1 Commits
0.3.2 ... 0.3.3

Author SHA1 Message Date
7a2f17bfe9 Use kiwix search completely, no fuzzy ranking
All checks were successful
Publish Library / Build NPM Project (push) Successful in 16s
Publish Library / Tag Version (push) Successful in 10s
2026-08-25 00:42:59 -04:00
3 changed files with 9 additions and 26 deletions

View File

@@ -1,6 +1,6 @@
{
"name": "@ztimson/zim-utils",
"version": "0.3.2",
"version": "0.3.3",
"description": "Native, dependency-light ZIM archive reader/searcher and Kiwix catalog downloader for Node.js",
"author": "Zak Timson",
"license": "MIT",

View File

@@ -6,8 +6,7 @@ import net from 'node:net';
import fs from 'node:fs';
import path from 'node:path';
import {fileURLToPath} from 'node:url';
import {decodeHtml, fromXml} from '@ztimson/utils';
import {fuzzyMatch, weightedScore} from './utils.js';
import {fromXml} from '@ztimson/utils';
const execFileAsync = promisify(execFile);
const __dirname = path.dirname(fileURLToPath(import.meta.url));
@@ -180,28 +179,21 @@ export class KiwixServer {
return {mimetype: res.headers.get('content-type'), data: Buffer.from(await res.arrayBuffer())};
}
/**
* Two-pass fulltext search across every local ZIM: xapian prefilter (kiwix-serve's
* own index), then fuzzy-reranked by title so the strongest matches surface first.
* Returns a flat array matching the catalog/list shape: {id, title, name, category, ..., href, score}
*/
/** Fulltext search across every local ZIM via kiwix-serve's own xapian index */
async search(terms, limit = 20) {
this.#assertRunning();
const termList = String(terms).split(',').map(t => t.trim()).filter(Boolean);
const termList = String(terms).split(/[,\s]+/).map(t => t.trim()).filter(Boolean);
if (!termList.length) return [];
const params = new URLSearchParams({pattern: termList.join(' '), format: 'xml', pageLength: String(limit)});
const res = await fetch(`${this.baseUrl}/search?${params}`);
if (!res.ok) return [];
const xml = await res.text();
let found = fromXml(xml);
found = found?.rss?.channel?.item || [];
const found = fromXml(await res.text())?.rss?.channel?.item || [];
const books = await this.list();
const bookMap = new Map(books.map(b => [b.title, b]));
const enriched = found.map(hit => {
return found.map(hit => {
const book = bookMap.get(hit.book.title);
if (!book) return null;
const prefix = `/content/${book.href}/`;
@@ -216,10 +208,8 @@ export class KiwixServer {
icon: book.icon,
viewer: this.baseUrl + hit.link,
summary: hit.description,
score: weightedScore(hit.title, termList) + weightedScore(hit.description, termList),
score: +hit.score || 0,
};
}).filter(hit => !!hit && hit.score > 0);
return enriched.toSorted((a, b) => b.score - a.score).slice(0, limit);
}).filter(Boolean);
}
}

View File

@@ -40,10 +40,3 @@ export function fuzzyMatch(target, ...terms) {
similarities,
};
}
export function weightedScore(text, termList) {
if (!text) return 0;
const covered = termList.reduce((sum, t) => sum + t.length, 0);
const coverage = Math.min(1, covered / text.length);
return fuzzyMatch(text, ...termList).max * coverage;
}