yujunjun/unstructured
1
0
Fork 0
mirror of https://github.com/Unstructured-IO/unstructured.git synced 2025-12-26 22:55:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

built(test): exclude version metadata from google drive test (#1682)

Browse Source
This commit is contained in:
David Potter 2023-10-07 19:34:32 -07:00 committed by GitHub
parent 46cb1b642a
commit 8b93217a33
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 1 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
test_unstructured_ingest/expected-structured-output/google-drive/117qrVqiCoR5EjYMsDHGdy3UMkEtKr9Q8-test-drive-doc.docx.json
View File

@ -5,7 +5,6 @@
"metadata": {
"data_source": {
"url": "https://drive.google.com/uc?id=117qrVqiCoR5EjYMsDHGdy3UMkEtKr9Q8&export=download",
"version": "15",
"record_locator": {
"drive_id": "1OQZ66OHBE30rNsNa7dweGLfRmXvkT_jr",
"file_id": "117qrVqiCoR5EjYMsDHGdy3UMkEtKr9Q8"
@ -29,7 +28,6 @@
"metadata": {
"data_source": {
"url": "https://drive.google.com/uc?id=117qrVqiCoR5EjYMsDHGdy3UMkEtKr9Q8&export=download",
"version": "15",
"record_locator": {
"drive_id": "1OQZ66OHBE30rNsNa7dweGLfRmXvkT_jr",
"file_id": "117qrVqiCoR5EjYMsDHGdy3UMkEtKr9Q8"

1
test_unstructured_ingest/expected-structured-output/google-drive/1SpQuE7jHz9nMt5hfQXsiok1SgIdRYX5o-fake.docx.json
View File

@ -5,7 +5,6 @@
"metadata": {
"data_source": {
"url": "https://drive.google.com/uc?id=1SpQuE7jHz9nMt5hfQXsiok1SgIdRYX5o&export=download",
"version": "17",
"record_locator": {
"drive_id": "1OQZ66OHBE30rNsNa7dweGLfRmXvkT_jr",
"file_id": "1SpQuE7jHz9nMt5hfQXsiok1SgIdRYX5o"

1
test_unstructured_ingest/expected-structured-output/google-drive/1cTKXAreuj-wYmL38nFnqKvz3X8UKcaMC-foo.txt.json
View File

@ -5,7 +5,6 @@
"metadata": {
"data_source": {
"url": "https://drive.google.com/uc?id=1cTKXAreuj-wYmL38nFnqKvz3X8UKcaMC&export=download",
"version": "9",
"record_locator": {
"drive_id": "1OQZ66OHBE30rNsNa7dweGLfRmXvkT_jr",
"file_id": "1cTKXAreuj-wYmL38nFnqKvz3X8UKcaMC"

2
test_unstructured_ingest/test-ingest-google-drive.sh
View File

@ -35,7 +35,7 @@ echo "$GCP_INGEST_SERVICE_KEY" >"$GCP_INGEST_SERVICE_KEY_FILE"
PYTHONPATH=. unstructured/ingest/main.py \
google-drive \
--download-dir "$DOWNLOAD_DIR" \
--metadata-exclude coordinates,filename,file_directory,metadata.data_source.date_processed,metadata.last_modified,metadata.detection_class_prob,metadata.parent_id,metadata.category_depth \
--metadata-exclude coordinates,filename,file_directory,metadata.data_source.date_processed,metadata.last_modified,metadata.detection_class_prob,metadata.parent_id,metadata.category_depth,metadata.data_source.version \
--num-processes "$max_processes" \
--strategy hi_res \
--preserve-downloads \