| DEPRECATED / NO LONGER WORKS |
|---|
Deprecation note: Google kept changing their URL system. Finally I decided to share science data the right way with Zenodo. For collaborations I use Google Team Drive and rclone. |
| Usage (no longer works) |
| ===== |
Copy the URL of the publicly-shared Google Drive folder from which you want to download files: |
|
| -o output directory |
| -c clobber (overwrite already downloaded files) |
-v verbose |
| =============== =========================== |
| file description |
| =============== =========================== |
| public.py Download publicly-shared Google Drive files without login |
| put.py use drive-google to upload to Google Drive |
| get.py use drive-google to download from Google Drive |
| rename_lower.py use drive-google to rename all files to lower case on Google Drive |
| deinit.py use drive-google to recursively deinit (disconnect) from Google Drive |
| pydrv.py use PyDrive (use public.py instead) |
Normal users don't care about these notes below.
I used golang-based drive-google instead of PyDrive. Even though the files may be shared publicly, with PyDrive you still need an API key client_secrets.json This may be acquired via Google API console API Manager > Credentials > Create Project (if not already created) -- any name is fine > Create Credentials > OAuth client ID > Web Application (any name is fine) Then in the Credentials list, at the far right of the key row, there is a download icon. Save it into your code directory as client_secrets.json Note, this gives the Python application a view into your entire Google Drive, so perhaps for best security create a throwaway account used just for this.
Currently, the code underlying PyDrive (google-api-python-client) has a memory leak, so the code automatically stops after it's used too much RAM. Keep an eye on it so your computer doesn't crash.
Used by Selenium to invoke a Chrome browser.
Linux:
sudo apt-get install chromium-chromedriver