Update PDF to Markdown converter API with NVIDIA L4 support

Dockerfile

@@ -25,11 +25,25 @@ RUN apt-get update && \
 
 RUN update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.10 1
 
-# Create app directory
+# Create user with UID 1000 (required for Hugging Face Spaces)
+RUN useradd -m -u 1000 user
+
+# Create necessary directories and set permissions
+RUN mkdir -p /app /app/docker_mineru /app/docker_mineru/output /app/docker_mineru/output/images && \
+    chown -R user:user /app
+
+# Switch to user
+USER user
+
+# Set home directory
+ENV HOME=/home/user \
+    PATH=/home/user/.local/bin:$PATH
+
+# Set working directory
 WORKDIR /app
 
-# Copy requirements first
-COPY requirements.txt .
+# Copy requirements first (with correct ownership)
+COPY --chown=user requirements.txt .
 
 # Install PyTorch dependencies with explicit compatible versions for NVIDIA L4
 RUN pip3 install --no-cache-dir --upgrade pip && \
@@ -37,8 +51,8 @@ RUN pip3 install --no-cache-dir --upgrade pip && \
     pip3 install --no-cache-dir transformers==4.36.2 && \
     pip3 install --no-cache-dir -r requirements.txt
 
-# Copy the rest of the application
-COPY . .
+# Copy the rest of the application with correct ownership
+COPY --chown=user . .
 
 # Expose port
 EXPOSE 7860

app/main.py

@@ -16,8 +16,8 @@ sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from pdf_converter import convert_pdf_to_md
 
 # --- Configuration for output directory ---
-# Always use the local output directory
-output_dir = "docker_mineru/output"
+# In Docker container, use /app prefix
+output_dir = "/app/docker_mineru/output"
 
 images_dir = os.path.join(output_dir, "images")