@se-studio/search 1.0.36 → 1.0.38

package/CHANGELOG.md

@@ -1,5 +1,27 @@
 # @se-studio/search
 
+## 1.0.38
+
+### Patch Changes
+
+- Version bump: patch for changed packages
+- Updated dependencies
+  - @se-studio/contentful-rest-api@1.0.132
+  - @se-studio/core-data-types@1.0.127
+  - @se-studio/markdown-renderer@1.0.91
+  - @se-studio/wordpress-rest-api@1.0.8
+
+## 1.0.37
+
+### Patch Changes
+
+- Version bump: patch for changed packages
+- Updated dependencies
+  - @se-studio/contentful-rest-api@1.0.131
+  - @se-studio/core-data-types@1.0.126
+  - @se-studio/markdown-renderer@1.0.90
+  - @se-studio/wordpress-rest-api@1.0.7
+
 ## 1.0.36
 
 ### Patch Changes

package/README.md

@@ -198,6 +198,30 @@ documentTransformer?: (doc: SearchDocument, contentData: ContentData) => SearchD
 
 The transformer runs on every chunk of a multi-chunk entry, so a 3-chunk page will call it 3 times.
 
+## Canonical URLs and custom indexes (`IIndexableContent`)
+
+For **Upstash Search**, each `SearchDocument` already carries **`metadata.href`** (full path), populated from the CMS model’s `href` in `buildSearchDocuments` — do not infer URLs from `slug` + content type at query time.
+
+If you maintain a **separate** full-text index (e.g. Lunr, blob JSON), use the shared **`IIndexableContent`** type from `@se-studio/search`. It requires:
+
+- **`href`** – full canonical path as used in `<a href>` (e.g. `/insights/my-article`)
+- **`content`**, **`title`**, **`id`**, **`contentType`** (string label), **`metadata`**
+
+Optional **`slug`** remains useful for tokenisation; do not rely on it alone to build user-facing links.
+
+Helpers:
+
+- **`stripMarkdownToPlainText`**, **`calculateReadingTime`** – `@se-studio/search/indexing`
+- **`indexArticleLinksToIndexableContent`** – builds `IIndexableContent[]` from `IArticleLink[]` using **one `MarkdownExporter.fetchContent()` call per article** (see below). Pass **`markdownContext`** (`config`, `siteConfig`, `urlCalculators`, optional `customConverters`); `contentContext` is supplied per entry from each `fetchContent` result.
+
+## Contentful indexing hazards (includes truncation)
+
+The Contentful REST API can **silently truncate** `includes.Entry` and `includes.Asset` when a single response is large — for example, batching many entries in **one** `getEntries` call with **`include`** and walking `includes` to resolve rich text or linked entries often yields **incomplete** trees. Symptoms include missing body text or featured images in the index, with **no error** from the API.
+
+**Recommended default:** fetch **each** entry individually via **`MarkdownExporter.fetchContent()`** (same pattern as `rebuildSearchIndex` in this package: one call per link/slug). That uses the normal per-entry REST/converter path instead of a multi-entry batched query.
+
+If you must batch requests, you need a strategy that does not depend on unbounded resolved includes in one response (e.g. lower depth, smaller batches, or follow-up fetches) — the package cannot fix API limits from TypeScript alone.
+
 ## Architecture
 
 - **Single Upstash database, two indexes**: `published` and `preview`
@@ -213,7 +237,7 @@ The transformer runs on every chunk of a multi-chunk entry, so a 3-chunk page wi
 |--------|---------|
 | `@se-studio/search` | Types only |
 | `@se-studio/search/client` | `createSearchClient()` |
-| `@se-studio/search/indexing` | `rebuildSearchIndex()`, `buildSearchDocument()` |
+| `@se-studio/search/indexing` | `rebuildSearchIndex()`, `buildSearchDocuments`, `indexArticleLinksToIndexableContent`, `stripMarkdownToPlainText`, `calculateReadingTime`, … |
 | `@se-studio/search/webhook` | `createSearchWebhookHandler()` |
 | `@se-studio/search/api` | `createSearchApiHandler()`, `createRebuildApiHandler()` |
 | `@se-studio/search/hooks` | `useSearch()` |

package/dist/index.d.ts

@@ -1,3 +1,4 @@
 export type { ContentData } from '@se-studio/markdown-renderer';
+export type { IIndexableContent } from './indexable-content';
 export type { ContentTypeIndexConfig, RebuildResult, SearchableContentType, SearchConfig, SearchDocument, SearchDocumentMetadata, SearchIndexConfig, SearchIndexingConfig, SearchOptions, SearchResponse, SearchResult, } from './types';
 //# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,YAAY,EAAE,WAAW,EAAE,MAAM,8BAA8B,CAAC;AAChE,YAAY,EACV,sBAAsB,EACtB,aAAa,EACb,qBAAqB,EACrB,YAAY,EACZ,cAAc,EACd,sBAAsB,EACtB,iBAAiB,EACjB,oBAAoB,EACpB,aAAa,EACb,cAAc,EACd,YAAY,GACb,MAAM,SAAS,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,YAAY,EAAE,WAAW,EAAE,MAAM,8BAA8B,CAAC;AAChE,YAAY,EAAE,iBAAiB,EAAE,MAAM,qBAAqB,CAAC;AAC7D,YAAY,EACV,sBAAsB,EACtB,aAAa,EACb,qBAAqB,EACrB,YAAY,EACZ,cAAc,EACd,sBAAsB,EACtB,iBAAiB,EACjB,oBAAoB,EACpB,aAAa,EACb,cAAc,EACd,YAAY,GACb,MAAM,SAAS,CAAC"}

package/dist/indexable-content.d.ts

@@ -0,0 +1,24 @@
+/**
+ * Flat document shape for building non-Upstash indexes (e.g. Lunr, blob-stored indexes).
+ * Use {@link SearchDocument} / {@link SearchDocumentMetadata} for Upstash Search instead.
+ *
+ * **href** must be the full path as used in `<a href>` (e.g. `/insights/my-article`), not a
+ * bare slug — URL topology must not be inferred from `slug` + `contentType` at query time.
+ */
+export interface IIndexableContent {
+    id: string;
+    /** Full canonical path for links, e.g. `/insights/my-article` */
+    href: string;
+    /** Logical type label (app-defined); e.g. `article`, `page`, `person` */
+    contentType: string;
+    title: string;
+    /** Plain-text body used for full-text indexing */
+    content: string;
+    /** Optional bare slug for tokenisation or legacy callers */
+    slug?: string;
+    date?: string;
+    tags?: string[];
+    author?: string;
+    metadata: Record<string, unknown>;
+}
+//# sourceMappingURL=indexable-content.d.ts.map

