Update Dockerfile

Dockerfile

@@ -1,37 +1,45 @@
 # ── Hugging Face Spaces / Docker deployment ───────────────────────────────────
 # Port 7860 is required for HF Spaces
-
+ 
 FROM python:3.11-slim
-
+ 
 # System deps for chromadb and sentence-transformers
 RUN apt-get update && apt-get install -y --no-install-recommends \
     gcc g++ libgomp1 && \
     rm -rf /var/lib/apt/lists/*
-
+ 
 WORKDIR /app
-
+ 
 # Install Python deps first (cached layer)
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
-
+ 
 # Copy all flat-uploaded files
 COPY . .
-
+ 
 # Reorganise flat upload into the directory structure the app expects:
 #   static/  → index.html, manifest.json, sw.js
 #   knowledge/ → data_engineering_patterns.pdf
 RUN mkdir -p static knowledge && \
     mv index.html manifest.json sw.js static/ 2>/dev/null || true && \
     mv data_engineering_patterns.pdf knowledge/ 2>/dev/null || true
-
+ 
 # HF Spaces runs as non-root
 RUN useradd -m -u 1000 user
 USER user
 ENV HOME=/home/user PATH=/home/user/.local/bin:$PATH
-
+ 
+# ── UTF-8 locale ──────────────────────────────────────────────────────────────
+# PDF extractors emit Unicode structural chars (U+2028, U+00A0, etc). Without
+# these env vars, Python defaults to ASCII inside a minimal slim container and
+# httpx raises UnicodeEncodeError when sending them upstream to Groq.
+ENV LANG=C.UTF-8 \
+    LC_ALL=C.UTF-8 \
+    PYTHONIOENCODING=utf-8
+ 
 EXPOSE 7860
-
+ 
 # PORT env var is set automatically by HF Spaces to 7860
 ENV PORT=7860
-
-CMD ["python", "app.py"]
+ 
+CMD ["python", "app.py"]