@fullstackdatasolutions/articles 1.2.3 → 1.3.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +46 -0
- package/README.md +313 -1
- package/dist/index.cjs +308 -79
- package/dist/index.cjs.map +1 -1
- package/dist/index.d.cts +267 -16
- package/dist/index.d.ts +267 -16
- package/dist/index.js +300 -79
- package/dist/index.js.map +1 -1
- package/dist/nextjs.cjs +325 -31
- package/dist/nextjs.cjs.map +1 -1
- package/dist/nextjs.d.cts +179 -2
- package/dist/nextjs.d.ts +179 -2
- package/dist/nextjs.js +325 -31
- package/dist/nextjs.js.map +1 -1
- package/dist/server.cjs +660 -50
- package/dist/server.cjs.map +1 -1
- package/dist/server.d.cts +333 -12
- package/dist/server.d.ts +333 -12
- package/dist/server.js +645 -50
- package/dist/server.js.map +1 -1
- package/package.json +1 -1
- package/src/ArticleAnswer.tsx +35 -0
- package/src/ArticleSchemas.tsx +263 -23
- package/src/AuthorArticlesPage.tsx +38 -8
- package/src/__tests__/ArticleAnswer.test.tsx +25 -0
- package/src/__tests__/ArticleSchemas.test.tsx +516 -0
- package/src/__tests__/AuthorArticlesPage.test.tsx +76 -0
- package/src/__tests__/authorUtils.test.ts +50 -0
- package/src/__tests__/markdown.test.ts +77 -1
- package/src/__tests__/nextjs.test.ts +31 -15
- package/src/__tests__/seoUtils.test.ts +279 -0
- package/src/__tests__/server-articles.test.ts +434 -1
- package/src/__tests__/validateArticles.test.ts +167 -6
- package/src/articleTypes.ts +57 -0
- package/src/articlesConfig.ts +176 -1
- package/src/authorUtils.ts +19 -1
- package/src/errorReporting.ts +1 -0
- package/src/index.ts +17 -1
- package/src/markdown.ts +100 -1
- package/src/nextjs.ts +7 -4
- package/src/seoUtils.ts +247 -26
- package/src/server-articles.ts +385 -25
- package/src/server.ts +35 -4
- package/src/validateArticles.ts +157 -12
package/dist/server.d.ts
CHANGED
|
@@ -15,6 +15,21 @@ interface HowToStep {
|
|
|
15
15
|
name: string;
|
|
16
16
|
text: string;
|
|
17
17
|
}
|
|
18
|
+
/**
|
|
19
|
+
* A real-world thing an article is about, emitted as schema.org `about`.
|
|
20
|
+
* `sameAs` should point at an authoritative identifier for the entity
|
|
21
|
+
* (Wikipedia, Wikidata, an official site) - that URL is what lets a consumer
|
|
22
|
+
* resolve "Pathfinder" the game system rather than guessing from the string.
|
|
23
|
+
*/
|
|
24
|
+
interface EntityReference {
|
|
25
|
+
name: string;
|
|
26
|
+
sameAs?: string;
|
|
27
|
+
}
|
|
28
|
+
/** An outbound source an article cites, emitted as schema.org `citation`. */
|
|
29
|
+
interface CitationReference {
|
|
30
|
+
name: string;
|
|
31
|
+
url?: string;
|
|
32
|
+
}
|
|
18
33
|
interface Article {
|
|
19
34
|
slug: string;
|
|
20
35
|
title: string;
|
|
@@ -39,6 +54,18 @@ interface Article {
|
|
|
39
54
|
toc?: TocItem[];
|
|
40
55
|
faq?: FaqItem[];
|
|
41
56
|
howTo?: HowToStep[];
|
|
57
|
+
/**
|
|
58
|
+
* A direct, self-contained answer to the article's core question, 40-60
|
|
59
|
+
* words. Rendered by `ArticleAnswer`, emitted as the Article schema's
|
|
60
|
+
* `abstract`, used as the `acceptedAnswer` when `articleType` is
|
|
61
|
+
* `'QAPage'`, and placed at the top of the article's markdown twin - the
|
|
62
|
+
* one passage most likely to be lifted verbatim by an answer engine.
|
|
63
|
+
*/
|
|
64
|
+
answer?: string;
|
|
65
|
+
/** Entities this article is about. Emitted as schema.org `about`. */
|
|
66
|
+
about?: EntityReference[];
|
|
67
|
+
/** Outbound sources. Emitted as schema.org `citation`. */
|
|
68
|
+
citation?: CitationReference[];
|
|
42
69
|
canonicalUrl?: string;
|
|
43
70
|
articleType?: string;
|
|
44
71
|
series?: string;
|
|
@@ -153,6 +180,13 @@ interface AuthorProfile {
|
|
|
153
180
|
promise?: string;
|
|
154
181
|
/** Structured long-form origin story - see `RichText`/`RichTextSection`. */
|
|
155
182
|
originStory?: RichText;
|
|
183
|
+
/**
|
|
184
|
+
* Topics this author writes about, emitted as Person `knowsAbout`. Omit to
|
|
185
|
+
* derive it from the categories of their own published articles - unlike
|
|
186
|
+
* `credentials`/`proof`, subject matter is verifiable from the corpus
|
|
187
|
+
* itself, so it does belong in structured data.
|
|
188
|
+
*/
|
|
189
|
+
knowsAbout?: string[];
|
|
156
190
|
/** Who this author's content/work is for, e.g. "New game masters", "Streaming DMs". */
|
|
157
191
|
servesWho?: string[];
|
|
158
192
|
/** Core beliefs/approach statements. */
|
|
@@ -165,6 +199,27 @@ interface AuthorProfile {
|
|
|
165
199
|
credentials?: string[];
|
|
166
200
|
/** Concrete, sourceable proof points. */
|
|
167
201
|
proof?: ProofItem[];
|
|
202
|
+
/**
|
|
203
|
+
* The one canonical URL identifying this person across every site they
|
|
204
|
+
* publish on. Used *only* to derive the Person `@id`, so the same author
|
|
205
|
+
* resolves to a single entity everywhere instead of one entity per site.
|
|
206
|
+
*
|
|
207
|
+
* Deliberately separate from `url`: that field also drives byline link
|
|
208
|
+
* targets, the author page's Open Graph URL, and its `CollectionPage` URL,
|
|
209
|
+
* so pointing it at another domain would send readers off-site and hand
|
|
210
|
+
* this site's author page a canonical belonging to a different one.
|
|
211
|
+
* `identityUrl` changes nothing a reader sees.
|
|
212
|
+
*
|
|
213
|
+
* Set the same value on every site. Falls back to `url`, then to this
|
|
214
|
+
* site's own author page.
|
|
215
|
+
*/
|
|
216
|
+
identityUrl?: string;
|
|
217
|
+
/**
|
|
218
|
+
* Additional profile URLs identifying this same person elsewhere, merged
|
|
219
|
+
* into the Person schema's `sameAs` alongside the derived social links.
|
|
220
|
+
* List the author's pages on the other sites here.
|
|
221
|
+
*/
|
|
222
|
+
sameAs?: string[];
|
|
168
223
|
/** Primary call-to-action rendered on the author's page. */
|
|
169
224
|
primaryCta?: {
|
|
170
225
|
label: string;
|
|
@@ -346,6 +401,54 @@ interface BreadcrumbsConfig {
|
|
|
346
401
|
/** Optional label overrides for built-in breadcrumb items. */
|
|
347
402
|
labels?: BreadcrumbLabels;
|
|
348
403
|
}
|
|
404
|
+
/** Payload passed to `ArticlesConfig.onAiCrawl`. */
|
|
405
|
+
interface AiCrawlEvent {
|
|
406
|
+
/** Article slug whose markdown twin was fetched. */
|
|
407
|
+
slug: string;
|
|
408
|
+
/** Matched crawler name (e.g. `'GPTBot'`), or `'unknown'` when the agent is not recognized. */
|
|
409
|
+
crawler: string;
|
|
410
|
+
/** Raw `User-Agent` header, or an empty string when absent. */
|
|
411
|
+
userAgent: string;
|
|
412
|
+
}
|
|
413
|
+
/**
|
|
414
|
+
* The publishing entity behind the site. Drives `OrganizationSchema`,
|
|
415
|
+
* `WebSiteSchema`, and the `publisher` reference on every article - so all
|
|
416
|
+
* three point at one `@id` instead of repeating an inline, unlinked
|
|
417
|
+
* `Organization` stub per page.
|
|
418
|
+
*/
|
|
419
|
+
interface OrganizationConfig {
|
|
420
|
+
/** schema.org type. Use `'Person'` for a personal brand. Default: `'Organization'`. */
|
|
421
|
+
type?: 'Organization' | 'Person';
|
|
422
|
+
/** Entity name. Falls back to `siteName`. */
|
|
423
|
+
name?: string;
|
|
424
|
+
/** Entity homepage. Falls back to `siteUrl`. */
|
|
425
|
+
url?: string;
|
|
426
|
+
/** Logo URL. Site-relative paths are resolved against `siteUrl`. */
|
|
427
|
+
logo?: string;
|
|
428
|
+
/** Entity description. Falls back to `ArticlesConfig.description`. */
|
|
429
|
+
description?: string;
|
|
430
|
+
/** Profile URLs that identify the same entity elsewhere (social, Crunchbase, Wikidata). */
|
|
431
|
+
sameAs?: string[];
|
|
432
|
+
/**
|
|
433
|
+
* The umbrella entity this site belongs to, for a network of sites run by
|
|
434
|
+
* one publisher. Emitted as `parentOrganization` - the link that lets
|
|
435
|
+
* authority earned by the network attach to each site in it, instead of
|
|
436
|
+
* each site standing alone.
