diff --git a/Dockerfile b/Dockerfile
index 3c85c43..5f9e4df 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -31,9 +31,12 @@ COPY . .
 
 RUN pip install --no-cache-dir -r requirements.lock
 
+# Set NLTK data directory environment variable explicitly
+ENV NLTK_DATA=/usr/local/share/nltk_data
+
 RUN playwright install --with-deps && \
-    python -c "from unstructured.nlp.tokenize import download_nltk_packages; download_nltk_packages()" && \
-    python -c "import nltk;nltk.download('punkt_tab'); nltk.download('averaged_perceptron_tagger_eng')" && \
+    mkdir -p $NLTK_DATA && \
+    python -c "import nltk; nltk.download('punkt', download_dir='$NLTK_DATA'); nltk.download('averaged_perceptron_tagger', download_dir='$NLTK_DATA')" && \
     python -c "from unstructured.partition.model_init import initialize; initialize()"
 
 EXPOSE 8001
diff --git a/app.py b/app.py
index f12fa25..e5ae9b1 100644
--- a/app.py
+++ b/app.py
@@ -151,6 +151,7 @@ class FileExtension(str, Enum):
     PPTX = ".pptx"
     PPT = ".ppt"
     JSON = ".json"
+    LOG = ".log"
     MD = ".md"
     MARKDOWN = ".markdown"