mirror of
https://github.com/Unstructured-IO/unstructured.git
synced 2025-08-20 06:39:41 +00:00
81af879038
### Description In an effort to speed up the ingest tests, bumping the num if processes to the max on the system for each
38 lines
1.4 KiB
Bash
Executable File
38 lines
1.4 KiB
Bash
Executable File
#!/usr/bin/env bash
|
|
|
|
set -e
|
|
|
|
SCRIPT_DIR=$(dirname "$(realpath "$0")")
|
|
cd "$SCRIPT_DIR"/.. || exit 1
|
|
OUTPUT_FOLDER_NAME=Sharepoint
|
|
OUTPUT_DIR=$SCRIPT_DIR/structured-output/$OUTPUT_FOLDER_NAME
|
|
DOWNLOAD_DIR=$SCRIPT_DIR/download/$OUTPUT_FOLDER_NAME
|
|
max_processes=${MAX_PROCESSES:=$(python -c "import os; print(os.cpu_count())")}
|
|
|
|
# shellcheck disable=SC1091
|
|
source "$SCRIPT_DIR"/cleanup.sh
|
|
trap 'cleanup_dir "$OUTPUT_DIR"' EXIT
|
|
|
|
if [ -z "$SHAREPOINT_CLIENT_ID" ] || [ -z "$SHAREPOINT_CRED" ]; then
|
|
echo "Skipping Sharepoint ingest test because the SHAREPOINT_CLIENT_ID or SHAREPOINT_CRED env var is not set."
|
|
exit 0
|
|
fi
|
|
# excluding metadata.last_modified since this will always update as date processed because the Sharepoint connector creates documents on the fly
|
|
PYTHONPATH=. ./unstructured/ingest/main.py \
|
|
sharepoint \
|
|
--download-dir "$DOWNLOAD_DIR" \
|
|
--metadata-exclude file_directory,metadata.data_source.date_processed,metadata.last_modified,metadata.detection_class_prob,metadata.parent_id,metadata.category_depth \
|
|
--num-processes "$max_processes" \
|
|
--strategy hi_res \
|
|
--preserve-downloads \
|
|
--reprocess \
|
|
--output-dir "$OUTPUT_DIR" \
|
|
--verbose \
|
|
--client-cred "$SHAREPOINT_CRED" \
|
|
--client-id "$SHAREPOINT_CLIENT_ID" \
|
|
--site "$SHAREPOINT_SITE" \
|
|
--path "Shared Documents" \
|
|
--recursive \
|
|
|
|
"$SCRIPT_DIR"/check-diff-expected-output.sh $OUTPUT_FOLDER_NAME
|