igneus / adiutor

Chant database assisting my work on the "In adiutorium" project.

Adiutor

Private chant database assisting work on the In adiutorium project. Internal tool, not intended for public deployment.

What it's used for

The application's main purpose is to transform a heap of sheet music into a database of pieces which can be conveniently searched and filtered by lyrics, tune and metadata. Many features are specific to the corpus being worked on (keeping track of relations between pieces, of settings for a few special psalms and canticles required to match the respective antiphons, of non-standard tunes worth occasional re-evaluation; continual focus on pieces marked as needing a revision).

The application is only used as visualization and accessibility tool, not to edit the data. The music corpus is maintained as a source code repository edited using standard music editing applications and the source code is regularly re-imported in the web app. (Hence the feature of opening a selected piece in an external text editor.)

Various publicly available corpora of Gregorian chant pieces are imported as additional reference material and for research purposes not directly related to the work on the vernacular chant corpus.

Setting up

Prerequisites

• Docker
• docker-compose
• for the parts not dockerized:
  • Bash
  • GNU Make (other Make implementations may work, too)
  • Ruby 3.x
  • LilyPond >= 2.18
  • Inkscape
  • LuaLaTeX and gregorio (only if you are going to import and render also gabc-based chant corpora)
  • Verovio (only if you are going to import and render also MEI-based chant corpora)
  • PHP (only if you are going to use the "open in editor" functionality)

Setup steps

Configure and start the dockerized web app

1. ./bin/init.sh
2. spin the application up by ./dc.sh up (Ctrl+C to stop, ./dc.sh down to remove the containers)
3. check that the application works:
   • http://localhost:3000/ serves Adiutor homepage
   • make test succeeds (or at least make test-rails test-python succeeds)
4. optionally create some user account: rake user:add[[email protected]] (only needed if you plan to use the data-modifying functionalities like reviewing parent-child mismatches and marking them resolved)

Import data

Data import is not dockerized, so these steps install dependencies and execute parts of the Rails application outside of Docker. (The dockerized database is still being used.)

1. in section 3 of .env configure local paths to the data of corpora you plan to import
2. cd rails_app
3. bundle install
4. check that the application works outside of Docker: bundle exec rake spec
5. bundle exec rake refresh to import (or re-import) the In adiutorium corpus
6. optionally import more chant corpora for comparative purposes (respective optional settings in .env must be set)
   • gabc-based corpora
     • bundle exec rake gregobase:load_dump gregobase:import to import relevant chants from a GregoBase DB dump https://github.com/gregorio-project/GregoBase
     • bundle exec rake nocturnale:import to import chants from the Nocturnale Romanum project https://github.com/Nocturnale-Romanum/nocturnale-romanum
     • bundle exec rake antiphonarius:import to import chants from the 1960 Liber antiphonarius https://github.com/ahinkley/liber-antiphonarius-1960 (note: if you import GregoBase, complete contents of this dataset are included, so there's no need to import both)
     • bundle exec rake antiphonale83:import to import chants from the 1983 Ordo cantus officii https://github.com/igneus/antiphonale83
   • MEI-based corpora
     • bundle exec rake hughes:import to import chants from the Andrew Hughes chant corpus https://github.com/DDMAL/Andrew-Hughes-Chant
     • bundle exec rake neuma:fetch neuma:import to import chants from the Sequentia collection http://neuma.huma-num.fr/home/corpus/sequentia/
7. bundle exec rake images to render each chant in notation (takes a lot of time to finish, requires LilyPond and Inkscape, for the gabc-based chant corpora also LuaLaTex and Gregorio)
8. cd .. (return to the project root directory)
9. make volpiano to generate normalized representations of melodies (required for music search and other features)