Update Dockerfile

Dockerfile

@@ -1,4 +1,3 @@
-# Use slim Python base image
 FROM python:3.10-slim
 
 # Install system-level dependencies
@@ -13,28 +12,36 @@ RUN apt-get update && apt-get install -y \
     ffmpeg \
     git \
     build-essential \
-    && rm -rf /var/lib/apt/lists/*
+    curl \
+ && rm -rf /var/lib/apt/lists/*
 
-# Set environment variables
+# Environment variables
 ENV PYTHONUNBUFFERED=1
 ENV PIP_NO_CACHE_DIR=1
 ENV HF_HUB_DISABLE_SYMLINKS_WARNING=1
+ENV LANG=C.UTF-8
+ENV LC_ALL=C.UTF-8
+ENV TESSERACT_PATH=/usr/bin/tesseract
+ENV PATH="${TESSERACT_PATH}:${PATH}"
 
-# Set working directory
+# Work directory
 WORKDIR /home/user/app
 
-# Copy dependencies and install
+# Install Python dependencies
 COPY requirements.txt .
 RUN pip install --upgrade pip && pip install -r requirements.txt
 
-# Copy app source files
+# Copy application code
 COPY . .
 
-# Download NLTK data (punkt tokenizer)
+# Download NLTK tokenizer
 RUN python -m nltk.downloader punkt
 
-# Expose Gradio's default port
+# Expose port for Gradio
 EXPOSE 7860
 
-# Launch the Gradio app
+# Optional: container healthcheck
+HEALTHCHECK CMD curl --fail http://localhost:7860 || exit 1
+
+# Start app
 CMD ["python", "app.py"]