package/dist/indexable-content.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"indexable-content.d.ts","sourceRoot":"","sources":["../src/indexable-content.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AACH,MAAM,WAAW,iBAAiB;IAChC,EAAE,EAAE,MAAM,CAAC;IACX,iEAAiE;IACjE,IAAI,EAAE,MAAM,CAAC;IACb,yEAAyE;IACzE,WAAW,EAAE,MAAM,CAAC;IACpB,KAAK,EAAE,MAAM,CAAC;IACd,kDAAkD;IAClD,OAAO,EAAE,MAAM,CAAC;IAChB,4DAA4D;IAC5D,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,IAAI,CAAC,EAAE,MAAM,EAAE,CAAC;IAChB,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,QAAQ,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;CACnC"}

package/dist/indexable-content.js

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=indexable-content.js.map

package/dist/indexable-content.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"indexable-content.js","sourceRoot":"","sources":["../src/indexable-content.ts"],"names":[],"mappings":""}

package/dist/indexing/article-indexable.d.ts

@@ -0,0 +1,48 @@
+import type { IArticleLink } from '@se-studio/core-data-types';
+import { type ContentData, type MarkdownConverterContext, type MarkdownExporter } from '@se-studio/markdown-renderer';
+import type { IIndexableContent } from '../indexable-content';
+/** Same fields as {@link MarkdownConverterContext} except `contentContext` (supplied per fetch). */
+export type ArticleMarkdownConverterParams = Omit<MarkdownConverterContext, 'contentContext'>;
+export interface ArticleIndexableHooks {
+    /**
+     * Return false to skip a link. Defaults to {@link defaultShouldIndexArticleLink}
+     * when omitted.
+     */
+    shouldIndexArticleLink?(link: IArticleLink): boolean;
+    /**
+     * Override plain body after markdown conversion (e.g. custom truncation).
+     */
+    mapPlainTextBody?(link: IArticleLink, plainText: string, contentData: ContentData | null): string;
+    /**
+     * Build metadata beyond defaults (`originalType`, `readingTime`, `link`).
+     */
+    buildMetadata?(link: IArticleLink, plainTextBody: string, contentData: ContentData | null): Record<string, unknown>;
+}
+export interface IndexArticleLinksOptions {
+    exporter: MarkdownExporter;
+    /**
+     * Passed into each `MarkdownConverter.convert()` together with `contentData.context`
+     * from `fetchContent` (the entry-specific context is required for correct markdown).
+     */
+    markdownContext: ArticleMarkdownConverterParams;
+    /**
+     * Passed to `fetchContent('article', slug, { articleType })` when `link.articleType?.slug`
+     * is missing — mirrors {@link rebuildSearchIndex} (`blog` fallback in rebuild).
+     */
+    defaultArticleTypeSlug?: string;
+    /** Label stored on each `IIndexableContent.contentType` (default: `article`). */
+    indexableContentType?: string;
+    hooks?: ArticleIndexableHooks;
+    onProgress?: (current: number, total: number, itemName: string) => Promise<void>;
+}
+/**
+ * Default article link filter: internal paths only, no download articles.
+ */
+export declare function defaultShouldIndexArticleLink(link: IArticleLink): boolean;
+/**
+ * Builds {@link IIndexableContent} for each article link using **one**
+ * `MarkdownExporter.fetchContent()` call per article (avoids Contentful `includes` truncation
+ * from batched entry queries).
+ */
+export declare function indexArticleLinksToIndexableContent(links: readonly IArticleLink[], options: IndexArticleLinksOptions): Promise<IIndexableContent[]>;
+//# sourceMappingURL=article-indexable.d.ts.map

package/dist/indexing/article-indexable.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"article-indexable.d.ts","sourceRoot":"","sources":["../../src/indexing/article-indexable.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAY,MAAM,4BAA4B,CAAC;AACzE,OAAO,EACL,KAAK,WAAW,EAGhB,KAAK,wBAAwB,EAC7B,KAAK,gBAAgB,EACtB,MAAM,8BAA8B,CAAC;AACtC,OAAO,KAAK,EAAE,iBAAiB,EAAE,MAAM,sBAAsB,CAAC;AAG9D,oGAAoG;AACpG,MAAM,MAAM,8BAA8B,GAAG,IAAI,CAAC,wBAAwB,EAAE,gBAAgB,CAAC,CAAC;AAE9F,MAAM,WAAW,qBAAqB;IACpC;;;OAGG;IACH,sBAAsB,CAAC,CAAC,IAAI,EAAE,YAAY,GAAG,OAAO,CAAC;IACrD;;OAEG;IACH,gBAAgB,CAAC,CAAC,IAAI,EAAE,YAAY,EAAE,SAAS,EAAE,MAAM,EAAE,WAAW,EAAE,WAAW,GAAG,IAAI,GAAG,MAAM,CAAC;IAClG;;OAEG;IACH,aAAa,CAAC,CACZ,IAAI,EAAE,YAAY,EAClB,aAAa,EAAE,MAAM,EACrB,WAAW,EAAE,WAAW,GAAG,IAAI,GAC9B,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;CAC5B;AAED,MAAM,WAAW,wBAAwB;IACvC,QAAQ,EAAE,gBAAgB,CAAC;IAC3B;;;OAGG;IACH,eAAe,EAAE,8BAA8B,CAAC;IAChD;;;OAGG;IACH,sBAAsB,CAAC,EAAE,MAAM,CAAC;IAChC,iFAAiF;IACjF,oBAAoB,CAAC,EAAE,MAAM,CAAC;IAC9B,KAAK,CAAC,EAAE,qBAAqB,CAAC;IAC9B,UAAU,CAAC,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,KAAK,OAAO,CAAC,IAAI,CAAC,CAAC;CAClF;AAkBD;;GAEG;AACH,wBAAgB,6BAA6B,CAAC,IAAI,EAAE,YAAY,GAAG,OAAO,CAQzE;AAMD;;;;GAIG;AACH,wBAAsB,mCAAmC,CACvD,KAAK,EAAE,SAAS,YAAY,EAAE,EAC9B,OAAO,EAAE,wBAAwB,GAChC,OAAO,CAAC,iBAAiB,EAAE,CAAC,CA0E9B"}

