |
+
chatLog.appendChild(div);
|
| 109 |
+
chatLog.scrollTop = chatLog.scrollHeight;
|
| 110 |
+
return id;
|
| 111 |
+
}
|
| 112 |
+
|
| 113 |
+
// Eventos
|
| 114 |
+
btnEnviar.addEventListener('click', enviarPregunta);
|
| 115 |
+
inputPregunta.addEventListener('keypress', (e) => {
|
| 116 |
+
if (e.key === 'Enter') enviarPregunta();
|
| 117 |
+
});
|
| 118 |
+
|
| 119 |
+
document.getElementById('btnLimpiar').addEventListener('click', () => {
|
| 120 |
+
chatLog.innerHTML = "";
|
| 121 |
+
historial = [];
|
| 122 |
+
agregarMensaje('system', 'Chat reiniciado.');
|
| 123 |
+
});
|
static/index.html
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
<!DOCTYPE html>
|
| 2 |
+
<html lang="es">
|
| 3 |
+
<head>
|
| 4 |
+
<meta charset="UTF-8">
|
| 5 |
+
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
| 6 |
+
<title>VisionAI | OCR & LLM Assistant</title>
|
| 7 |
+
<link rel="stylesheet" href="/static/style.css?v=1.1">
|
| 8 |
+
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.0/css/all.min.css">
|
| 9 |
+
</head>
|
| 10 |
+
<body>
|
| 11 |
+
<div class="app-container">
|
| 12 |
+
<!-- Sidebar: Control de Imagen -->
|
| 13 |
+
<aside class="sidebar">
|
| 14 |
+
<div class="logo">
|
| 15 |
+
<i class="fa-solid fa-eye"></i>
|
| 16 |
+
<span>VisionAI</span>
|
| 17 |
+
</div>
|
| 18 |
+
|
| 19 |
+
<div class="upload-box" id="dropZone">
|
| 20 |
+
<input type="file" id="imageInput" accept="image/*" hidden>
|
| 21 |
+
<label for="imageInput" class="upload-label">
|
| 22 |
+
<i class="fa-solid fa-cloud-arrow-up"></i>
|
| 23 |
+
<span>Sube una imagen o arrastra</span>
|
| 24 |
+
<small>PNG, JPG hasta 10MB</small>
|
| 25 |
+
</label>
|
| 26 |
+
</div>
|
| 27 |
+
|
| 28 |
+
<div id="previewWrapper" class="hidden">
|
| 29 |
+
<div class="preview-card">
|
| 30 |
+
<img id="imagePreview" src="" alt="Preview">
|
| 31 |
+
<div id="ocrBadge" class="badge">Procesando...</div>
|
| 32 |
+
</div>
|
| 33 |
+
<div class="ocr-text-container">
|
| 34 |
+
<label>Texto extraído:</label>
|
| 35 |
+
<div id="extractedText">Sube una imagen para ver el texto...</div>
|
| 36 |
+
</div>
|
| 37 |
+
</div>
|
| 38 |
+
</aside>
|
| 39 |
+
|
| 40 |
+
<!-- Main: Chat Interface -->
|
| 41 |
+
<main class="chat-area">
|
| 42 |
+
<div class="chat-header">
|
| 43 |
+
<h2>Asistente de Documentos</h2>
|
| 44 |
+
<button id="btnLimpiar" class="btn-icon" title="Limpiar chat">
|
| 45 |
+
<i class="fa-solid fa-trash-can"></i>
|
| 46 |
+
</button>
|
| 47 |
+
</div>
|
| 48 |
+
|
| 49 |
+
<div id="chatLog" class="chat-log">
|
| 50 |
+
<div class="message system">
|
| 51 |
+
¡Bienvenido! Sube una imagen con texto para empezar a hacerme preguntas sobre ella.
