quran-search-engine

Search Configuration

The main search function accepts two optional configuration layers allowing fine-tuned control over behavior and pagination limits.

SearchOptions

Toggles linguistic matching logic. Enabling Lemma & Root requires you to pass valid morphologyMap and wordMap payloads.

export type AdvancedSearchOptions = {
  // Required mapping variables
  lemma: boolean;
  root: boolean;

  // Optional matching variables
  fuzzy?: boolean; // Default: true
  isRegex?: boolean;
  semantic?: boolean;

  // Optional geographic filtering parameters
  suraId?: number;
  juzId?: number;
  suraName?: string;
  sura_name_en?: string;
  sura_name_romanization?: string;
};

export type SearchOptions = AdvancedSearchOptions;

• Fuzzy Matching: By default fuzzy is allowed to fallback if exact/lemma/root checks fail. Pass { fuzzy: false } to strictly enforce absolute dict matches.
• Regex Search: Passing { isRegex: true } processes the query as a regular expression instead of standard token matching. The pattern is validated for syntactic correctness and checked against known ReDoS-prone patterns (nested quantifiers, overlapping alternation) before execution. Throws InvalidRegexError for invalid or unsafe patterns. Regex search bypasses all linguistic pipelines (lemma, root, fuzzy) and matches directly against the normalized standard text field. Can be combined with suraId, juzId, or suraName to narrow the search scope.
• Semantic Search: Enable { semantic: true } to integrate with AI embeddings conditionally if valid data map shapes are configured.
• Filtering Options: Narrow searches spatially via explicit { suraId: 2, juzId: 3 } etc.

PaginationOptions

Control bounds limit constraints natively. Omitting these defaults to logical page limits (typically page 1, limit 50).

export type PaginationOptions = {
  page?: number; // Starting return page.
  limit?: number; // Count return offset per page limit.
};

This site is open source. Improve this page.