package/dist/indexing/article-indexable.js

@@ -0,0 +1,95 @@
+import { cleanMarkdownText, MarkdownConverter, } from '@se-studio/markdown-renderer';
+import { calculateReadingTime, stripMarkdownToPlainText } from './readability';
+function tagTitle(t) {
+    return t.title ?? t.name ?? t.slug ?? '';
+}
+function defaultTags(link) {
+    const raw = link.tags ?? [];
+    const names = raw.map(tagTitle).filter(Boolean);
+    return names.length > 0 ? names : ['article'];
+}
+function defaultAuthor(link) {
+    const a = link.author;
+    if (!a)
+        return undefined;
+    return a.name ?? a.title ?? undefined;
+}
+/**
+ * Default article link filter: internal paths only, no download articles.
+ */
+export function defaultShouldIndexArticleLink(link) {
+    if (!link.href || link.href.startsWith('http')) {
+        return false;
+    }
+    if (link.download) {
+        return false;
+    }
+    return true;
+}
+function resolveArticleTypeSlug(link, fallback) {
+    return link.articleType?.slug ?? fallback;
+}
+/**
+ * Builds {@link IIndexableContent} for each article link using **one**
+ * `MarkdownExporter.fetchContent()` call per article (avoids Contentful `includes` truncation
+ * from batched entry queries).
+ */
+export async function indexArticleLinksToIndexableContent(links, options) {
+    const { exporter, markdownContext, defaultArticleTypeSlug = 'blog', indexableContentType = 'article', hooks, onProgress, } = options;
+    const shouldIndex = hooks?.shouldIndexArticleLink ?? defaultShouldIndexArticleLink;
+    const converter = new MarkdownConverter();
+    const results = [];
+    const indexable = links.filter((l) => shouldIndex(l));
+    const total = indexable.length;
+    let current = 0;
+    for (const link of indexable) {
+        current++;
+        const articleType = resolveArticleTypeSlug(link, defaultArticleTypeSlug);
+        let contentData = null;
+        try {
+            contentData = await exporter.fetchContent('article', link.slug, { articleType });
+        }
+        catch {
+            contentData = null;
+        }
+        let plainBody = '';
+        if (contentData) {
+            const markdown = converter.convert(contentData, {
+                ...markdownContext,
+                contentContext: contentData.context,
+            });
+            plainBody = stripMarkdownToPlainText(cleanMarkdownText(markdown));
+        }
+        if (hooks?.mapPlainTextBody) {
+            plainBody = hooks.mapPlainTextBody(link, plainBody, contentData);
+        }
+        const readingMinutes = plainBody.trim() ? calculateReadingTime(plainBody) : undefined;
+        const tagNames = defaultTags(link);
+        const authorName = defaultAuthor(link);
+        const baseMeta = {
+            originalType: 'article',
+            link: link.href,
+            ...(readingMinutes !== undefined ? { readingTime: `${readingMinutes} min read` } : {}),
+        };
+        const metadata = hooks?.buildMetadata
+            ? { ...baseMeta, ...hooks.buildMetadata(link, plainBody, contentData) }
+            : baseMeta;
+        results.push({
+            id: link.id,
+            href: link.href,
+            contentType: indexableContentType,
+            title: link.title ?? link.text ?? '',
+            content: plainBody,
+            slug: link.slug,
+            date: link.date ?? undefined,
+            tags: tagNames,
+            author: authorName,
+            metadata,
+        });
+        if (onProgress) {
+            await onProgress(current, total, link.title ?? link.text ?? '');
+        }
+    }
+    return results;
+}
+//# sourceMappingURL=article-indexable.js.map