|
| 52 |
+
</div>
|
| 53 |
+
</div>
|
| 54 |
+
|
| 55 |
+
<div class="input-container">
|
| 56 |
+
<div class="input-wrapper">
|
| 57 |
+
<input type="text" id="pregunta" placeholder="Escribe tu duda aquí..." disabled>
|
| 58 |
+
<button id="btnEnviar" disabled>
|
| 59 |
+
<i class="fa-solid fa-paper-plane"></i>
|
| 60 |
+
</button>
|
| 61 |
+
</div>
|
| 62 |
+
</div>
|
| 63 |
+
</main>
|
| 64 |
+
</div>
|
| 65 |
+
<script src="/static/app.js"></script>
|
| 66 |
+
</body>
|
| 67 |
+
</html>
|
static/style.css
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
:root {
|
| 2 |
+
--primary: #2563eb;
|
| 3 |
+
--primary-hover: #1d4ed8;
|
| 4 |
+
--bg-app: #f8fafc;
|
| 5 |
+
--bg-sidebar: #ffffff;
|
| 6 |
+
--text-main: #1e293b;
|
| 7 |
+
--text-muted: #64748b;
|
| 8 |
+
--border: #e2e8f0;
|
| 9 |
+
}
|
| 10 |
+
|
| 11 |
+
* { box-sizing: border-box; margin: 0; padding: 0; }
|
| 12 |
+
body { font-family: 'Inter', system-ui, sans-serif; background: var(--bg-app); color: var(--text-main); height: 100vh; overflow: hidden; }
|
| 13 |
+
|
| 14 |
+
.app-container { display: flex; height: 100vh; }
|
| 15 |
+
|
| 16 |
+
/* Sidebar */
|
| 17 |
+
.sidebar { width: 350px; background: var(--bg-sidebar); border-right: 1px solid var(--border); padding: 24px; display: flex; flex-direction: column; gap: 20px; }
|
| 18 |
+
.logo { display: flex; align-items: center; gap: 12px; font-size: 1.5rem; font-weight: 700; color: var(--primary); margin-bottom: 10px; }
|
| 19 |
+
|
| 20 |
+
.upload-box { border: 2px dashed var(--border); border-radius: 12px; padding: 30px 20px; text-align: center; transition: all 0.3s; cursor: pointer; }
|
| 21 |
+
.upload-box:hover { border-color: var(--primary); background: #f0f7ff; }
|
| 22 |
+
.upload-label { cursor: pointer; display: flex; flex-direction: column; gap: 8px; color: var(--text-muted); }
|
| 23 |
+
.upload-label i { font-size: 2rem; color: var(--primary); }
|
| 24 |
+
|
| 25 |
+
.preview-card { position: relative; margin-top: 10px; border-radius: 8px; overflow: hidden; border: 1px solid var(--border); }
|
| 26 |
+
#imagePreview { width: 100%; display: block; max-height: 200px; object-fit: contain; background: #000; }
|
| 27 |
+
.badge { position: absolute; bottom: 8px; right: 8px; background: rgba(0,0,0,0.7); color: white; padding: 4px 10px; border-radius: 20px; font-size: 0.7rem; }
|
| 28 |
+
|
| 29 |
+
.ocr-text-container { margin-top: 15px; font-size: 0.85rem; }
|
| 30 |
+
#extractedText { background: #f1f5f9; padding: 12px; border-radius: 8px; height: 120px; overflow-y: auto; color: var(--text-muted); margin-top: 5px; border: 1px solid var(--border); line-height: 1.4; }
|
| 31 |
+
|
| 32 |
+
/* Chat Area */
|
| 33 |
+
.chat-area { flex: 1; display: flex; flex-direction: column; position: relative; }
|
| 34 |
+
.chat-header { padding: 20px 30px; border-bottom: 1px solid var(--border); display: flex; justify-content: space-between; align-items: center; background: white; }
|
| 35 |
+
.chat-log { flex: 1; padding: 30px; overflow-y: auto; display: flex; flex-direction: column; gap: 20px; }
|
| 36 |
+
|
| 37 |
+
.message { max-width: 80%; padding: 12px 16px; border-radius: 12px; line-height: 1.5; font-size: 0.95rem; }
|
| 38 |
+
.message.system { align-self: center; background: #f1f5f9; color: var(--text-muted); font-size: 0.8rem; }
|
| 39 |
+
.message.user { align-self: flex-end; background: var(--primary); color: white; border-bottom-right-radius: 2px; }
|
| 40 |
+
.message.assistant { align-self: flex-start; background: white; border: 1px solid var(--border); border-bottom-left-radius: 2px; box-shadow: 0 2px 5px rgba(0,0,0,0.05); }
|
| 41 |
+
|
| 42 |
+
/* Input */
|
| 43 |
+
.input-container { padding: 30px; background: linear-gradient(transparent, var(--bg-app) 20%); }
|
| 44 |
+
.input-wrapper { display: flex; gap: 12px; background: white; padding: 8px; border-radius: 15px; box-shadow: 0 4px 15px rgba(0,0,0,0.08); border: 1px solid var(--border); }
|
| 45 |
+
input[type="text"] { flex: 1; border: none; padding: 10px 15px; outline: none; font-size: 1rem; }
|
| 46 |
+
button#btnEnviar { background: var(--primary); color: white; border: none; width: 45px; height: 45px; border-radius: 12px; cursor: pointer; transition: 0.3s; }
|
| 47 |
+
button#btnEnviar:disabled { background: var(--border); cursor: not-allowed; }
|
| 48 |
+
button#btnEnviar:hover:not(:disabled) { background: var(--primary-hover); transform: scale(1.05); }
|
| 49 |
+
|
| 50 |
+
.hidden { display: none; }
|
| 51 |
+
.btn-icon { background: none; border: none; color: var(--text-muted); cursor: pointer; font-size: 1.1rem; }
|
| 52 |
+
.btn-icon:hover { color: #ef4444; }
|