Update Dockerfile

Dockerfile

@@ -1,37 +1,37 @@
 FROM python:3.10-slim
 
-# system deps (optional but nice)
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    git curl && rm -rf /var/lib/apt/lists/*
+RUN apt-get update && apt-get install -y --no-install-recommends git curl && \
+    rm -rf /var/lib/apt/lists/*
 
 WORKDIR /app
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
-# copy app
 COPY . .
 
-# cache models/datasets to persistent disk on Spaces
+# Use /data for caches and our index
 ENV HF_HOME=/data/.huggingface
-
-# default envs (override in Space Settings if you like)
+ENV DATA_ROOT=/data
 ENV CORPUS_REPO=Azizahalq/materialmind-corpus
-ENV INDEX_UUID=space
 ENV EMB_PROVIDER=hf
 ENV EMB_MODEL=BAAI/bge-small-en-v1.5
-ENV INDEX_DIR=MaterialMind/index/chroma_v3/${INDEX_UUID}
+ENV INDEX_UUID=space
 ENV INDEX_COLLECTION=materialmind
+ENV INDEX_DIR=/data/MaterialMind/index/chroma_v3/${INDEX_UUID}
 ENV PORT=7860
 
-# On container start:
-# 1) if the index folder doesn't exist, build it from the HF dataset
-# 2) start the Flask app with gunicorn
+# On start: build index under /data if missing, then run gunicorn
 CMD bash -lc '\
-  if [ ! -d "MaterialMind/index/chroma_v3/${INDEX_UUID}" ]; then \
-    echo "[BOOT] Building index from $CORPUS_REPO into ${INDEX_DIR}..."; \
-    python build_index_from_hf.py --repo ${CORPUS_REPO} --split train --uuid ${INDEX_UUID} --collection ${INDEX_COLLECTION}; \
+  if [ ! -d "/data/MaterialMind/index/chroma_v3/${INDEX_UUID}" ]; then \
+    echo "[BOOT] Building index from $CORPUS_REPO into /data/MaterialMind/index/chroma_v3/${INDEX_UUID}..."; \
+    python build_index_from_hf.py \
+      --repo ${CORPUS_REPO} \
+      --split train \
+      --out_dir /data/MaterialMind/index/chroma_v3 \
+      --uuid ${INDEX_UUID} \
+      --collection ${INDEX_COLLECTION}; \
   else \
-    echo "[BOOT] Found existing index at ${INDEX_DIR}"; \
+    echo "[BOOT] Found existing index at /data/MaterialMind/index/chroma_v3/${INDEX_UUID}"; \
   fi; \
   gunicorn -w 2 -k gthread -t 120 -b 0.0.0.0:${PORT} app:app \
 '