package/dist/indexing/article-indexable.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"article-indexable.js","sourceRoot":"","sources":["../../src/indexing/article-indexable.ts"],"names":[],"mappings":"AACA,OAAO,EAEL,iBAAiB,EACjB,iBAAiB,GAGlB,MAAM,8BAA8B,CAAC;AAEtC,OAAO,EAAE,oBAAoB,EAAE,wBAAwB,EAAE,MAAM,eAAe,CAAC;AA2C/E,SAAS,QAAQ,CAAC,CAAW;IAC3B,OAAO,CAAC,CAAC,KAAK,IAAI,CAAC,CAAC,IAAI,IAAI,CAAC,CAAC,IAAI,IAAI,EAAE,CAAC;AAC3C,CAAC;AAED,SAAS,WAAW,CAAC,IAAkB;IACrC,MAAM,GAAG,GAAG,IAAI,CAAC,IAAI,IAAI,EAAE,CAAC;IAC5B,MAAM,KAAK,GAAG,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;IAChD,OAAO,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC;AAChD,CAAC;AAED,SAAS,aAAa,CAAC,IAAkB;IACvC,MAAM,CAAC,GAAG,IAAI,CAAC,MAAM,CAAC;IACtB,IAAI,CAAC,CAAC;QAAE,OAAO,SAAS,CAAC;IACzB,OAAO,CAAC,CAAC,IAAI,IAAI,CAAC,CAAC,KAAK,IAAI,SAAS,CAAC;AACxC,CAAC;AAED;;GAEG;AACH,MAAM,UAAU,6BAA6B,CAAC,IAAkB;IAC9D,IAAI,CAAC,IAAI,CAAC,IAAI,IAAI,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,MAAM,CAAC,EAAE,CAAC;QAC/C,OAAO,KAAK,CAAC;IACf,CAAC;IACD,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAC;QAClB,OAAO,KAAK,CAAC;IACf,CAAC;IACD,OAAO,IAAI,CAAC;AACd,CAAC;AAED,SAAS,sBAAsB,CAAC,IAAkB,EAAE,QAAgB;IAClE,OAAO,IAAI,CAAC,WAAW,EAAE,IAAI,IAAI,QAAQ,CAAC;AAC5C,CAAC;AAED;;;;GAIG;AACH,MAAM,CAAC,KAAK,UAAU,mCAAmC,CACvD,KAA8B,EAC9B,OAAiC;IAEjC,MAAM,EACJ,QAAQ,EACR,eAAe,EACf,sBAAsB,GAAG,MAAM,EAC/B,oBAAoB,GAAG,SAAS,EAChC,KAAK,EACL,UAAU,GACX,GAAG,OAAO,CAAC;IAEZ,MAAM,WAAW,GAAG,KAAK,EAAE,sBAAsB,IAAI,6BAA6B,CAAC;IACnF,MAAM,SAAS,GAAG,IAAI,iBAAiB,EAAE,CAAC;IAC1C,MAAM,OAAO,GAAwB,EAAE,CAAC;IACxC,MAAM,SAAS,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,WAAW,CAAC,CAAC,CAAC,CAAC,CAAC;IACtD,MAAM,KAAK,GAAG,SAAS,CAAC,MAAM,CAAC;IAE/B,IAAI,OAAO,GAAG,CAAC,CAAC;IAChB,KAAK,MAAM,IAAI,IAAI,SAAS,EAAE,CAAC;QAC7B,OAAO,EAAE,CAAC;QACV,MAAM,WAAW,GAAG,sBAAsB,CAAC,IAAI,EAAE,sBAAsB,CAAC,CAAC;QACzE,IAAI,WAAW,GAAuB,IAAI,CAAC;QAC3C,IAAI,CAAC;YACH,WAAW,GAAG,MAAM,QAAQ,CAAC,YAAY,CAAC,SAAS,EAAE,IAAI,CAAC,IAAI,EAAE,EAAE,WAAW,EAAE,CAAC,CAAC;QACnF,CAAC;QAAC,MAAM,CAAC;YACP,WAAW,GAAG,IAAI,CAAC;QACrB,CAAC;QAED,IAAI,SAAS,GAAG,EAAE,CAAC;QACnB,IAAI,WAAW,EAAE,CAAC;YAChB,MAAM,QAAQ,GAAG,SAAS,CAAC,OAAO,CAAC,WAAW,EAAE;gBAC9C,GAAG,eAAe;gBAClB,cAAc,EAAE,WAAW,CAAC,OAAO;aACpC,CAAC,CAAC;YACH,SAAS,GAAG,wBAAwB,CAAC,iBAAiB,CAAC,QAAQ,CAAC,CAAC,CAAC;QACpE,CAAC;QAED,IAAI,KAAK,EAAE,gBAAgB,EAAE,CAAC;YAC5B,SAAS,GAAG,KAAK,CAAC,gBAAgB,CAAC,IAAI,EAAE,SAAS,EAAE,WAAW,CAAC,CAAC;QACnE,CAAC;QAED,MAAM,cAAc,GAAG,SAAS,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,oBAAoB,CAAC,SAAS,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;QAEtF,MAAM,QAAQ,GAAG,WAAW,CAAC,IAAI,CAAC,CAAC;QACnC,MAAM,UAAU,GAAG,aAAa,CAAC,IAAI,CAAC,CAAC;QAEvC,MAAM,QAAQ,GAA4B;YACxC,YAAY,EAAE,SAAS;YACvB,IAAI,EAAE,IAAI,CAAC,IAAI;YACf,GAAG,CAAC,cAAc,KAAK,SAAS,CAAC,CAAC,CAAC,EAAE,WAAW,EAAE,GAAG,cAAc,WAAW,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;SACvF,CAAC;QAEF,MAAM,QAAQ,GAAG,KAAK,EAAE,aAAa;YACnC,CAAC,CAAC,EAAE,GAAG,QAAQ,EAAE,GAAG,KAAK,CAAC,aAAa,CAAC,IAAI,EAAE,SAAS,EAAE,WAAW,CAAC,EAAE;YACvE,CAAC,CAAC,QAAQ,CAAC;QAEb,OAAO,CAAC,IAAI,CAAC;YACX,EAAE,EAAE,IAAI,CAAC,EAAE;YACX,IAAI,EAAE,IAAI,CAAC,IAAI;YACf,WAAW,EAAE,oBAAoB;YACjC,KAAK,EAAE,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,IAAI,IAAI,EAAE;YACpC,OAAO,EAAE,SAAS;YAClB,IAAI,EAAE,IAAI,CAAC,IAAI;YACf,IAAI,EAAE,IAAI,CAAC,IAAI,IAAI,SAAS;YAC5B,IAAI,EAAE,QAAQ;YACd,MAAM,EAAE,UAAU;YAClB,QAAQ;SACT,CAAC,CAAC;QAEH,IAAI,UAAU,EAAE,CAAC;YACf,MAAM,UAAU,CAAC,OAAO,EAAE,KAAK,EAAE,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,IAAI,IAAI,EAAE,CAAC,CAAC;QAClE,CAAC;IACH,CAAC;IAED,OAAO,OAAO,CAAC;AACjB,CAAC"}

package/dist/indexing/content-extractor.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"content-extractor.d.ts","sourceRoot":"","sources":["../../src/indexing/content-extractor.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,WAAW,EAAE,wBAAwB,EAAE,MAAM,8BAA8B,CAAC;AAY1F;;;GAGG;AACH,wBAAgB,sBAAsB,CAAC,WAAW,EAAE,MAAM,EAAE,iBAAiB,EAAE,MAAM,GAAG,MAAM,CAE7F;AAuCD;;;;;;GAMG;AACH,wBAAgB,uBAAuB,CACrC,WAAW,EAAE,WAAW,EACxB,gBAAgB,EAAE,wBAAwB,EAC1C,gBAAgB,EAAE,MAAM,GACvB,MAAM,EAAE,CA8BV;AAED;;;GAGG;AACH,wBAAgB,kBAAkB,CAAC,KAAK,EAAE,MAAM,EAAE,WAAW,EAAE,MAAM,GAAG,MAAM,CAE7E"}
+{"version":3,"file":"content-extractor.d.ts","sourceRoot":"","sources":["../../src/indexing/content-extractor.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,WAAW,EAAE,wBAAwB,EAAE,MAAM,8BAA8B,CAAC;AAa1F;;;GAGG;AACH,wBAAgB,sBAAsB,CAAC,WAAW,EAAE,MAAM,EAAE,iBAAiB,EAAE,MAAM,GAAG,MAAM,CAE7F;AAED;;;;;;GAMG;AACH,wBAAgB,uBAAuB,CACrC,WAAW,EAAE,WAAW,EACxB,gBAAgB,EAAE,wBAAwB,EAC1C,gBAAgB,EAAE,MAAM,GACvB,MAAM,EAAE,CA8BV;AAED;;;GAGG;AACH,wBAAgB,kBAAkB,CAAC,KAAK,EAAE,MAAM,EAAE,WAAW,EAAE,MAAM,GAAG,MAAM,CAE7E"}

