Playbills annotation project for LibCrowds, designed for use with the libcrowds-bs4-pybossa-theme.
Choose a set of tasks from tasks.json (e.g. "titles")
and either an Aleph system number from the file
ark_and_aleph_system_numbers.csv or a
JSON file where the info field contains the keys aleph_sys_no, image_ark
and regions.
When generating tasks from an Aleph system number a new task will be created for each permutation of image and task in the chosen set. When generating tasks from a JSON file (which will probably be the result of a previous annotation project) a new task will be created for all permutations of an image, each region now associated with that image and each task in the chosen task set. The idea being that we can chain tasks to link particular categories of data together and reduce the size of each task. For example, annotate all actors associated with a title, rather than annotate all actors on a page.
To generate a new project and push it to the server install and configure pbs, then:
pip install -r requirements.txt
python generate_project.py <task set> [--sysno=<sysno> or --json=<path>]
cd gen
pbs create_project
pbs add_tasks --tasks-file=tasks.csv
pbs update-task-redundancy --redundancy 3
pbs update_project
Now visit the project settings page and update the category, webhook and thumbnail. The project is now ready to be published.
Each task run will store the region and transcription data. Once these have been compared and processed the final result associated with each task will be updated to store the annotations according to the W3C Annotation Data Model.
Some of the ark identifiers in ark_and_aleph_system_numbers.csv do not point to images that can be retreived via the BL IIIF API. For now, these rows are being moved to bad_arks.csv. If you receive an error message stating that a bad ark has been find while generating the project just copy that row over to bad_arks.csv and run the script again. We'll deal with these later!