Update Dockerfile

Dockerfile

@@ -3,41 +3,51 @@ FROM python:3.10-slim
 # Set working directory to /app
 WORKDIR /app
 
-# 1. Install System Dependencies
+# 1. Set Environment Variables for Llama Build
+# This forces the installer to compile from source with OpenBLAS (CPU speed boost)
+ENV PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1 \
+    CMAKE_ARGS="-DGGML_BLAS=ON -DGGML_OPENBLAS=ON" \
+    FORCE_CMAKE=1 \
+    STREAMLIT_SERVER_PORT=7860 \
+    STREAMLIT_SERVER_ADDRESS=0.0.0.0 \
+    STREAMLIT_SERVER_FILE_WATCHER_TYPE=none
+
+# 2. Install System Dependencies (Merged)
+# Added: cmake, g++, libopenblas-dev (For Llama)
+# Kept: tesseract-ocr, poppler-utils (For your document tools)
 RUN apt-get update && apt-get install -y \
     build-essential \
+    cmake \
+    g++ \
+    libopenblas-dev \
     curl \
     git \
     tesseract-ocr \
     poppler-utils \
     && rm -rf /var/lib/apt/lists/*
 
-# 2. Install Python Libraries
-# Copy requirements from root (since it's not in src)
+# 3. Install Python Libraries
+# We separate llama-cpp-python to ensure it builds with the flags set above
 COPY requirements.txt .
-RUN pip3 install --no-cache-dir -r requirements.txt
 
-# 3. Download NLP Models
-RUN python -m spacy download en_core_web_sm
-RUN python -m nltk.downloader stopwords wordnet omw-1.4
+RUN pip3 install --no-cache-dir --upgrade pip && \
+    pip3 install --no-cache-dir llama-cpp-python && \
+    pip3 install --no-cache-dir -r requirements.txt
 
-# 4. Copy Application Code
-# This copies the 'src' folder and everything else into /app
+# 4. Download NLP Models (Spacy/NLTK)
+RUN python3 -m spacy download en_core_web_sm
+RUN python3 -m nltk.downloader stopwords wordnet omw-1.4
+
+# 5. Copy Application Code
 COPY . .
 
-# 5. Global Permissions Fix
-# This ensures HF user has write access to /app/src/chroma_db if it gets created there
+# 6. Global Permissions Fix
+# Ensures write access for user data/uploads
 RUN chmod -R 777 /app
 
-# 6. CONFIGURATION
-# Force Streamlit to use the correct port
-ENV STREAMLIT_SERVER_PORT=7860
-ENV STREAMLIT_SERVER_ADDRESS=0.0.0.0
-ENV STREAMLIT_SERVER_FILE_WATCHER_TYPE=none
-
-# Expose the port
+# 7. Expose the port
 EXPOSE 7860
 
-# 7. Start the App
-# CRITICAL FIX: We tell Streamlit to run the file inside the 'src' folder
+# 8. Start the App
 ENTRYPOINT ["streamlit", "run", "src/app.py", "--server.port=7860", "--server.address=0.0.0.0", "--server.enableCORS=false", "--server.enableXsrfProtection=false"]