@growth-labs/cms 0.1.0

package/dist/engine/validator/checks/word-gdocs-paste-artifacts.d.ts

@@ -0,0 +1,23 @@
+/**
+ * `word-gdocs-paste-artifacts`
+ *
+ * Detect Microsoft Word / Google Docs paste residue that the import
+ * pipeline failed to strip. These markers either render as visible junk
+ * (the comment markers) or burn unused class attributes into the DOM
+ * (Mso* class names, mso-* style properties).
+ *
+ * Block-severity: when these appear, the body is not safe to publish
+ * raw — they have triggered measurable layout regressions in the past
+ * (the `Mso*` classes inherit Word's default font stack, which leaks
+ * Times New Roman into the prose font).
+ */
+import type { ArticleBodyInput, ValidationError } from '../types.js';
+export declare function check(input: ArticleBodyInput): ValidationError[];
+/**
+ * Pure repair: drop Mso* class attributes, Start* fragment comments,
+ * and mso-* CSS rules. The repair is conservative — we keep the
+ * surrounding element and strip the offending attribute / declaration
+ * rather than removing the whole node.
+ */
+export declare function repair(value: string): string;
+//# sourceMappingURL=word-gdocs-paste-artifacts.d.ts.map

package/dist/engine/validator/checks/word-gdocs-paste-artifacts.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"word-gdocs-paste-artifacts.d.ts","sourceRoot":"","sources":["../../../../src/engine/validator/checks/word-gdocs-paste-artifacts.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;GAYG;AACH,OAAO,KAAK,EAAE,gBAAgB,EAAE,eAAe,EAAE,MAAM,aAAa,CAAA;AAMpE,wBAAgB,KAAK,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CAmBhE;AAED;;;;;GAKG;AACH,wBAAgB,MAAM,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAyB5C"}

package/dist/engine/validator/checks/word-gdocs-paste-artifacts.js

@@ -0,0 +1,47 @@
+import { makeError } from '../types.js';
+const PATTERN = /\sclass\s*=\s*"?Mso\w+|<!--\s*Start(?:Fragment|HTML|Selection)\s*-->|mso-(?:margin|para|list|ascii|fareast|hansi|bidi)-/i;
+export function check(input) {
+    const errors = [];
+    const checkField = (value, fieldName) => {
+        if (!value)
+            return;
+        const match = PATTERN.exec(value);
+        if (!match)
+            return;
+        errors.push(makeError('word-gdocs-paste-artifacts', `${fieldName} contains Word/Google Docs paste residue (matched '${match[0]}'). Strip Mso* classes and mso-* styles before publishing.`, { start: match.index, end: match.index + match[0].length }));
+    };
+    checkField(input.body_markdown, 'body_markdown');
+    checkField(input.body_html, 'body_html');
+    return errors;
+}
+/**
+ * Pure repair: drop Mso* class attributes, Start* fragment comments,
+ * and mso-* CSS rules. The repair is conservative — we keep the
+ * surrounding element and strip the offending attribute / declaration
+ * rather than removing the whole node.
+ */
+export function repair(value) {
+    return (value
+        // Drop class attributes that contain Mso* tokens.
+        .replace(/\sclass\s*=\s*"([^"]*)"/gi, (_match, classes) => {
+        const filtered = classes
+            .split(/\s+/)
+            .filter((token) => !/^Mso\w+/i.test(token))
+            .join(' ')
+            .trim();
+        return filtered ? ` class="${filtered}"` : '';
+    })
+        .replace(/\sclass\s*=\s*'([^']*)'/gi, (_match, classes) => {
+        const filtered = classes
+            .split(/\s+/)
+            .filter((token) => !/^Mso\w+/i.test(token))
+            .join(' ')
+            .trim();
+        return filtered ? ` class='${filtered}'` : '';
+    })
+        // Drop Start*/End* fragment markers entirely.
+        .replace(/<!--\s*(?:Start|End)(?:Fragment|HTML|Selection)\s*-->/gi, '')
+        // Drop mso-* CSS declarations.
