Media Downloader

[UnwashedMeme]

Request coming out of idigBio Summit VI is for a way to download large segments of media images for local review/processing.

Desired Workflow

• A user would need a working python environment with the idigbio-python-client installed.
• The user would copy a file from the examples folder and modify that to fit their particular need. E.g.
  • Change the search critera
  • Specify a txt/csv file listing media record uuids they would like downloaded
  • Specify an image size {webview, fullsize, thumbnail}; default: webview
  • Specify an output directory; default to a subdir of current dir
  • Specify an output filename format, default to: {mediarecord_uuid}-{size}.jpg
• Run the python file to perform the download
  • It should have some basic output of the progress
  • It should resume gracefully if interrupted

Work to be done:

• In the python client core:
  • Implement a downloader that takes a list of mediarecord uuids, size, directory, filename format.
  • Ensure that it can be fed from search results.
• Write an example file that shows the use with several of the different options in the above workflow
• Test and document
• publish

[danstoner]

Passing on additional clarification from user:

Happy Holidays! In working with a couple datasets from iDigBio, I've discovered that it would be helpful to be able to input a list of GUIDs and get out the images associated with them. This would be useful for when I need to code label data (e.g., date) that the data provider did not include in the dwc fields.

Just a thought from someone working with the data! Thanks, as always, for your work. ~Katie

[danstoner]

Given the amount of user-preference built into the workflow of downloading media, it did not feel right to me to try to put that functionality into the core client itself.

I implemented this requested feature by writing scripts and documentation in examples/fetch_media.

fetch_media.py can do most of the above.

released in v0.8.4. bf4a3670273425401feee7160c51aafa376e6f69

Closing.