|
|
437
|
+
*/
|
|
438
|
+
parentOrganization?: {
|
|
439
|
+
name: string;
|
|
440
|
+
url: string;
|
|
441
|
+
};
|
|
442
|
+
/**
|
|
443
|
+
* Search URL template for the `WebSite` `SearchAction`, e.g.
|
|
444
|
+
* `'/search?q={search_term_string}'`. Omitted by default - the library's
|
|
445
|
+
* own search is client-side with no crawlable results URL, so declaring one
|
|
446
|
+
* that does not exist would be a false claim. Only set this if the app
|
|
447
|
+
* actually serves search results at that URL. Must contain the literal
|
|
448
|
+
* `{search_term_string}` placeholder.
|
|
449
|
+
*/
|
|
450
|
+
searchUrlTemplate?: string;
|
|
451
|
+
}
|
|
349
452
|
/** Top-level configuration object. Pass one instance to every library component. */
|
|
350
453
|
interface ArticlesConfig {
|
|
351
454
|
/** Canonical base URL of the site, used in metadata and JSON-LD. Example: `'https://yoursite.com'` */
|
|
@@ -408,6 +511,94 @@ interface ArticlesConfig {
|
|
|
408
511
|
* step rather than a build-time failure otherwise.
|
|
409
512
|
*/
|
|
410
513
|
paths?: Record<string, PathDefinition>;
|
|
514
|
+
/**
|
|
515
|
+
* Template for the `<title>` tag on article, category, series, and author
|
|
516
|
+
* pages. Supports `{title}` and `{siteName}` placeholders.
|
|
517
|
+
* Default: `'{title} | {siteName}'`.
|
|
518
|
+
*
|
|
519
|
+
* Google truncates a result title around 60 characters, and a site name
|
|
520
|
+
* suffix spends that budget on every page. Set `'{title}'` to drop it when
|
|
521
|
+
* your titles are already long and your brand draws little search volume -
|
|
522
|
+
* the suffix is only earning its characters if people search for the brand.
|
|
523
|
+
*/
|
|
524
|
+
titleTemplate?: string;
|
|
525
|
+
/**
|
|
526
|
+
* BCP 47 language tag for the site's content. Emitted as the Article
|
|
527
|
+
* schema's `inLanguage` and the RSS channel `<language>`. Default: `'en'`.
|
|
528
|
+
*/
|
|
529
|
+
language?: string;
|
|
530
|
+
/**
|
|
531
|
+
* Set to `false` when articles sit behind a paywall or registration wall.
|
|
532
|
+
* Default: `true`, emitted as the Article schema's `isAccessibleForFree` -
|
|
533
|
+
* an explicit "this is readable" signal, since consumers that cannot tell
|
|
534
|
+
* tend to skip suspected-paywalled sources.
|
|
535
|
+
*/
|
|
536
|
+
isAccessibleForFree?: boolean;
|
|
537
|
+
/**
|
|
538
|
+
* CSS selectors marking the parts of an article suitable for text-to-speech,
|
|
539
|
+
* emitted as the Article schema's `speakable`. Omitted by default - the
|
|
540
|
+
* correct selectors depend on the consuming app's own markup, and guessing
|
|
541
|
+
* them would point at elements that may not exist.
|
|
542
|
+
*/
|
|
543
|
+
speakableSelectors?: string[];
|
|
544
|
+
/**
|
|
545
|
+
* Set to `true` to derive `FAQPage` entries from question-shaped `##`
|
|
546
|
+
* headings and the paragraph that follows each one. Default: `false` -
|
|
547
|
+
* turning prose into structured data without the author's intent can
|
|
548
|
+
* promote a rhetorical heading into a published Q&A pair, so this is
|
|
549
|
+
* opt-in. Explicit `faq` frontmatter always wins over derived entries.
|
|
550
|
+
*/
|
|
551
|
+
deriveFaqFromHeadings?: boolean;
|
|
552
|
+
/**
|
|
553
|
+
* Shared entity vocabulary, keyed by an app-chosen slug. Article `about`
|
|
554
|
+
* entries may reference a key here instead of repeating a name/`sameAs`
|
|
555
|
+
* pair - the point of `about` is that a consumer can resolve one entity
|
|
556
|
+
* across a corpus, which free-text names spelled three different ways
|
|
557
|
+
* defeat. `validateArticles` warns on `about` keys with no registry entry.
|
|
558
|
+
*/
|
|
559
|
+
entities?: Record<string, EntityReference>;
|
|
560
|
+
/**
|
|
561
|
+
* Where `lastmod` comes from when frontmatter omits it.
|
|
562
|
+
* - `'published'` (default): reuse the publish date, as before 1.3.0.
