You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Roundtrip canonical USFM para styles (existing editor functionality)
Roundtrip chapter and verse markup including verse ranges (existing editor functionality)
Roundtrip headings (existing editor functionality, important UI issues to explore)
mt (not editable)
Out of Scope (unless easy)
character styles
milestones that span paragraphs
cp, vp, cl...
footnotes and xrefs
word-level markup, attributes etc
introductions
extended markup such as sidebars
General Approach
Proskomma generates AGHAST for the base sequence (main/canonical or introduction) including headings. Other grafts are present as references, eg mt. The AGHAST record or its wrapper includes the heading ids.
Handling headings as discrete sequences within markup where blocks do not contain content is going to be weird, one way or another. Maybe
Heading starts with heading class milestone, similar to, say \p
Heading ends with next heading/para milestone or EOF
Start-heading milestone contains sequence id or null (for new heading)
ie headings behave like paragraph milestones in the editor, except that they may carry a sequence id.
The editor should read back the AGHAST produced after each mutation, to avoid any skew, and because...
The editor produces similar AGHAST for mutations, including the base sequence id. Unchanged or edited headings include the id. New headings have no id (Proskomma will create a new sequence and return the id in the next AGHAST read.) Sequences for headings that are no longer present will be GC'd by Proskomma.
Pk Tasks
Add mutation input cleaning (akin to what happens when USFM is read, eg move widow and orphan verse and chapter tags)
Generate cv-structured AGHAST for main sequence using Pk render model
Convert cv-structured AGHAST into Pk input format using utility function
No description provided.
The text was updated successfully, but these errors were encountered: