@chronicle.app/marvin

Chronicle plugin for reading sessions and annotations from Marvin, the e-reader app. It reads the CSV files Marvin exports. Pass the file with --input and pick the record type with --type:

chronicle extract marvin --type reading-sessions --input sessions.csv
chronicle extract marvin --type annotations --input annotations.csv

Each type reads its own export file. The reading-sessions export has one row per part of a session (Session ID, Book Title, Book Author, Date Created, Part Time in seconds). Rows are grouped by session: part times are summed and the latest Date Created ends the session. Each session becomes a ReadAction whose startTime is the end time minus the total part time, with the Book as its object. Sessions come out oldest first. The limit also caps the CSV rows read before grouping, so pass --limit 0 to read a whole export.

The annotations export has one row per highlight or note (ID, Title, Author, Date or Created, HighlightText, EntryText). A highlight becomes a QuoteAction whose result is a Quotation that is part of the Book. A note becomes an AnnotateAction whose result is a Comment about the highlight, or about the book when the row has no highlight. Rows with neither are skipped.

Books have no IDs in the export, so a Book is keyed by its title and its author's name. Marvin escapes commas in author names as \,; they are unescaped. The export does not identify the reader, so the agent is the marvin self person, marked as you with sameAs: ['@me'].

Tests use small synthetic CSV files with made-up books and quotes, and never read the host's data or accounts.