|
|
563
|
+
* - `'none'`: leave `lastmod` unset, so `dateModified` is omitted rather
|
|
564
|
+
* than repeating a stale publish date.
|
|
565
|
+
* - `'fileMtime'`: read the article file's modification time. Accurate
|
|
566
|
+
* locally; on a CI runner that clones fresh, every file's mtime is the
|
|
567
|
+
* checkout time, which would report the whole corpus as updated today.
|
|
568
|
+
* Only use it where the build preserves mtimes.
|
|
569
|
+
*/
|
|
570
|
+
lastmodFallback?: 'published' | 'none' | 'fileMtime';
|
|
571
|
+
/**
|
|
572
|
+
* Called when an AI crawler fetches an article's markdown twin. The one
|
|
573
|
+
* choke point where those requests land, so it is the only place a site
|
|
574
|
+
* can measure whether any of its AI-readable content is being read, and
|
|
575
|
+
* by which bot. No PII: the payload carries the slug, the matched crawler
|
|
576
|
+
* name, and the raw user agent string only.
|
|
577
|
+
*/
|
|
578
|
+
onAiCrawl?: (event: AiCrawlEvent) => void;
|
|
579
|
+
/**
|
|
580
|
+
* The publishing entity behind the site. Omit to keep the pre-1.3.0 inline
|
|
581
|
+
* `{'@type':'Organization', name: siteName}` publisher stub on articles.
|
|
582
|
+
* Set it to emit `OrganizationSchema`/`WebSiteSchema` in the root layout and
|
|
583
|
+
* have every article, author, and collection page reference the same `@id`.
|
|
584
|
+
*/
|
|
585
|
+
organization?: OrganizationConfig;
|
|
586
|
+
/**
|
|
587
|
+
* Default `aiCrawl` value for articles whose frontmatter omits the key.
|
|
588
|
+
* Default: `false` (every article stays opted out unless it sets
|
|
589
|
+
* `aiCrawl: true`). Set to `true` on a site whose goal is being cited by
|
|
590
|
+
* answer engines to opt the whole corpus in at once; per-article
|
|
591
|
+
* `aiCrawl: false` still wins and keeps that article blocked.
|
|
592
|
+
*/
|
|
593
|
+
aiCrawlDefault?: boolean;
|
|
594
|
+
/**
|
|
595
|
+
* Set to `false` to serve article markdown twins as the bare body, with no
|
|
596
|
+
* attribution header. Default: `true` - the twin is prefixed with the
|
|
597
|
+
* title, excerpt, canonical source URL, dates, author, and site name so a
|
|
598
|
+
* model reading `/articles/[slug].md` can attribute it. Only applies when
|
|
599
|
+
* a config is available (i.e. via `getArticleMarkdownResponse`).
|
|
600
|
+
*/
|
|
601
|
+
markdownTwinHeader?: boolean;
|
|
411
602
|
/**
|
|
412
603
|
* Vendor-neutral event callback (Phase 27F). Fired by components/hooks at
|
|
413
604
|
* meaningful reader-journey moments (see `ArticleEvent` in `events.ts`).
|
|
@@ -417,6 +608,18 @@ interface ArticlesConfig {
|
|
|
417
608
|
onEvent?: ArticleEventHandler;
|
|
418
609
|
}
|
|
419
610
|
declare function getBreadcrumbsConfig(config: ArticlesConfig): BreadcrumbsConfig;
|
|
611
|
+
/** Stable `@id` for the site's publishing entity. */
|
|
612
|
+
declare function getOrganizationId(config: Pick<ArticlesConfig, 'siteUrl'>): string;
|
|
613
|
+
/** Stable `@id` for the site itself. */
|
|
614
|
+
declare function getWebSiteId(config: Pick<ArticlesConfig, 'siteUrl'>): string;
|
|
615
|
+
/** Stable `@id` for an author, so every article by them resolves to one Person. */
|
|
616
|
+
declare function getPersonId(authorUrl: string): string;
|
|
617
|
+
/**
|
|
618
|
+
* Applies `ArticlesConfig.titleTemplate` to a page title. Kept here rather
|
|
619
|
+
* than inlined at each call site so article, category, series, and author
|
|
620
|
+
* pages can never drift apart on how the site name is appended.