+        .replace(/mso-(?:margin|para|list|ascii|fareast|hansi|bidi)-[^;"'>]*;?/gi, ''));
+}
+//# sourceMappingURL=word-gdocs-paste-artifacts.js.map

package/dist/engine/validator/checks/word-gdocs-paste-artifacts.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"word-gdocs-paste-artifacts.js","sourceRoot":"","sources":["../../../../src/engine/validator/checks/word-gdocs-paste-artifacts.ts"],"names":[],"mappings":"AAcA,OAAO,EAAE,SAAS,EAAE,MAAM,aAAa,CAAA;AAEvC,MAAM,OAAO,GACZ,0HAA0H,CAAA;AAE3H,MAAM,UAAU,KAAK,CAAC,KAAuB;IAC5C,MAAM,MAAM,GAAsB,EAAE,CAAA;IAEpC,MAAM,UAAU,GAAG,CAAC,KAAoB,EAAE,SAAiB,EAAE,EAAE;QAC9D,IAAI,CAAC,KAAK;YAAE,OAAM;QAClB,MAAM,KAAK,GAAG,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;QACjC,IAAI,CAAC,KAAK;YAAE,OAAM;QAClB,MAAM,CAAC,IAAI,CACV,SAAS,CACR,4BAA4B,EAC5B,GAAG,SAAS,sDAAsD,KAAK,CAAC,CAAC,CAAC,4DAA4D,EACtI,EAAE,KAAK,EAAE,KAAK,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAC1D,CACD,CAAA;IACF,CAAC,CAAA;IAED,UAAU,CAAC,KAAK,CAAC,aAAa,EAAE,eAAe,CAAC,CAAA;IAChD,UAAU,CAAC,KAAK,CAAC,SAAS,EAAE,WAAW,CAAC,CAAA;IACxC,OAAO,MAAM,CAAA;AACd,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,MAAM,CAAC,KAAa;IACnC,OAAO,CACN,KAAK;QACJ,kDAAkD;SACjD,OAAO,CAAC,2BAA2B,EAAE,CAAC,MAAM,EAAE,OAAe,EAAE,EAAE;QACjE,MAAM,QAAQ,GAAG,OAAO;aACtB,KAAK,CAAC,KAAK,CAAC;aACZ,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,UAAU,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;aAC1C,IAAI,CAAC,GAAG,CAAC;aACT,IAAI,EAAE,CAAA;QACR,OAAO,QAAQ,CAAC,CAAC,CAAC,WAAW,QAAQ,GAAG,CAAC,CAAC,CAAC,EAAE,CAAA;IAC9C,CAAC,CAAC;SACD,OAAO,CAAC,2BAA2B,EAAE,CAAC,MAAM,EAAE,OAAe,EAAE,EAAE;QACjE,MAAM,QAAQ,GAAG,OAAO;aACtB,KAAK,CAAC,KAAK,CAAC;aACZ,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,UAAU,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;aAC1C,IAAI,CAAC,GAAG,CAAC;aACT,IAAI,EAAE,CAAA;QACR,OAAO,QAAQ,CAAC,CAAC,CAAC,WAAW,QAAQ,GAAG,CAAC,CAAC,CAAC,EAAE,CAAA;IAC9C,CAAC,CAAC;QACF,8CAA8C;SAC7C,OAAO,CAAC,yDAAyD,EAAE,EAAE,CAAC;QACvE,+BAA+B;SAC9B,OAAO,CAAC,gEAAgE,EAAE,EAAE,CAAC,CAC/E,CAAA;AACF,CAAC"}

package/dist/engine/validator/index.d.ts

@@ -0,0 +1,75 @@
+/**
+ * article-body-validator (modular layout)
+ *
+ * The validator is the single source of truth for ingest-time formatting
+ * QA. Each `checks/<code>.ts` file implements one independent detector;
+ * this index module composes them into the public API used by the
+ * publish guard and the publishing engine.
+ *
+ * Ported verbatim from fulcrum-labs/fronts
+ * src/lib/article-body-validator/index.ts (inventory §1: extract). The
+ * validator runs on PRE-strip source; renderer-side cleanup happens AFTER
+ * the validator has emitted its report so ingest bugs stay visible.
+ */
+import * as bareUrl from './checks/bare-url-not-autolinked.js';
+import * as brokenFootnoteLabel from './checks/broken-footnote-label.js';
+import * as doubleEncoded from './checks/double-encoded-entities.js';
+import * as emptyAlt from './checks/empty-alt-text.js';
+import * as headingHierarchy from './checks/heading-hierarchy-skip.js';
+import * as htmlCommentLeak from './checks/html-comment-leak.js';
+import * as iframeMissingDims from './checks/iframe-missing-dims-and-wrapper.js';
+import * as invisibleControl from './checks/invisible-control-chars.js';
+import * as paywallMarker from './checks/paywall-marker-leak.js';
+import * as rawBlockHtml from './checks/raw-block-html.js';
+import * as unresolvedAnchor from './checks/unresolved-footnote-anchor.js';
+import * as wordGdocs from './checks/word-gdocs-paste-artifacts.js';
+import type { ArticleBodyInput, Severity, ValidationError, ValidationErrorCode, ValidationResult, ValidatorOptions } from './types.js';
+import { DEFAULT_OPTIONS, SEVERITY } from './types.js';
+/**
+ * Astro 500 error template fingerprints. The build-time guard looks for
+ * the same set, so the static-output check and the pre-publish check
+ * agree on what a baked server-error page looks like.
+ */
+export declare const ASTRO_SERVER_ERROR_PATTERNS: readonly RegExp[];
+export declare function checkDataUriArtifact(input: ArticleBodyInput): ValidationError[];
+export declare function checkBakedAstroErrorTemplate(input: ArticleBodyInput): ValidationError[];
+export declare function checkEscapedRawHtml(input: ArticleBodyInput): ValidationError[];
+export declare function checkMalformedDataImageMarkdown(input: ArticleBodyInput): ValidationError[];
+export declare function checkUnbalancedMarkdownEmphasis(input: ArticleBodyInput): ValidationError[];
+export declare function checkSingleSentenceParagraphsPattern(input: ArticleBodyInput, options?: ValidatorOptions): ValidationError[];
+export declare function checkBackslashOverEscape(input: ArticleBodyInput): ValidationError[];
+export declare function checkHtmlEntityInHeading(input: ArticleBodyInput): ValidationError[];
+export declare function checkEmptyBody(input: ArticleBodyInput): ValidationError[];
+export declare const checkBrokenFootnoteLabel: typeof brokenFootnoteLabel.check;
+export declare const checkUnresolvedFootnoteAnchor: typeof unresolvedAnchor.check;
+export declare const checkWordGdocsPasteArtifacts: typeof wordGdocs.check;
+export declare const checkHtmlCommentLeak: typeof htmlCommentLeak.check;
+export declare const checkRawBlockHtml: typeof rawBlockHtml.check;
+export declare const checkDoubleEncodedEntities: typeof doubleEncoded.check;
+export declare const checkInvisibleControlChars: typeof invisibleControl.check;
+export declare const checkBareUrlNotAutolinked: typeof bareUrl.check;
+export declare const checkEmptyAltText: typeof emptyAlt.check;
+export declare const checkIframeMissingDimsAndWrapper: typeof iframeMissingDims.check;
+export declare const checkHeadingHierarchySkip: typeof headingHierarchy.check;
+export declare const checkPaywallMarkerLeak: typeof paywallMarker.check;
+export interface RunOptions extends ValidatorOptions {
+    requireBody: boolean;
+}
+export declare const DEFAULT_RUN_OPTIONS: RunOptions;
+/**
+ * Run every check in a deterministic order and aggregate the errors.
+ * Structural problems (empty body, baked-error template) come first;
+ * the body-string checks follow; the noisiest heuristic checks
+ * (paragraph-split pattern) come last.
+ */
+export declare function validateArticleBody(input: ArticleBodyInput, options?: Partial<RunOptions>): ValidationResult;
+/**
+ * Filter a list of errors down to those with `block` severity. Callers
+ * use this to decide between 422 (block) and 200 + warnings (warn-only).
+ */
+export declare function blockErrors(errors: ValidationError[]): ValidationError[];
+export declare function warnErrors(errors: ValidationError[]): ValidationError[];
+export declare function formatValidationErrors(errors: ValidationError[]): string;
+export type { ArticleBodyInput, Severity, ValidationError, ValidationErrorCode, ValidationResult, ValidatorOptions, };
+export { DEFAULT_OPTIONS, SEVERITY };
+//# sourceMappingURL=index.d.ts.map

package/dist/engine/validator/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/engine/validator/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;GAYG;AACH,OAAO,KAAK,OAAO,MAAM,qCAAqC,CAAA;AAC9D,OAAO,KAAK,mBAAmB,MAAM,mCAAmC,CAAA;AACxE,OAAO,KAAK,aAAa,MAAM,qCAAqC,CAAA;AACpE,OAAO,KAAK,QAAQ,MAAM,4BAA4B,CAAA;AACtD,OAAO,KAAK,gBAAgB,MAAM,oCAAoC,CAAA;AACtE,OAAO,KAAK,eAAe,MAAM,+BAA+B,CAAA;AAChE,OAAO,KAAK,iBAAiB,MAAM,6CAA6C,CAAA;AAChF,OAAO,KAAK,gBAAgB,MAAM,qCAAqC,CAAA;AACvE,OAAO,KAAK,aAAa,MAAM,iCAAiC,CAAA;AAChE,OAAO,KAAK,YAAY,MAAM,4BAA4B,CAAA;AAC1D,OAAO,KAAK,gBAAgB,MAAM,wCAAwC,CAAA;AAC1E,OAAO,KAAK,SAAS,MAAM,wCAAwC,CAAA;AAEnE,OAAO,KAAK,EACX,gBAAgB,EAChB,QAAQ,EACR,eAAe,EACf,mBAAmB,EACnB,gBAAgB,EAChB,gBAAgB,EAChB,MAAM,YAAY,CAAA;AACnB,OAAO,EAAE,eAAe,EAAa,QAAQ,EAAE,MAAM,YAAY,CAAA;AAMjE;;;;GAIG;AACH,eAAO,MAAM,2BAA2B,EAAE,SAAS,MAAM,EAKxD,CAAA;AAQD,wBAAgB,oBAAoB,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CAkB/E;AAED,wBAAgB,4BAA4B,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CAoBvF;AAED,wBAAgB,mBAAmB,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CAmB9E;AAED,wBAAgB,+BAA+B,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CAY1F;AAED,wBAAgB,+BAA+B,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CAoC1F;AAED,wBAAgB,oCAAoC,CACnD,KAAK,EAAE,gBAAgB,EACvB,OAAO,GAAE,gBAAkC,GACzC,eAAe,EAAE,CA4BnB;AAuBD,wBAAgB,wBAAwB,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CAgBnF;AAID,wBAAgB,wBAAwB,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CA4BnF;AAED,wBAAgB,cAAc,CAAC,KAAK,EAAE,gBAAgB,GAAG,eAAe,EAAE,CAUzE;AAMD,eAAO,MAAM,wBAAwB,kCAA4B,CAAA;AACjE,eAAO,MAAM,6BAA6B,+BAAyB,CAAA;AACnE,eAAO,MAAM,4BAA4B,wBAAkB,CAAA;AAC3D,eAAO,MAAM,oBAAoB,8BAAwB,CAAA;AACzD,eAAO,MAAM,iBAAiB,2BAAqB,CAAA;AACnD,eAAO,MAAM,0BAA0B,4BAAsB,CAAA;AAC7D,eAAO,MAAM,0BAA0B,+BAAyB,CAAA;AAChE,eAAO,MAAM,yBAAyB,sBAAgB,CAAA;AACtD,eAAO,MAAM,iBAAiB,uBAAiB,CAAA;AAC/C,eAAO,MAAM,gCAAgC,gCAA0B,CAAA;AACvE,eAAO,MAAM,yBAAyB,+BAAyB,CAAA;AAC/D,eAAO,MAAM,sBAAsB,4BAAsB,CAAA;AAMzD,MAAM,WAAW,UAAW,SAAQ,gBAAgB;IACnD,WAAW,EAAE,OAAO,CAAA;CACpB;AAED,eAAO,MAAM,mBAAmB,EAAE,UAGjC,CAAA;AAED;;;;;GAKG;AACH,wBAAgB,mBAAmB,CAClC,KAAK,EAAE,gBAAgB,EACvB,OAAO,GAAE,OAAO,CAAC,UAAU,CAAM,GAC/B,gBAAgB,CA6BlB;AAED;;;GAGG;AACH,wBAAgB,WAAW,CAAC,MAAM,EAAE,eAAe,EAAE,GAAG,eAAe,EAAE,CAExE;AAED,wBAAgB,UAAU,CAAC,MAAM,EAAE,eAAe,EAAE,GAAG,eAAe,EAAE,CAEvE;AAED,wBAAgB,sBAAsB,CAAC,MAAM,EAAE,eAAe,EAAE,GAAG,MAAM,CAGxE;AAED,YAAY,EACX,gBAAgB,EAChB,QAAQ,EACR,eAAe,EACf,mBAAmB,EACnB,gBAAgB,EAChB,gBAAgB,GAChB,CAAA;AACD,OAAO,EAAE,eAAe,EAAE,QAAQ,EAAE,CAAA"}

package/dist/engine/validator/index.js

@@ -0,0 +1,313 @@
+/**
+ * article-body-validator (modular layout)
+ *
+ * The validator is the single source of truth for ingest-time formatting
+ * QA. Each `checks/<code>.ts` file implements one independent detector;
+ * this index module composes them into the public API used by the
+ * publish guard and the publishing engine.
+ *
+ * Ported verbatim from fulcrum-labs/fronts
+ * src/lib/article-body-validator/index.ts (inventory §1: extract). The
+ * validator runs on PRE-strip source; renderer-side cleanup happens AFTER
+ * the validator has emitted its report so ingest bugs stay visible.
+ */
+import * as bareUrl from './checks/bare-url-not-autolinked.js';
+import * as brokenFootnoteLabel from './checks/broken-footnote-label.js';
+import * as doubleEncoded from './checks/double-encoded-entities.js';
+import * as emptyAlt from './checks/empty-alt-text.js';
+import * as headingHierarchy from './checks/heading-hierarchy-skip.js';
+import * as htmlCommentLeak from './checks/html-comment-leak.js';
+import * as iframeMissingDims from './checks/iframe-missing-dims-and-wrapper.js';
+import * as invisibleControl from './checks/invisible-control-chars.js';
+import * as paywallMarker from './checks/paywall-marker-leak.js';
+import * as rawBlockHtml from './checks/raw-block-html.js';
+import * as unresolvedAnchor from './checks/unresolved-footnote-anchor.js';
+import * as wordGdocs from './checks/word-gdocs-paste-artifacts.js';
+import { DEFAULT_OPTIONS, makeError, SEVERITY } from './types.js';
+/* -------------------------------------------------------------------------- */
+/*                     Phase 0 / Phase 1 — preserved checks                   */
+/* -------------------------------------------------------------------------- */
+/**
+ * Astro 500 error template fingerprints. The build-time guard looks for
+ * the same set, so the static-output check and the pre-publish check
+ * agree on what a baked server-error page looks like.
+ */
+export const ASTRO_SERVER_ERROR_PATTERNS = [
+    /<title>\s*Server Error(?:\s*\|\s*Fronts)?\s*<\/title>/i,
+    /Fronts encountered an unexpected error\./i,
+    /The request reached Fronts but the response failed before rendering completed\./i,
+    /Something broke upstream\./i,
+];
+const LEGIT_ESCAPE_CHARS = '*_[]{}|`#+!>\\-';
+function isBlank(value) {
+    return !value || value.trim().length === 0;
+}
+export function checkDataUriArtifact(input) {
+    const errors = [];
+    const pattern = /data:image\/[a-z0-9+.-]+;base64,/i;
+    const checkField = (value, fieldName) => {
+        if (!value)
+            return;
+        const match = pattern.exec(value);
+        if (!match)
+            return;
+        errors.push(makeError('data-uri-artifact', `Article body contains an inline data:image base64 payload in ${fieldName}; bodies must reference R2 URLs only.`, { start: match.index, end: match.index + match[0].length }));
+    };
+    checkField(input.body_markdown, 'body_markdown');
+    checkField(input.body_html, 'body_html');
+    return errors;
+}
+export function checkBakedAstroErrorTemplate(input) {
+    const errors = [];
+    const checkField = (value, fieldName) => {
+        if (!value)
+            return;
+        for (const pattern of ASTRO_SERVER_ERROR_PATTERNS) {
+            const match = pattern.exec(value);
+            if (!match)
+                continue;
+            errors.push(makeError('baked-astro-error-template', `Article body in ${fieldName} matches a baked Astro server-error template fingerprint.`, { start: match.index, end: match.index + match[0].length }));
+            return;
+        }
+    };
+    checkField(input.body_markdown, 'body_markdown');
+    checkField(input.body_html, 'body_html');
+    return errors;
+}
+export function checkEscapedRawHtml(input) {
+    const errors = [];
+    if (!input.body_markdown)
+        return errors;
+    const patterns = [
+        { tag: 'iframe', re: /&lt;iframe[\s>]/i },
+        { tag: 'img', re: /&lt;img[\s>]/i },
+    ];
+    for (const { tag, re } of patterns) {
+        const match = re.exec(input.body_markdown);
+        if (!match)
+            continue;
+        errors.push(makeError('escaped-raw-html', `body_markdown contains escaped raw <${tag}> markup — this will render as literal text. Decode the entities before saving.`, { start: match.index, end: match.index + match[0].length }));
