modified: app.py

This commit is contained in:
Russell Ballestrini 2025-04-11 09:12:21 -04:00
parent c63cfe0de6
commit 36e2c681f9
2 changed files with 6 additions and 2 deletions

View file

@ -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

1
app.py
View file

@ -151,6 +151,7 @@ class FileExtension(str, Enum):
PPTX = ".pptx"
PPT = ".ppt"
JSON = ".json"
LOG = ".log"
MD = ".md"
MARKDOWN = ".markdown"