Update app.py

app.py

@@ -1,33 +1,30 @@
-import gradio as gr
-from paddleocr import PaddleOCR
-import cv2
-import numpy as np
+# 1. Use Python 3.10 (Most stable for Paddle)
+FROM python:3.10
 
-# Initialize OCR engine (loads models into memory once)
-# lang='en' supports English. Change to 'ch' for Chinese/English mix.
-ocr = PaddleOCR(use_angle_cls=True, lang='en', use_gpu=False)
+# 2. Set working directory
+WORKDIR /app
 
-def run_ocr(image):
-    if image is None:
-        return "No image provided"
+# 3. INSTALL MISSING GRAPHICS LIBRARIES (This fixes the "Starting" crash)
+RUN apt-get update && apt-get install -y \
+    libgl1 \
+    libglib2.0-0 \
+    libsm6 \
+    libxext6 \
+    libxrender-dev \
+    && rm -rf /var/lib/apt/lists/*
 
-    # Gradio passes image as numpy array, perfect for PaddleOCR
-    result = ocr.ocr(image, cls=True)
+# 4. Copy requirements and install
+COPY requirements.txt requirements.txt
+RUN pip install --no-cache-dir --upgrade pip && \
+    pip install --no-cache-dir -r requirements.txt
 
-    txts = []
-    if result and result[0]:
-        # Extract just the text from the complex result
-        txts = [line[1][0] for line in result[0]]
+# 5. Pre-download the models so they don't timeout your browser
+# (This runs a dummy scan during build to cache the files)
+RUN pip install paddlepaddle==2.6.2 paddleocr==2.7.3 numpy<2.0.0
+RUN python -c "from paddleocr import PaddleOCR; PaddleOCR(use_angle_cls=False, lang='en', use_gpu=False)"
 
-    return "\n".join(txts)
+# 6. Copy App Code
+COPY . .
 
-# Create the interface
-# The 'api_name' is crucial - it's how your HTML file will find this function
-demo = gr.Interface(
-    fn=run_ocr,
-    inputs=gr.Image(type="numpy"),
-    outputs="text",
-)
-
-if __name__ == "__main__":
-    demo.launch()
+# 7. Start the Server
+CMD ["python", "app.py"]