Skip to main content

Search

GospeLib provides two complementary search capabilities: full-text search powered by Typesense and smart reference parsing that recognizes scripture citations and navigates directly to passages.

Overview​

Full-Text Search​

Typesense indexes all scripture passages, topics, and lexicon entries for sub-50ms search responses. Users can search across the entire canon or filter by:

  • Book -- Limit to a specific book (e.g., only Book of Mormon)
  • Testament -- Old Testament, New Testament, or Restoration scripture
  • Corpus -- Biblical, Book of Mormon, Doctrine and Covenants, Pearl of Great Price, Pseudepigrapha

Results are ranked by relevance with highlighted matching text. Pagination supports browsing through large result sets.

Reference Lookup​

The reference parser recognizes standard scripture citation formats and navigates directly to the passage without requiring a full-text search:

InputParsed As
John 3:16Gospel of John, chapter 3, verse 16
1 Ne. 3:71 Nephi, chapter 3, verse 7
D&C 89:1-4Doctrine and Covenants, section 89, verses 1-4
Moses 1Book of Moses, chapter 1 (full chapter)
Gen 1:1-2:3Genesis, chapter 1 verse 1 through chapter 2 verse 3

Reference parsing happens client-side via the @gospelib/scripture-ref package, providing instant navigation without a network round-trip.

Search UI​

The search interface is accessible from:

  1. Search bar in the application header -- Standard search input with type-ahead suggestions
  2. Command palette (Cmd+K) -- Three-tier ranked results:
    • Tier 1 (<80ms): Fuzzy search over local passage index
    • Tier 2 (50-200ms): FalkorDB context-aware suggestions based on current reading position
    • Tier 3 (200-1000ms): AI-powered suggestions for complex queries

The command palette supports prefix modes for targeted search:

PrefixSearches
(none)Everything
>Commands and actions
@Symbols (people, places)
#Topics
?AI-powered questions
/Plugin commands

Rate Limits​

Search is available on all tiers with different rate limits:

PlanSearch Requests
Reader (free)20/min
Scholar200/min
Academic1000/min

Deep Dive: Typesense Integration​

The ingest pipeline includes a Typesense sync stage that indexes passages into three Typesense collections:

  • passages -- Scripture text with book, chapter, verse, corpus, and translation metadata
  • topics -- Topical Guide and Bible Dictionary entries
  • lexicon -- Strong's Hebrew and Greek dictionary entries

The content service exposes search through GET /search with query, filters, and pagination parameters. Typesense handles tokenization, stemming, and typo tolerance automatically.