Create and visualize textual notes annotations

[jacobwegner]

e.g. https://beyond-translation.perseus.org/reader/urn:cts:engLit:mds822-32.tpsthl-1599.pdl-eng:1-6?mode=commentaries

Assumes:

• A tokenized version of the underlying base text (e.g. mds822-32.tpsthl-1599.pdl-eng.txt *
• A file similar to text_annotations_mds822-32.tpsthl-1599.pdl-eng.json.
• The file should have an array of commentary objects with each commentary object having the following properties:
  • references ties the passage to the annotation
  • fragment is the value to show in the textual notes widget list ("live with mee" in the screenshot above)
  • commentary is the HTML to be displayed (<span>If thou wilt live</span>)
  • ve_refs is an array of the token ve_ref values (used for the highlighting functionality when the "Textual Notes" display mode is activated)
  • idx as a zero-based index. This will be used for sorting each "commentary" / textual note in the textual notes widget list
  • urn is the URN for each commentary object (e.g. urn:cite2:beyond-tranlsation:commentary.atlas_v1:amedsaid1831-dw042.1)
  • witnesses as an array with
  • value as the abbreviated value, e.g. Rs
  • label as the witness label, e.g. MS Rodenbach

* We can use these instructions to generate a CSV similar to https://gist.github.com/jacobwegner/3a96e1763b7bc22d827680db1351a377)

[jacobwegner]

@jchill-git I've left some breadcrumbs above; there are a few more things I'd like to surface as documentation on other features supported by the widget or to simplify things (one example being that we could have a "default" witness if we don't have multiple "sources").

These annotations would be loaded via the scaife_viewer.atlas.importers.text_annotations.import_text_annotations pipeline in settings.SV_ATLAS_INGESTION_PIPELINE or via:

# assumes python manage.py shell_plus

from scaife_viewer.atlas.importers.text_annotations.import_text_annotations
import_text_annotations(reset=True)

[jacobwegner]

(I think longer term we may have a commentary / textual notes-specific pipeline and annotation format, but what we have should work ok)

I'd mentioned that the "fragment" might be required for matching; I think it is best to have it be the same thing as the "value" of any of the tokens in ve_ref, but it is not required here. It is required for an variant of the CommentaryWidget used in Scaife Viewer 1, but not for Beyond Translation.