|
|
621
|
+
*/
|
|
622
|
+
declare function formatPageTitle(title: string, config: Pick<ArticlesConfig, 'siteName' | 'titleTemplate'>): string;
|
|
420
623
|
|
|
421
624
|
declare function sanitizeImagePath(rawPath: string, articleSlug: string): string | null;
|
|
422
625
|
declare function getAvailableArticleSlugs(): string[];
|
|
@@ -429,11 +632,64 @@ declare function getAdjacentArticles(currentSlug: string): Promise<{
|
|
|
429
632
|
previous: Article | null;
|
|
430
633
|
next: Article | null;
|
|
431
634
|
}>;
|
|
432
|
-
declare function getArticleMarkdown(slug: string): Promise<string | null>;
|
|
433
|
-
|
|
635
|
+
declare function getArticleMarkdown(slug: string, config?: ArticlesConfig): Promise<string | null>;
|
|
636
|
+
/**
|
|
637
|
+
* Crawlers that read content on behalf of an answer engine. Used both to
|
|
638
|
+
* write robots.txt rules and to classify markdown-twin fetches for
|
|
639
|
+
* `ArticlesConfig.onAiCrawl` - one list, so the two can never disagree about
|
|
640
|
+
* what counts as an AI crawler.
|
|
641
|
+
*/
|
|
642
|
+
declare const AI_CRAWLERS: readonly ["GPTBot", "ChatGPT-User", "OAI-SearchBot", "CCBot", "ClaudeBot", "Claude-User", "Claude-SearchBot", "anthropic-ai", "PerplexityBot", "Perplexity-User", "Google-Extended", "Applebot-Extended", "Bytespider", "Amazonbot", "meta-externalagent", "cohere-ai", "DuckAssistBot", "MistralAI-User"];
|
|
643
|
+
/** Matches a `User-Agent` against `AI_CRAWLERS`, returning the crawler name or `null`. */
|
|
644
|
+
declare function matchAiCrawler(userAgent: string): string | null;
|
|
645
|
+
/**
|
|
646
|
+
* Attribution header prepended to an article's markdown twin.
|
|
647
|
+
*
|
|
648
|
+
* The twin is what an AI crawler actually reads, and `matter()` strips every
|
|
649
|
+
* frontmatter field before it is served - so without this the model gets an
|
|
650
|
+
* anonymous body with no title, date, author, or link back to the canonical
|
|
651
|
+
* page. Skips its own `# {title}` line when the body already opens with the
|
|
652
|
+
* same H1, so the common "body repeats the title" layout doesn't end up with
|
|
653
|
+
* two.
|
|
654
|
+
*/
|
|
655
|
+
declare function buildMarkdownTwinHeader(article: Article, config: ArticlesConfig, body: string): string;
|
|
656
|
+
/** Minimal shape of a request's headers - avoids depending on `next/server` here. */
|
|
657
|
+
type RequestHeaders = Readonly<{
|
|
658
|
+
get(name: string): string | null;
|
|
659
|
+
}>;
|
|
660
|
+
/** Markdown twin for `/articles/category/[category]`. `null` when the category has no articles. */
|
|
661
|
+
declare function getCategoryMarkdown(categorySlug: string, config: ArticlesConfig): Promise<string | null>;
|
|
662
|
+
/**
|
|
663
|
+
* Markdown twin for `/articles/authors/[author]`.
|
|
664
|
+
*
|
|
665
|
+
* Carries the author's bio, promise, principles, and sourced proof alongside
|
|
666
|
+
* their article list - the "who is this and why trust them" context that
|
|
667
|
+
* otherwise exists only inside React components, and the exact question asked
|
|
668
|
+
* before anything they wrote gets cited. `credentials` are included as the
|
|
669
|
+
* author's own stated claims; unlike JSON-LD, prose can attribute a claim
|
|
670
|
+
* without asserting it as a verified fact.
|
|
671
|
+
*/
|
|
672
|
+
declare function getAuthorMarkdown(authorSlug: string, config: ArticlesConfig): Promise<string | null>;
|
|
673
|
+
/** Markdown twin for `/articles/series/[series]`. `null` when the series has no articles. */
|
|
674
|
+
declare function getSeriesMarkdown(seriesSlug: string, config: ArticlesConfig): Promise<string | null>;
|
|
675
|
+
/**
|
|
676
|
+
* Serves the markdown twin for any `/articles/...` path - an article, or a
|
|
677
|
+
* category/author/series listing.
|
|
678
|
+
*
|
|
679
|
+
* Dispatching on the slug prefix here rather than adding three more app
|
|
680
|
+
* routes keeps the existing single rewrite (`/articles/:path*.md`) working
|
|
681
|
+
* unchanged: without it, `/articles/category/campaigns.md` matches that
|
|
682
|
+
* rewrite, reaches the article handler, and 404s.