package/dist/indexing/content-extractor.js

@@ -1,5 +1,6 @@
 import { cleanMarkdownText, MarkdownConverter } from '@se-studio/markdown-renderer';
 import { debugLog } from '../debug';
+import { stripMarkdownToPlainText } from './readability';
 /** Upstash Search hard limit for total content per document. */
 const UPSTASH_CONTENT_LIMIT = 4096;
 /** Buffer for JSON key names and structural overhead in the content object. */
@@ -12,33 +13,6 @@ const CHUNK_OVERLAP = 300;
 export function calculateMaxBodyLength(titleLength, descriptionLength) {
     return Math.max(0, UPSTASH_CONTENT_LIMIT - titleLength - descriptionLength - JSON_OVERHEAD);
 }
-/**
- * Strips markdown syntax to produce plain text suitable for search indexing.
- * Removes headings markers, bold/italic markers, links, images, and frontmatter.
- */
-function stripMarkdown(md) {
-    let text = md;
-    // Remove YAML frontmatter
-    text = text.replace(/^---[\s\S]*?---\n*/m, '');
-    // Remove images: ![alt](url)
-    text = text.replace(/!\[.*?\]\(.*?\)/g, '');
-    // Convert links to text: [text](url) → text
-    text = text.replace(/\[([^\]]*)\]\([^)]*\)/g, '$1');
-    // Remove heading markers: ### Heading → Heading
-    text = text.replace(/^#{1,6}\s+/gm, '');
-    // Remove bold/italic markers
-    text = text.replace(/\*{1,3}([^*]+)\*{1,3}/g, '$1');
-    text = text.replace(/_{1,3}([^_]+)_{1,3}/g, '$1');
-    // Remove horizontal rules
-    text = text.replace(/^[-*_]{3,}\s*$/gm, '');
-    // Remove list markers
-    text = text.replace(/^[\s]*[-*+]\s+/gm, '');
-    text = text.replace(/^[\s]*\d+\.\s+/gm, '');
-    // Collapse whitespace
-    text = text.replace(/\n{3,}/g, '\n\n');
-    text = text.trim();
-    return text;
-}
 /**
  * Converts CMS content to plain text, then splits it into chunks that each
  * fit within `maxCharsPerChunk`, with `CHUNK_OVERLAP` characters of overlap
@@ -50,7 +24,7 @@ export function extractSearchableChunks(contentData, converterContext, maxCharsP
     const converter = new MarkdownConverter();
     const markdown = converter.convert(contentData, converterContext);
     debugLog('extractor', `Markdown length: ${markdown.length} chars`);
-    const plainText = stripMarkdown(cleanMarkdownText(markdown));
+    const plainText = stripMarkdownToPlainText(cleanMarkdownText(markdown));
     debugLog('extractor', `Plain text length: ${plainText.length} chars (chunk limit ${maxCharsPerChunk})`);
     if (plainText.length <= maxCharsPerChunk) {
         return [plainText];

