blearn

Utilities to work with BlackBoard Learn in offline mode (i.e., not with its API).

Package under construction:

• No that much documentation yet.
  • Best entry points are CLI tools starting with blearn- (they have docs).
• No tests yet (it would require widespread availability of test courses for known Learn versions to make them truly reproducible).
• Things can easily break with Learn updates.
• Offline marking of non-anonymous submissions still needs some love to handle reporting of some cases (e.g., what happens with repeated submissions).

Installation

Regular installation for python packages only available at git apply. The package is being developed with flit, so flit install --symlink works great. Since the package includes web automation routines with playwright, it may be necessary to also install its tools with playwright install after this package is installed to make browsers available.

Functionality

• Parse offline grading sheet.
• Parse bundled submission in zip file or extracted zip file:
  • Metadata based on file names.
  • Metadata from the content in the log .txt file of the submission.
• Expand grading sheet and bundled submission into normalised folders:
  • All submitted files per user are bundled together if they were not.
  • Links to per-user folder are added to the grading sheet for ease of use.
• Support for anonymous marking that is not made available by Learn (!).

⚠️ If you are using macOS: Apple's sandboxing, and its implementation in Microsoft Office, might prevent hyperlinks from working. To see if this is the case, move all associated files to ~/Library/Group Containers/<code>.Office/. Because Microsoft Office has rights for this folder, hyperlinks should then work. (Note that, as of macOS 12.3 (Monterey), giving Excel Full disk access does not solve this issue.)

⚠️ If you are using folders synced to the cloud: In cases like OneDrive on Windows and Excel, the OS might redirect paths to files synced to the cloud by a URL. In that case, it might be preferable to work on a fully offline location.