|
|
683
|
+
*/
|
|
684
|
+
declare function getMarkdownTwinResponse(slug: string, config: ArticlesConfig, options?: Readonly<{
|
|
685
|
+
headers?: RequestHeaders;
|
|
686
|
+
}>): Promise<Response>;
|
|
687
|
+
declare function getArticleMarkdownResponse(slug: string, config: ArticlesConfig, options?: Readonly<{
|
|
688
|
+
headers?: RequestHeaders;
|
|
689
|
+
}>): Promise<Response>;
|
|
434
690
|
declare function getArticleMarkdownUrl(article: Pick<Article, 'slug' | 'aiCrawl'>, config?: Pick<ArticlesConfig, 'siteUrl'>): string | undefined;
|
|
435
691
|
declare function getArticleAiHeaders(article: Pick<Article, 'slug' | 'aiCrawl'>, config?: Pick<ArticlesConfig, 'siteUrl'>): Record<string, string>;
|
|
436
|
-
declare function getAiRobotsTxtRules(): Promise<string>;
|
|
692
|
+
declare function getAiRobotsTxtRules(config?: ArticlesConfig): Promise<string>;
|
|
437
693
|
declare function searchArticles(query: string, config?: ArticlesConfig): Promise<Article[]>;
|
|
438
694
|
declare function categoryToSlug(category: string): string;
|
|
439
695
|
declare function getAllCategories(): Promise<CategoryInfo[]>;
|
|
@@ -478,7 +734,39 @@ interface RelatedContentResult {
|
|
|
478
734
|
*/
|
|
479
735
|
declare function getRelatedContent(article: Article, config: ArticlesConfig, limit?: number): Promise<RelatedContentResult>;
|
|
480
736
|
|
|
481
|
-
|
|
737
|
+
/**
|
|
738
|
+
* `options.fullContent` adds `<content:encoded>` with each article's rendered
|
|
739
|
+
* HTML, for feeds meant to be ingested rather than previewed - an
|
|
740
|
+
* excerpt-only feed gives a consumer nothing to work with. Off by default:
|
|
741
|
+
* it requires articles loaded with `htmlContent` (i.e. via
|
|
742
|
+
* `getArticleMetadata`, not `getAllArticles`' summaries), and articles
|
|
743
|
+
* without it are simply emitted without the element.
|
|
744
|
+
*/
|
|
745
|
+
declare function generateRssFeed(articles: Article[], config: ArticlesConfig, options?: Readonly<{
|
|
746
|
+
fullContent?: boolean;
|
|
747
|
+
}>): string;
|
|
748
|
+
/**
|
|
749
|
+
* `llms.txt` index - the emerging convention for pointing an LLM at a site's
|
|
750
|
+
* canonical, markdown-native content (https://llmstxt.org). Lists only
|
|
751
|
+
* articles opted in via `aiCrawl` (see `ArticlesConfig.aiCrawlDefault`),
|
|
752
|
+
* grouped by category, linking to each article's `.md` twin rather than its
|
|
753
|
+
* HTML page.
|
|
754
|
+
*
|
|
755
|
+
* Wire it up in the consuming app as `app/llms.txt/route.ts`:
|
|
756
|
+
* export async function GET() {
|
|
757
|
+
* return new Response(generateLlmsTxt(await getAllArticles(siteConfig), siteConfig), {
|
|
758
|
+
* headers: { 'Content-Type': 'text/plain; charset=utf-8' },
|
|
759
|
+
* })
|
|
760
|
+
* }
|
|
761
|
+
*/
|
|
762
|
+
declare function generateLlmsTxt(articles: Article[], config: ArticlesConfig): string;
|
|
763
|
+
/**
|
|
764
|
+
* `llms-full.txt` - every opted-in article's full markdown twin, headers
|
|
765
|
+
* included, concatenated into one document. Larger and slower to build than
|
|
766
|
+
* `generateLlmsTxt`; generate it in a route handler or at build time, not on
|
|
767
|
+
* every request.
|
|
768
|
+
*/
|
|
769
|
+
declare function generateLlmsFullTxt(articles: Article[], config: ArticlesConfig): Promise<string>;
|
|
482
770
|
declare function generateArticleStaticParams(): {
|
|
483
771
|
slug: string;
|
|
484
772
|
}[];
|
|
@@ -570,8 +858,17 @@ interface ContentSlotBoundaries {
|
|
|
570
858
|
}
|
|
571
859
|
declare function getContentSlotBoundaries(markdown: string): ContentSlotBoundaries | null;
|
|
572
860
|
declare function extractToc(markdown: string): Promise<TocItem[]>;
|
|
861
|
+
/**
|
|
862
|
+
* Derives `FaqItem`s from question-shaped `##` headings and the prose that
|
|
863
|
+
* follows each one, for `ArticlesConfig.deriveFaqFromHeadings`.
|
|
864
|
+
*
|
|
865
|
+
* Line-based on purpose: it runs for every article at summary time, and the
|
|
866
|
+
* shapes it must reject (fenced code, a heading with no prose under it) are
|
|
867
|
+
* cheaper to detect by scanning than by walking a parsed AST.
