paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-04-02 13:45:10 -05:00
Code Issues Packages Projects Releases Wiki Activity

Adds skipping of NLTK data download if the feature appears disabled

Browse Source
This commit is contained in:
Trenton H 2022-10-05 11:01:45 -07:00
parent d1a17480ea
commit d08eb0c66b
1 changed files with 15 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
docker/docker-entrypoint.sh
View File

@ -56,15 +56,24 @@ map_folders() {
nltk_data () {
# Store the NLTK data outside the Docker container
local nltk_data_dir="${DATA_DIR}/nltk"
readonly truthy_things=("yes y 1 t true")
# Download or update the snowball stemmer data
python3 -W ignore::RuntimeWarning -m nltk.downloader -d "${nltk_data_dir}" snowball_data
# If not set, or it looks truthy
if [[ -z "${PAPERLESS_ENABLE_NLTK}" ]] || [[ "${truthy_things[*]}" =~ ${PAPERLESS_ENABLE_NLTK,} ]]; then
# Download or update the stopwords corpus
python3 -W ignore::RuntimeWarning -m nltk.downloader -d "${nltk_data_dir}" stopwords
# Download or update the snowball stemmer data
python3 -W ignore::RuntimeWarning -m nltk.downloader -d "${nltk_data_dir}" snowball_data
# Download or update the punkt tokenizer data
python3 -W ignore::RuntimeWarning -m nltk.downloader -d "${nltk_data_dir}" punkt
# Download or update the stopwords corpus
python3 -W ignore::RuntimeWarning -m nltk.downloader -d "${nltk_data_dir}" stopwords
# Download or update the punkt tokenizer data
python3 -W ignore::RuntimeWarning -m nltk.downloader -d "${nltk_data_dir}" punkt
else
echo "Skipping NLTK data download"
fi
}