+    }
+    return errors;
+}
+export function checkMalformedDataImageMarkdown(input) {
+    if (!input.body_markdown)
+        return [];
+    const pattern = /\[!\[[^\]]*\]\(data:image\/[a-z0-9+.-]+;base64,/i;
+    const match = pattern.exec(input.body_markdown);
+    if (!match)
+        return [];
+    return [
+        makeError('malformed-data-image-markdown', 'body_markdown contains the [![](data:image/...)] artifact from a broken import. Replace the inline payload with an R2 URL.', { start: match.index, end: match.index + match[0].length }),
+    ];
+}
+export function checkUnbalancedMarkdownEmphasis(input) {
+    if (!input.body_markdown)
+        return [];
+    const errors = [];
+    const lines = input.body_markdown.split('\n');
+    let lineStart = 0;
+    for (const rawLine of lines) {
+        const len = rawLine.length + 1;
+        const lineStartHere = lineStart;
+        lineStart += len;
+        const stripped = rawLine.replace(/`[^`]*`/g, '');
+        const noStrong = stripped.replace(/\*\*/g, '').replace(/__/g, '');
+        const noEscaped = noStrong.replace(/\\[*_]/g, '');
+        const stars = (noEscaped.match(/\*/g) || []).length;
+        const unders = (noEscaped.match(/(?<=\w)_(?=\w)|(?<!\w)_|_(?!\w)/g) || []).length;
+        if (stars % 2 === 1) {
+            const idx = noEscaped.indexOf('*');
+            errors.push(makeError('unbalanced-markdown-emphasis', "Line has an unbalanced '*' emphasis marker.", {
+                start: lineStartHere + (idx >= 0 ? idx : 0),
+                end: lineStartHere + rawLine.length,
+            }));
+            break;
+        }
+        if (unders % 2 === 1) {
+            const idx = noEscaped.indexOf('_');
+            errors.push(makeError('unbalanced-markdown-emphasis', "Line has an unbalanced '_' emphasis marker.", {
+                start: lineStartHere + (idx >= 0 ? idx : 0),
+                end: lineStartHere + rawLine.length,
+            }));
+            break;
+        }
+    }
+    return errors;
+}
+export function checkSingleSentenceParagraphsPattern(input, options = DEFAULT_OPTIONS) {
+    if (!input.body_markdown)
+        return [];
+    const paragraphs = input.body_markdown
+        .split(/\n{2,}/)
+        .map((p) => p.trim())
+        .filter((p) => p.length > 0);
+    if (paragraphs.length < options.singleSentenceParagraphRun)
+        return [];
+    let run = 0;
+    let runStartIndex = -1;
+    for (let i = 0; i < paragraphs.length; i += 1) {
+        const para = paragraphs[i];
+        if (isLikelyProseParagraph(para) && countSentences(para) <= 1) {
+            if (run === 0)
+                runStartIndex = i;
+            run += 1;
+            if (run >= options.singleSentenceParagraphRun) {
+                return [
+                    makeError('single-sentence-paragraphs-pattern', `Found ${run}+ consecutive single-sentence paragraphs starting at paragraph ${runStartIndex + 1}; this usually indicates an over-split import.`),
+                ];
+            }
+        }
+        else {
+            run = 0;
+            runStartIndex = -1;
+        }
+    }
+    return [];
+}
+function isLikelyProseParagraph(paragraph) {
+    const trimmed = paragraph.trim();
+    if (trimmed.startsWith('#'))
+        return false;
+    if (trimmed.startsWith('>'))
+        return false;
+    if (trimmed.startsWith('- ') || /^\d+\.\s/.test(trimmed))
+        return false;
+    if (trimmed.startsWith('```'))
+        return false;
+    if (trimmed.startsWith('<') && trimmed.endsWith('>'))
+        return false;
+    if (trimmed.startsWith('![') || trimmed.startsWith('['))
+        return false;
+    if (trimmed.length < 20)
+        return false;
+    return true;
+}
+function countSentences(paragraph) {
+    const stripped = paragraph
+        .replace(/`[^`]*`/g, '')
+        .replace(/\*\*([^*]+)\*\*/g, '$1')
+        .replace(/\*([^*]+)\*/g, '$1');
+    const matches = stripped.match(/[.!?](?:\s+|$)/g);
+    return matches ? matches.length : 0;
+}
+export function checkBackslashOverEscape(input) {