package/dist/indexing/content-extractor.js.map

@@ -1 +1 @@
-{"version":3,"file":"content-extractor.js","sourceRoot":"","sources":["../../src/indexing/content-extractor.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,iBAAiB,EAAE,iBAAiB,EAAE,MAAM,8BAA8B,CAAC;AACpF,OAAO,EAAE,QAAQ,EAAE,MAAM,UAAU,CAAC;AAEpC,gEAAgE;AAChE,MAAM,qBAAqB,GAAG,IAAI,CAAC;AAEnC,+EAA+E;AAC/E,MAAM,aAAa,GAAG,GAAG,CAAC;AAE1B,MAAM,aAAa,GAAG,GAAG,CAAC;AAE1B;;;GAGG;AACH,MAAM,UAAU,sBAAsB,CAAC,WAAmB,EAAE,iBAAyB;IACnF,OAAO,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,qBAAqB,GAAG,WAAW,GAAG,iBAAiB,GAAG,aAAa,CAAC,CAAC;AAC9F,CAAC;AAED;;;GAGG;AACH,SAAS,aAAa,CAAC,EAAU;IAC/B,IAAI,IAAI,GAAG,EAAE,CAAC;IAEd,0BAA0B;IAC1B,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,qBAAqB,EAAE,EAAE,CAAC,CAAC;IAE/C,6BAA6B;IAC7B,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,kBAAkB,EAAE,EAAE,CAAC,CAAC;IAE5C,4CAA4C;IAC5C,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,wBAAwB,EAAE,IAAI,CAAC,CAAC;IAEpD,gDAAgD;IAChD,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,cAAc,EAAE,EAAE,CAAC,CAAC;IAExC,6BAA6B;IAC7B,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,wBAAwB,EAAE,IAAI,CAAC,CAAC;IACpD,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,sBAAsB,EAAE,IAAI,CAAC,CAAC;IAElD,0BAA0B;IAC1B,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,kBAAkB,EAAE,EAAE,CAAC,CAAC;IAE5C,sBAAsB;IACtB,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,kBAAkB,EAAE,EAAE,CAAC,CAAC;IAC5C,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,kBAAkB,EAAE,EAAE,CAAC,CAAC;IAE5C,sBAAsB;IACtB,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,SAAS,EAAE,MAAM,CAAC,CAAC;IACvC,IAAI,GAAG,IAAI,CAAC,IAAI,EAAE,CAAC;IAEnB,OAAO,IAAI,CAAC;AACd,CAAC;AAED;;;;;;GAMG;AACH,MAAM,UAAU,uBAAuB,CACrC,WAAwB,EACxB,gBAA0C,EAC1C,gBAAwB;IAExB,MAAM,SAAS,GAAG,IAAI,iBAAiB,EAAE,CAAC;IAC1C,MAAM,QAAQ,GAAG,SAAS,CAAC,OAAO,CAAC,WAAW,EAAE,gBAAgB,CAAC,CAAC;IAClE,QAAQ,CAAC,WAAW,EAAE,oBAAoB,QAAQ,CAAC,MAAM,QAAQ,CAAC,CAAC;IACnE,MAAM,SAAS,GAAG,aAAa,CAAC,iBAAiB,CAAC,QAAQ,CAAC,CAAC,CAAC;IAC7D,QAAQ,CACN,WAAW,EACX,sBAAsB,SAAS,CAAC,MAAM,uBAAuB,gBAAgB,GAAG,CACjF,CAAC;IAEF,IAAI,SAAS,CAAC,MAAM,IAAI,gBAAgB,EAAE,CAAC;QACzC,OAAO,CAAC,SAAS,CAAC,CAAC;IACrB,CAAC;IACD,IAAI,gBAAgB,IAAI,aAAa,IAAI,gBAAgB,IAAI,CAAC,EAAE,CAAC;QAC/D,OAAO,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,gBAAgB,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;IAC7D,CAAC;IAED,MAAM,MAAM,GAAa,EAAE,CAAC;IAC5B,IAAI,MAAM,GAAG,CAAC,CAAC;IACf,MAAM,IAAI,GAAG,gBAAgB,GAAG,aAAa,CAAC;IAE9C,OAAO,MAAM,GAAG,SAAS,CAAC,MAAM,EAAE,CAAC;QACjC,MAAM,KAAK,GAAG,SAAS,CAAC,KAAK,CAAC,MAAM,EAAE,MAAM,GAAG,gBAAgB,CAAC,CAAC;QACjE,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;QACnB,MAAM,IAAI,IAAI,CAAC;QACf,IAAI,MAAM,GAAG,aAAa,IAAI,SAAS,CAAC,MAAM;YAAE,MAAM;IACxD,CAAC;IAED,QAAQ,CAAC,WAAW,EAAE,cAAc,MAAM,CAAC,MAAM,oBAAoB,aAAa,GAAG,CAAC,CAAC;IACvF,OAAO,MAAM,CAAC;AAChB,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,kBAAkB,CAAC,KAAa,EAAE,WAAmB;IACnE,OAAO,iBAAiB,CAAC,GAAG,KAAK,KAAK,WAAW,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC;AAC9D,CAAC"}
+{"version":3,"file":"content-extractor.js","sourceRoot":"","sources":["../../src/indexing/content-extractor.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,iBAAiB,EAAE,iBAAiB,EAAE,MAAM,8BAA8B,CAAC;AACpF,OAAO,EAAE,QAAQ,EAAE,MAAM,UAAU,CAAC;AACpC,OAAO,EAAE,wBAAwB,EAAE,MAAM,eAAe,CAAC;AAEzD,gEAAgE;AAChE,MAAM,qBAAqB,GAAG,IAAI,CAAC;AAEnC,+EAA+E;AAC/E,MAAM,aAAa,GAAG,GAAG,CAAC;AAE1B,MAAM,aAAa,GAAG,GAAG,CAAC;AAE1B;;;GAGG;AACH,MAAM,UAAU,sBAAsB,CAAC,WAAmB,EAAE,iBAAyB;IACnF,OAAO,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,qBAAqB,GAAG,WAAW,GAAG,iBAAiB,GAAG,aAAa,CAAC,CAAC;AAC9F,CAAC;AAED;;;;;;GAMG;AACH,MAAM,UAAU,uBAAuB,CACrC,WAAwB,EACxB,gBAA0C,EAC1C,gBAAwB;IAExB,MAAM,SAAS,GAAG,IAAI,iBAAiB,EAAE,CAAC;IAC1C,MAAM,QAAQ,GAAG,SAAS,CAAC,OAAO,CAAC,WAAW,EAAE,gBAAgB,CAAC,CAAC;IAClE,QAAQ,CAAC,WAAW,EAAE,oBAAoB,QAAQ,CAAC,MAAM,QAAQ,CAAC,CAAC;IACnE,MAAM,SAAS,GAAG,wBAAwB,CAAC,iBAAiB,CAAC,QAAQ,CAAC,CAAC,CAAC;IACxE,QAAQ,CACN,WAAW,EACX,sBAAsB,SAAS,CAAC,MAAM,uBAAuB,gBAAgB,GAAG,CACjF,CAAC;IAEF,IAAI,SAAS,CAAC,MAAM,IAAI,gBAAgB,EAAE,CAAC;QACzC,OAAO,CAAC,SAAS,CAAC,CAAC;IACrB,CAAC;IACD,IAAI,gBAAgB,IAAI,aAAa,IAAI,gBAAgB,IAAI,CAAC,EAAE,CAAC;QAC/D,OAAO,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,gBAAgB,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;IAC7D,CAAC;IAED,MAAM,MAAM,GAAa,EAAE,CAAC;IAC5B,IAAI,MAAM,GAAG,CAAC,CAAC;IACf,MAAM,IAAI,GAAG,gBAAgB,GAAG,aAAa,CAAC;IAE9C,OAAO,MAAM,GAAG,SAAS,CAAC,MAAM,EAAE,CAAC;QACjC,MAAM,KAAK,GAAG,SAAS,CAAC,KAAK,CAAC,MAAM,EAAE,MAAM,GAAG,gBAAgB,CAAC,CAAC;QACjE,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;QACnB,MAAM,IAAI,IAAI,CAAC;QACf,IAAI,MAAM,GAAG,aAAa,IAAI,SAAS,CAAC,MAAM;YAAE,MAAM;IACxD,CAAC;IAED,QAAQ,CAAC,WAAW,EAAE,cAAc,MAAM,CAAC,MAAM,oBAAoB,aAAa,GAAG,CAAC,CAAC;IACvF,OAAO,MAAM,CAAC;AAChB,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,kBAAkB,CAAC,KAAa,EAAE,WAAmB;IACnE,OAAO,iBAAiB,CAAC,GAAG,KAAK,KAAK,WAAW,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC;AAC9D,CAAC"}