|
|
868
|
+
*/
|
|
869
|
+
declare function deriveFaqFromHeadings(markdown: string): FaqItem[];
|
|
573
870
|
|
|
574
|
-
type ArticlesErrorCode = 'article-directory-read-failed' | 'article-load-failed' | 'article-markdown-load-failed' | 'markdown-conversion-failed' | 'unsafe-image-path';
|
|
871
|
+
type ArticlesErrorCode = 'article-directory-read-failed' | 'article-load-failed' | 'article-markdown-load-failed' | 'ai-crawl-handler-failed' | 'markdown-conversion-failed' | 'unsafe-image-path';
|
|
575
872
|
type ArticlesErrorContext = Readonly<Record<string, string | number | boolean | undefined>>;
|
|
576
873
|
type ArticlesErrorReport = Readonly<{
|
|
577
874
|
code: ArticlesErrorCode;
|
|
@@ -620,6 +917,22 @@ type ArticleTOCProps = Readonly<{
|
|
|
620
917
|
}>;
|
|
621
918
|
declare function ArticleTOC({ toc, className }: ArticleTOCProps): react_jsx_runtime.JSX.Element | null;
|
|
622
919
|
|
|
920
|
+
type ArticleAnswerProps = Readonly<{
|
|
921
|
+
article: Pick<Article, 'answer'>;
|
|
922
|
+
/** Heading shown above the answer. Default: `'The short answer'`. */
|
|
923
|
+
label?: string;
|
|
924
|
+
className?: string;
|
|
925
|
+
}>;
|
|
926
|
+
/**
|
|
927
|
+
* Renders `article.answer` as a callout above the article body.
|
|
928
|
+
*
|
|
929
|
+
* The same text is emitted as the Article schema's `abstract` and placed at
|
|
930
|
+
* the top of the article's markdown twin, so the passage an answer engine is
|
|
931
|
+
* most likely to lift is also the one a reader sees first. Returns `null`
|
|
932
|
+
* when the article has no `answer`, so it is safe to render unconditionally.
|
|
933
|
+
*/
|
|
934
|
+
declare function ArticleAnswer({ article, label, className, }: ArticleAnswerProps): react_jsx_runtime.JSX.Element | null;
|
|
935
|
+
|
|
623
936
|
type ValidationSeverity = 'error' | 'warning';
|
|
624
937
|
interface ValidationIssue {
|
|
625
938
|
severity: ValidationSeverity;
|
|
@@ -636,13 +949,21 @@ interface ValidationResult {
|
|
|
636
949
|
}
|
|
637
950
|
/**
|
|
638
951
|
* Validates a loaded article set + config. Warnings cover optional
|
|
639
|
-
* discovery-field issues (missing excerpt/date, over-length
|
|
640
|
-
* fields, category slug collisions)
|
|
641
|
-
* (
|
|
642
|
-
*
|
|
952
|
+
* discovery-field issues (missing excerpt/date, over-length rendered title
|
|
953
|
+
* and meta description, over-length social fields, category slug collisions) and answer-engine readiness
|
|
954
|
+
* (`no-answer`, `thin-content`, `missing-about`, `unknown-entity`,
|
|
955
|
+
* `stale-content`, `orphan-article`); errors cover broken reader journeys (duplicate canonical
|
|
956
|
+
* URLs, unknown author references, series order collisions, missing/draft
|
|
957
|
+
* path references, unsafe URL schemes).
|
|
958
|
+
*
|
|
959
|
+
* `options.now` overrides the clock used by the `stale-content` check.
|
|
643
960
|
*/
|
|
644
|
-
declare function validateArticles(articles: Article[], config: ArticlesConfig
|
|
961
|
+
declare function validateArticles(articles: Article[], config: ArticlesConfig, options?: Readonly<{
|
|
962
|
+
now?: Date;
|
|
963
|
+
}>): ValidationResult;
|
|
645
964
|
/** Convenience wrapper: loads every article via `getAllArticles(config)` (fs-dependent) then validates. Suitable for a consuming app's own `scripts/validate-articles.ts` invoked in CI before publish. */
|
|
646
|
-
declare function validateAllArticles(config: ArticlesConfig
|
|
965
|
+
declare function validateAllArticles(config: ArticlesConfig, options?: Readonly<{
|
|
966
|
+
now?: Date;
|
|
967
|
+
}>): Promise<ValidationResult>;
|
|
647
968
|
|
|
648
|
-
