mirror of
https://github.com/Unstructured-IO/unstructured.git
synced 2025-07-15 21:15:55 +00:00
11cdd8d71f
### Description In an effort to mitigate resource consumption when running CI tests, cleanup download dir for ingest tests after each one.
45 lines
1.5 KiB
Bash
Executable File
45 lines
1.5 KiB
Bash
Executable File
#!/usr/bin/env bash
|
|
set -e
|
|
|
|
# Description: This test checks if all the processed content is the same as the expected outputs
|
|
|
|
SCRIPT_DIR=$(dirname "$(realpath "$0")")
|
|
cd "$SCRIPT_DIR"/.. || exit 1
|
|
|
|
OUTPUT_FOLDER_NAME=confluence-diff
|
|
OUTPUT_DIR=$SCRIPT_DIR/structured-output/$OUTPUT_FOLDER_NAME
|
|
DOWNLOAD_DIR=$SCRIPT_DIR/download/$OUTPUT_FOLDER_NAME
|
|
max_processes=${MAX_PROCESSES:=$(python3 -c "import os; print(os.cpu_count())")}
|
|
CI=${CI:-"false"}
|
|
|
|
# shellcheck disable=SC1091
|
|
source "$SCRIPT_DIR"/cleanup.sh
|
|
function cleanup() {
|
|
cleanup_dir "$OUTPUT_DIR"
|
|
if [ "$CI" == "true" ]; then
|
|
cleanup_dir "$DOWNLOAD_DIR"
|
|
fi
|
|
}
|
|
trap cleanup EXIT
|
|
|
|
if [ -z "$CONFLUENCE_USER_EMAIL" ] || [ -z "$CONFLUENCE_API_TOKEN" ]; then
|
|
echo "Skipping Confluence ingest test because the CONFLUENCE_USER_EMAIL or CONFLUENCE_API_TOKEN env var is not set."
|
|
exit 0
|
|
fi
|
|
|
|
PYTHONPATH=. ./unstructured/ingest/main.py \
|
|
confluence \
|
|
--download-dir "$DOWNLOAD_DIR" \
|
|
--metadata-exclude filename,file_directory,metadata.data_source.date_processed,metadata.last_modified,metadata.detection_class_prob,metadata.parent_id,metadata.category_depth \
|
|
--num-processes "$max_processes" \
|
|
--preserve-downloads \
|
|
--reprocess \
|
|
--output-dir "$OUTPUT_DIR" \
|
|
--verbose \
|
|
--url https://unstructured-ingest-test.atlassian.net \
|
|
--user-email "$CONFLUENCE_USER_EMAIL" \
|
|
--api-token "$CONFLUENCE_API_TOKEN" \
|
|
--spaces testteamsp,MFS
|
|
|
|
"$SCRIPT_DIR"/check-diff-expected-output.sh $OUTPUT_FOLDER_NAME
|