package/dist/indexing/index.d.ts

@@ -1,5 +1,8 @@
+export type { ArticleIndexableHooks, ArticleMarkdownConverterParams, IndexArticleLinksOptions, } from './article-indexable';
+export { defaultShouldIndexArticleLink, indexArticleLinksToIndexableContent, } from './article-indexable';
 export { calculateMaxBodyLength, extractSearchableChunks, extractShallowText, } from './content-extractor';
 export { allChunkIds, baseEntryId, buildSearchDocuments, chunkDocumentId, shouldIndex, } from './document-builder';
+export { calculateReadingTime, DEFAULT_WORDS_PER_MINUTE, stripMarkdownToPlainText, } from './readability';
 export { rebuildSearchIndex } from './rebuild';
 export { rebuildSearchIndexWordPress } from './rebuild-wordpress';
 //# sourceMappingURL=index.d.ts.map

package/dist/indexing/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/indexing/index.ts"],"names":[],"mappings":"AAAA,OAAO,EACL,sBAAsB,EACtB,uBAAuB,EACvB,kBAAkB,GACnB,MAAM,qBAAqB,CAAC;AAC7B,OAAO,EACL,WAAW,EACX,WAAW,EACX,oBAAoB,EACpB,eAAe,EACf,WAAW,GACZ,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EAAE,kBAAkB,EAAE,MAAM,WAAW,CAAC;AAC/C,OAAO,EAAE,2BAA2B,EAAE,MAAM,qBAAqB,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/indexing/index.ts"],"names":[],"mappings":"AAAA,YAAY,EACV,qBAAqB,EACrB,8BAA8B,EAC9B,wBAAwB,GACzB,MAAM,qBAAqB,CAAC;AAC7B,OAAO,EACL,6BAA6B,EAC7B,mCAAmC,GACpC,MAAM,qBAAqB,CAAC;AAC7B,OAAO,EACL,sBAAsB,EACtB,uBAAuB,EACvB,kBAAkB,GACnB,MAAM,qBAAqB,CAAC;AAC7B,OAAO,EACL,WAAW,EACX,WAAW,EACX,oBAAoB,EACpB,eAAe,EACf,WAAW,GACZ,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EACL,oBAAoB,EACpB,wBAAwB,EACxB,wBAAwB,GACzB,MAAM,eAAe,CAAC;AACvB,OAAO,EAAE,kBAAkB,EAAE,MAAM,WAAW,CAAC;AAC/C,OAAO,EAAE,2BAA2B,EAAE,MAAM,qBAAqB,CAAC"}

package/dist/indexing/index.js

@@ -1,5 +1,7 @@
+export { defaultShouldIndexArticleLink, indexArticleLinksToIndexableContent, } from './article-indexable';
 export { calculateMaxBodyLength, extractSearchableChunks, extractShallowText, } from './content-extractor';
 export { allChunkIds, baseEntryId, buildSearchDocuments, chunkDocumentId, shouldIndex, } from './document-builder';
+export { calculateReadingTime, DEFAULT_WORDS_PER_MINUTE, stripMarkdownToPlainText, } from './readability';
 export { rebuildSearchIndex } from './rebuild';
 export { rebuildSearchIndexWordPress } from './rebuild-wordpress';
 //# sourceMappingURL=index.js.map

package/dist/indexing/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../src/indexing/index.ts"],"names":[],"mappings":"AAAA,OAAO,EACL,sBAAsB,EACtB,uBAAuB,EACvB,kBAAkB,GACnB,MAAM,qBAAqB,CAAC;AAC7B,OAAO,EACL,WAAW,EACX,WAAW,EACX,oBAAoB,EACpB,eAAe,EACf,WAAW,GACZ,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EAAE,kBAAkB,EAAE,MAAM,WAAW,CAAC;AAC/C,OAAO,EAAE,2BAA2B,EAAE,MAAM,qBAAqB,CAAC"}
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../src/indexing/index.ts"],"names":[],"mappings":"AAKA,OAAO,EACL,6BAA6B,EAC7B,mCAAmC,GACpC,MAAM,qBAAqB,CAAC;AAC7B,OAAO,EACL,sBAAsB,EACtB,uBAAuB,EACvB,kBAAkB,GACnB,MAAM,qBAAqB,CAAC;AAC7B,OAAO,EACL,WAAW,EACX,WAAW,EACX,oBAAoB,EACpB,eAAe,EACf,WAAW,GACZ,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EACL,oBAAoB,EACpB,wBAAwB,EACxB,wBAAwB,GACzB,MAAM,eAAe,CAAC;AACvB,OAAO,EAAE,kBAAkB,EAAE,MAAM,WAAW,CAAC;AAC/C,OAAO,EAAE,2BAA2B,EAAE,MAAM,qBAAqB,CAAC"}

package/dist/indexing/readability.d.ts

@@ -0,0 +1,14 @@
+/** Default reading speed for {@link calculateReadingTime}. */
+export declare const DEFAULT_WORDS_PER_MINUTE = 200;
+/**
+ * Estimates reading time in minutes from plain text (e.g. after markdown stripping).
+ * Empty or whitespace-only input returns `1` (same convention as marketing-site reading time).
+ */
+export declare function calculateReadingTime(text: string, wordsPerMinute?: number): number;
+/**
+ * Strip markdown syntax leaving only readable words for search indexing or word counts.
+ * Removes frontmatter, heading markers, bold/italic, links (keeping text), images,
+ * fenced and inline code, blockquotes, list markers, and horizontal rules.
+ */
+export declare function stripMarkdownToPlainText(markdown: string): string;
+//# sourceMappingURL=readability.d.ts.map

package/dist/indexing/readability.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"readability.d.ts","sourceRoot":"","sources":["../../src/indexing/readability.ts"],"names":[],"mappings":"AAAA,8DAA8D;AAC9D,eAAO,MAAM,wBAAwB,MAAM,CAAC;AAE5C;;;GAGG;AACH,wBAAgB,oBAAoB,CAClC,IAAI,EAAE,MAAM,EACZ,cAAc,GAAE,MAAiC,GAChD,MAAM,CAGR;AAED;;;;GAIG;AACH,wBAAgB,wBAAwB,CAAC,QAAQ,EAAE,MAAM,GAAG,MAAM,CA8BjE"}