export { type Article, ArticleContent, type ArticleEvent, type ArticleEventHandler, type ArticleEventName, type ArticleSlotContent, type ArticleSlotContext, ArticleTOC, type ArticleViewedEvent, type ArticlesConfig, type ArticlesErrorCode, type ArticlesErrorContext, type ArticlesErrorHandler, type ArticlesErrorReport, type AuthorClickedEvent, type AuthorProfile, type AuthorSocial, type BreadcrumbItem, type CategoryInfo, type ContentSlotBoundaries, type CtaClickedEvent, type CtaViewedEvent, type LinkTargetStrategy, type ListingPagination, type ListingPaginationContext, type MeaningfulReadEvent, type PaginatedArticles, type PaginationLinks, type PathDefinition, type PathStepAdvancedEvent, type RelatedArticleClickedEvent, type RelatedContentResult, type RelatedContentSource, type SharedEvent, type TocItem, type ValidationIssue, type ValidationResult, type ValidationSeverity, buildArticleBreadcrumbs, buildAuthorBreadcrumbs, buildCategoryBreadcrumbs, buildPageUrl, buildPaginationLinks, categoryToSlug, emitArticleEvent, extractToc, generateArticleMetadata, generateArticleStaticParams, generateArticlesIndexMetadata, generateArticlesIndexPageMetadata, generateAuthorMetadata, generateAuthorPageMetadata, generateAuthorStaticParams, generateCategoryMetadata, generateCategoryPageMetadata, generateCategoryStaticParams, generateListingPageStaticParams, generateRssFeed, generateSeriesMetadata, generateSeriesStaticParams, getAdjacentArticles, getAdjacentArticlesInSeries, getAiRobotsTxtRules, getAllArticles, getAllAuthors, getAllCategories, getArticleAiHeaders, getArticleAuthors, getArticleMarkdown, getArticleMarkdownResponse, getArticleMarkdownUrl, getArticleMetadata, getArticleSitemapEntries, getArticlesByAuthor, getArticlesByCategory, getArticlesBySeries, getAuthorBySlug, getAvailableArticleSlugs, getBreadcrumbsConfig, getContentSlotBoundaries, getPath, getPathArticles, getRelatedArticlesByCategory, getRelatedContent, getTotalPages, isPageOutOfRange, markdownToHtml, paginateArticles, parsePageParam, resolveAuthorAvatar, resolveSearchMetadata, resolveSocialMetadata, sanitizeImagePath, searchArticles, setArticlesErrorHandler, validateAllArticles, validateArticles };
|
|
969
|
+
export { AI_CRAWLERS, type AiCrawlEvent, type Article, ArticleAnswer, ArticleContent, type ArticleEvent, type ArticleEventHandler, type ArticleEventName, type ArticleSlotContent, type ArticleSlotContext, ArticleTOC, type ArticleViewedEvent, type ArticlesConfig, type ArticlesErrorCode, type ArticlesErrorContext, type ArticlesErrorHandler, type ArticlesErrorReport, type AuthorClickedEvent, type AuthorProfile, type AuthorSocial, type BreadcrumbItem, type CategoryInfo, type CitationReference, type ContentSlotBoundaries, type CtaClickedEvent, type CtaViewedEvent, type EntityReference, type FaqItem, type HowToStep, type LinkTargetStrategy, type ListingPagination, type ListingPaginationContext, type MeaningfulReadEvent, type OrganizationConfig, type PaginatedArticles, type PaginationLinks, type PathDefinition, type PathStepAdvancedEvent, type RelatedArticleClickedEvent, type RelatedContentResult, type RelatedContentSource, type RequestHeaders, type SharedEvent, type TocItem, type ValidationIssue, type ValidationResult, type ValidationSeverity, buildArticleBreadcrumbs, buildAuthorBreadcrumbs, buildCategoryBreadcrumbs, buildMarkdownTwinHeader, buildPageUrl, buildPaginationLinks, categoryToSlug, deriveFaqFromHeadings, emitArticleEvent, extractToc, formatPageTitle, generateArticleMetadata, generateArticleStaticParams, generateArticlesIndexMetadata, generateArticlesIndexPageMetadata, generateAuthorMetadata, generateAuthorPageMetadata, generateAuthorStaticParams, generateCategoryMetadata, generateCategoryPageMetadata, generateCategoryStaticParams, generateListingPageStaticParams, generateLlmsFullTxt, generateLlmsTxt, generateRssFeed, generateSeriesMetadata, generateSeriesStaticParams, getAdjacentArticles, getAdjacentArticlesInSeries, getAiRobotsTxtRules, getAllArticles, getAllAuthors, getAllCategories, getArticleAiHeaders, getArticleAuthors, getArticleMarkdown, getArticleMarkdownResponse, getArticleMarkdownUrl, getArticleMetadata, getArticleSitemapEntries, getArticlesByAuthor, getArticlesByCategory, getArticlesBySeries, getAuthorBySlug, getAuthorMarkdown, getAvailableArticleSlugs, getBreadcrumbsConfig, getCategoryMarkdown, getContentSlotBoundaries, getMarkdownTwinResponse, getOrganizationId, getPath, getPathArticles, getPersonId, getRelatedArticlesByCategory, getRelatedContent, getSeriesMarkdown, getTotalPages, getWebSiteId, isPageOutOfRange, markdownToHtml, matchAiCrawler, paginateArticles, parsePageParam, resolveAuthorAvatar, resolveSearchMetadata, resolveSocialMetadata, sanitizeImagePath, searchArticles, setArticlesErrorHandler, validateAllArticles, validateArticles };
|