Fix: Run apt-get as root before switching to user in Dockerfile

Dockerfile

@@ -3,41 +3,44 @@
 # Use a base image with Python installed. Python 3.10 is fine.
 FROM python:3.10-slim-buster
 
-# Create the user and set it for subsequent commands
-RUN useradd -m -u 1000 user
-USER user
-
 # Set the working directory inside the container
 WORKDIR /app
 
-# Set the Hugging Face cache directory to a path where the 'user' has write permissions.
-# This prevents permission issues with /root/.cache and similar system paths.
-ENV HF_HOME=/app/huggingface_cache
-
-# Create the cache directory and ensure the 'user' owns it
-RUN mkdir -p $HF_HOME && chown user:user $HF_HOME
-
 # Install system dependencies needed for pdf2image (Poppler) and pytesseract (Tesseract)
+# These RUN commands need to be executed as root.
+# Added Arabic and French Tesseract language packs as per your request.
 RUN apt-get update && apt-get install -y \
     libpoppler-dev \
     tesseract-ocr \
     tesseract-ocr-eng \
     tesseract-ocr-tur \
-    tesseract-ocr-fra \
     tesseract-ocr-ara \
-    # Add other languages if needed, e.g., tesseract-ocr-ara tesseract-ocr-fra
+    tesseract-ocr-fra \
+    # Add other languages if needed
     && rm -rf /var/lib/apt/lists/*
 
+# Now, create the user and switch to it for non-root operations.
+# All subsequent commands (COPY, RUN for pip, CMD) will be executed as 'user'.
+RUN useradd -m -u 1000 user
+USER user
+
+# Set the Hugging Face cache directory to a path where the 'user' has write permissions.
+ENV HF_HOME=/app/huggingface_cache
+
+# Create the cache directory and ensure the 'user' owns it
+RUN mkdir -p $HF_HOME && chown user:user $HF_HOME
+
 # Copy requirements.txt and install Python dependencies
+# Ensure the user has ownership of copied files.
 COPY --chown=user:user requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
 # Clean up any old Hugging Face Hub lock files before copying the application code.
-# This helps prevent PermissionErrors if a previous download attempt was interrupted.
 RUN rm -rf "${HF_HOME}/hub/tmp"
 RUN find "${HF_HOME}/hub/models--" -name "*.lock" -type f -delete || true
 
 # Copy the rest of your application code
+# Ensure the user has ownership of copied files.
 COPY --chown=user:user . /app
 
 # Set environment variables for Tesseract and potentially for Python's sqlite3 module