package/dist/indexing/readability.js

@@ -0,0 +1,45 @@
+/** Default reading speed for {@link calculateReadingTime}. */
+export const DEFAULT_WORDS_PER_MINUTE = 200;
+/**
+ * Estimates reading time in minutes from plain text (e.g. after markdown stripping).
+ * Empty or whitespace-only input returns `1` (same convention as marketing-site reading time).
+ */
+export function calculateReadingTime(text, wordsPerMinute = DEFAULT_WORDS_PER_MINUTE) {
+    const words = text.trim().split(/\s+/).filter(Boolean);
+    return Math.max(1, Math.round(words.length / wordsPerMinute));
+}
+/**
+ * Strip markdown syntax leaving only readable words for search indexing or word counts.
+ * Removes frontmatter, heading markers, bold/italic, links (keeping text), images,
+ * fenced and inline code, blockquotes, list markers, and horizontal rules.
+ */
+export function stripMarkdownToPlainText(markdown) {
+    return (markdown
+        // Remove YAML frontmatter block
+        .replace(/^---[\s\S]*?---\n?/, '')
+        // Fenced blocks before inline backticks (avoids corrupting ``` fences)
+        .replace(/```[\s\S]*?```/g, '')
+        // Remove images before links (order matters)
+        .replace(/!\[[^\]]*\]\([^)]+\)/g, '')
+        // Remove links, keep display text
+        .replace(/\[([^\]]*)\]\([^)]*\)/g, '$1')
+        // Remove heading markers
+        .replace(/^#{1,6}\s+/gm, '')
+        // Remove bold/italic (triple, double, single asterisk/underscore)
+        .replace(/\*{1,3}([^*\n]*)\*{1,3}/g, '$1')
+        .replace(/_{1,3}([^_\n]*)_{1,3}/g, '$1')
+        // Remove inline code backticks
+        .replace(/`([^`]+)`/g, '$1')
+        // Remove blockquote markers
+        .replace(/^>\s*/gm, '')
+        // Remove unordered list markers
+        .replace(/^[-*+]\s+/gm, '')
+        // Remove ordered list markers
+        .replace(/^\d+\.\s+/gm, '')
+        // Remove horizontal rules
+        .replace(/^[-*_]{3,}\s*$/gm, '')
+        // Collapse excess whitespace
+        .replace(/\n{3,}/g, '\n\n')
+        .trim());
+}
+//# sourceMappingURL=readability.js.map

package/dist/indexing/readability.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"readability.js","sourceRoot":"","sources":["../../src/indexing/readability.ts"],"names":[],"mappings":"AAAA,8DAA8D;AAC9D,MAAM,CAAC,MAAM,wBAAwB,GAAG,GAAG,CAAC;AAE5C;;;GAGG;AACH,MAAM,UAAU,oBAAoB,CAClC,IAAY,EACZ,iBAAyB,wBAAwB;IAEjD,MAAM,KAAK,GAAG,IAAI,CAAC,IAAI,EAAE,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;IACvD,OAAO,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,MAAM,GAAG,cAAc,CAAC,CAAC,CAAC;AAChE,CAAC;AAED;;;;GAIG;AACH,MAAM,UAAU,wBAAwB,CAAC,QAAgB;IACvD,OAAO,CACL,QAAQ;QACN,gCAAgC;SAC/B,OAAO,CAAC,oBAAoB,EAAE,EAAE,CAAC;QAClC,uEAAuE;SACtE,OAAO,CAAC,iBAAiB,EAAE,EAAE,CAAC;QAC/B,6CAA6C;SAC5C,OAAO,CAAC,uBAAuB,EAAE,EAAE,CAAC;QACrC,kCAAkC;SACjC,OAAO,CAAC,wBAAwB,EAAE,IAAI,CAAC;QACxC,yBAAyB;SACxB,OAAO,CAAC,cAAc,EAAE,EAAE,CAAC;QAC5B,kEAAkE;SACjE,OAAO,CAAC,0BAA0B,EAAE,IAAI,CAAC;SACzC,OAAO,CAAC,wBAAwB,EAAE,IAAI,CAAC;QACxC,+BAA+B;SAC9B,OAAO,CAAC,YAAY,EAAE,IAAI,CAAC;QAC5B,4BAA4B;SAC3B,OAAO,CAAC,SAAS,EAAE,EAAE,CAAC;QACvB,gCAAgC;SAC/B,OAAO,CAAC,aAAa,EAAE,EAAE,CAAC;QAC3B,8BAA8B;SAC7B,OAAO,CAAC,aAAa,EAAE,EAAE,CAAC;QAC3B,0BAA0B;SACzB,OAAO,CAAC,kBAAkB,EAAE,EAAE,CAAC;QAChC,6BAA6B;SAC5B,OAAO,CAAC,SAAS,EAAE,MAAM,CAAC;SAC1B,IAAI,EAAE,CACV,CAAC;AACJ,CAAC"}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@se-studio/search",
-  "version": "1.0.36",
+  "version": "1.0.38",
   "description": "AI-powered site search with Upstash Search for Next.js marketing sites",
   "repository": {
     "type": "git",
@@ -43,22 +43,22 @@
   ],
   "dependencies": {
     "@upstash/search": "^0.1.7",
-    "@se-studio/contentful-rest-api": "1.0.128",
-    "@se-studio/core-data-types": "1.0.125",
-    "@se-studio/markdown-renderer": "1.0.89",
-    "@se-studio/wordpress-rest-api": "1.0.6"
+    "@se-studio/contentful-rest-api": "1.0.132",
+    "@se-studio/core-data-types": "1.0.127",
+    "@se-studio/markdown-renderer": "1.0.91",
+    "@se-studio/wordpress-rest-api": "1.0.8"
   },
   "peerDependencies": {
     "next": ">=15.5.0",
     "react": "^19.0.0"
   },
   "devDependencies": {
-    "@biomejs/biome": "^2.4.10",
-    "@types/node": "^22.19.15",
+    "@biomejs/biome": "^2.4.12",
+    "@types/node": "^22.19.17",
     "@types/react": "^19.2.14",
-    "next": "^15.5.14",
+    "next": "^15.5.15",
     "typescript": "^6.0.2",
-    "vitest": "^4.1.2"
+    "vitest": "^4.1.4"
   },
   "scripts": {
     "build": "tsc --project tsconfig.build.json",