fix

Dockerfile

@@ -12,8 +12,13 @@ WORKDIR /app
 RUN apt-get update && apt-get install -y \
     gcc \
     g++ \
+    curl \
     && rm -rf /var/lib/apt/lists/*
 
+# Create NLTK data directory with proper permissions
+RUN mkdir -p /usr/local/nltk_data && chmod 755 /usr/local/nltk_data
+ENV NLTK_DATA=/usr/local/nltk_data
+
 # Copy requirements first for better Docker layer caching
 COPY requirements.txt .
 
@@ -21,11 +26,31 @@ COPY requirements.txt .
 RUN pip install --no-cache-dir --upgrade pip && \
     pip install --no-cache-dir -r requirements.txt
 
+# Download all potentially needed NLTK data during build
+# This ensures we have permissions and avoids runtime download issues
+RUN python -c "import nltk; \
+    nltk.download('punkt', download_dir='/usr/local/nltk_data', quiet=True); \
+    nltk.download('punkt_tab', download_dir='/usr/local/nltk_data', quiet=True); \
+    nltk.download('stopwords', download_dir='/usr/local/nltk_data', quiet=True); \
+    nltk.download('averaged_perceptron_tagger', download_dir='/usr/local/nltk_data', quiet=True); \
+    print('NLTK data download completed successfully')"
+
+# Verify NLTK data was downloaded correctly
+RUN python -c "import nltk; \
+    try: \
+        nltk.data.find('tokenizers/punkt'); \
+        print('NLTK punkt tokenizer found successfully'); \
+    except LookupError: \
+        print('Warning: NLTK punkt tokenizer not found'); \
+        exit(1)"
+
 # Copy application code
 COPY app.py .
 
-# Create a non-root user for security (optional but recommended)
-RUN useradd -m -u 1000 appuser && chown -R appuser:appuser /app
+# Create a non-root user for security but ensure they can access NLTK data
+RUN useradd -m -u 1000 appuser && \
+    chown -R appuser:appuser /app && \
+    chmod -R 755 /usr/local/nltk_data
 USER appuser
 
 # Expose the port

app.py

@@ -1,5 +1,24 @@
 import os
 import gradio as gr
+
+# Handle NLTK setup early with proper error handling
+try:
+    import nltk
+    # Ensure NLTK data is available, try to download if missing
+    try:
+        nltk.data.find('tokenizers/punkt')
+    except LookupError:
+        print("NLTK punkt tokenizer not found, attempting to download...")
+        try:
+            nltk.download('punkt', quiet=True)
+            nltk.download('punkt_tab', quiet=True)
+        except Exception as e:
+            print(f"Warning: Could not download NLTK data: {e}")
+            print("This may cause issues with text processing")
+except ImportError:
+    print("NLTK not available, continuing without it")
+
+# Now import LlamaIndex components
 from llama_index.core import SimpleDirectoryReader, VectorStoreIndex
 from llama_index.embeddings.cohere import CohereEmbedding
 from llama_index.llms.groq import Groq

requirements.txt

@@ -13,4 +13,7 @@ transformers>=4.30.0,<5.0.0
 
 # Add explicit dependencies that might be causing issues
 pydantic>=2.0.0,<3.0.0
-fastapi>=0.100.0,<1.0.0
+fastapi>=0.100.0,<1.0.0
+
+# Explicitly include NLTK with a compatible version
+nltk>=3.8,<4.0