Search configuration for client-side full-text search
When enabled, a search index is generated during site rendering. The index is sharded for efficient incremental loading and supports Unicode text with diacritic folding.
Can be a simple boolean or a detailed configuration object, e.g.
# Enable search with defaults
[site]
search = true
# Customize search indexing
[site.search]
enabled = true
include-types = ["Heading", "Paragraph"]
exclude-routes = ["/api/**"]
Enable search index generation
When true, a search index is generated during site rendering. Default: false
Node types to include in the search index
Specifies which Stencila node types should be indexed. When not specified, defaults to common content types: ["Heading", "Paragraph", "Datatable", "CodeChunk", "Figure", "Table"]
Available types include: Article, Heading, Paragraph, Datatable, CodeChunk, Figure, Table, and other Stencila node types.
Route patterns to exclude from indexing
Glob patterns for routes that should not be indexed. Useful for excluding API documentation, internal pages, etc.
Example: ["/api/**", "/internal/**"]
Maximum text length per search entry
Text content longer than this will be truncated. Default: 500 characters
Enable fuzzy search support
When true, pre-computed trigrams are included in the search index to enable fuzzy matching (finding results with typos). This increases index size by approximately 1KB per entry. Default: true
This documentation was generated from site.rs by generate.rs.
