Textbites is a Python library for representing textual resources. It provides a interface for browsing and searching along with several implementations and unit tests.
Textbites is used in Nutritious, a Django-based web application for browsing and tagging textual content.
Textbites' API includes two primary classes: Resource and Reference.
Resources are containers of textual content which provide a top-most reference to that resource as well a "reference locator" service which parses a string reference and returns the associated reference object. A resource could contain additional meta-data such as the type of resource.
References represent an iterator-like pointer into the resource. They provide methods to:
- Retrieve the text at that location.
- Retrieve a human-readable version of the reference string.
- Navigate by traversing to the parent or child references.
- Search over the textual content at that location returning matching references.
The Library class can dynamically load resources from a directory by detecting type by extension (e.g. JSON or TSV). Several resources are included in the project and are located in the "data" directory.
Several implementations of the API are included. See comments in source code for more details.
- SimpleBooks
- Book, Chapter, ChapterRange, Line, LineRange
- Bible (collection of SimpleBooks)
- Quotes
- ViewBooks (alternate design)
Over 100 unit tests can be run with bin/run_tests.sh.
Currently, all implementations are static, in-memory resources. This is a big limitation for many kinds of resources (e.g. web bookmarks), and will most likely be addressed by subclassing it within a framework which provides ORM like Django.