+    if (!input.body_markdown)
+        return [];
+    const pattern = /\\([^\s\\])/g;
+    for (const match of input.body_markdown.matchAll(pattern)) {
+        const ch = match[1] ?? '';
+        if (LEGIT_ESCAPE_CHARS.includes(ch))
+            continue;
+        const start = match.index ?? 0;
+        return [
+            makeError('backslash-over-escape', `body_markdown contains backslash-over-escape (e.g. '\\${ch}') from a broken import. Strip the legacy escapes before publishing.`, { start, end: start + match[0].length }),
+        ];
+    }
+    return [];
+}
+const HEADING_ENTITY_PATTERN = /&(?:quot|apos|amp|lt|gt|#\d+);/i;
+export function checkHtmlEntityInHeading(input) {
+    if (!input.body_markdown)
+        return [];
+    const lines = input.body_markdown.split('\n');
+    let cursor = 0;
+    for (const rawLine of lines) {
+        const lineLen = rawLine.length + 1;
+        const trimmed = rawLine.trimStart();
+        if (/^#{1,6}\s+/.test(trimmed)) {
+            const headingText = trimmed.replace(/^#{1,6}\s+/, '');
+            const match = HEADING_ENTITY_PATTERN.exec(headingText);
+            if (match) {
+                const entityStart = cursor +
+                    (rawLine.length - trimmed.length) +
+                    (trimmed.length - headingText.length) +
+                    (match.index ?? 0);
+                return [
+                    makeError('html-entity-in-heading', `Heading contains an HTML entity ('${match[0]}'); headings must be raw characters or the TOC will display the entity code.`, { start: entityStart, end: entityStart + match[0].length }),
+                ];
+            }
+        }
+        cursor += lineLen;
+    }
+    return [];
+}
+export function checkEmptyBody(input) {
+    if (isBlank(input.body_markdown) && isBlank(input.body_html)) {
+        return [
+            makeError('empty-body', 'Both body_markdown and body_html are empty — refusing to publish an article with no content.'),
+        ];
+    }
+    return [];
+}
+/* -------------------------------------------------------------------------- */
+/*                       Phase 2 — re-exported check fns                      */
+/* -------------------------------------------------------------------------- */
+export const checkBrokenFootnoteLabel = brokenFootnoteLabel.check;
+export const checkUnresolvedFootnoteAnchor = unresolvedAnchor.check;
+export const checkWordGdocsPasteArtifacts = wordGdocs.check;
+export const checkHtmlCommentLeak = htmlCommentLeak.check;
+export const checkRawBlockHtml = rawBlockHtml.check;
+export const checkDoubleEncodedEntities = doubleEncoded.check;
+export const checkInvisibleControlChars = invisibleControl.check;
+export const checkBareUrlNotAutolinked = bareUrl.check;
+export const checkEmptyAltText = emptyAlt.check;
+export const checkIframeMissingDimsAndWrapper = iframeMissingDims.check;
+export const checkHeadingHierarchySkip = headingHierarchy.check;
+export const checkPaywallMarkerLeak = paywallMarker.check;
+export const DEFAULT_RUN_OPTIONS = {
+    ...DEFAULT_OPTIONS,
+    requireBody: true,
+};
+/**
+ * Run every check in a deterministic order and aggregate the errors.
+ * Structural problems (empty body, baked-error template) come first;
+ * the body-string checks follow; the noisiest heuristic checks
+ * (paragraph-split pattern) come last.
+ */
+export function validateArticleBody(input, options = {}) {
+    const opts = { ...DEFAULT_RUN_OPTIONS, ...options };
+    const errors = [];
+    if (opts.requireBody)
+        errors.push(...checkEmptyBody(input));
+    errors.push(...checkBakedAstroErrorTemplate(input));
+    errors.push(...checkMalformedDataImageMarkdown(input));
+    errors.push(...checkDataUriArtifact(input));
+    errors.push(...checkEscapedRawHtml(input));
+    errors.push(...checkHtmlEntityInHeading(input));
+    errors.push(...checkBackslashOverEscape(input));
+    // Phase 2.
+    errors.push(...checkBrokenFootnoteLabel(input));
+    errors.push(...checkUnresolvedFootnoteAnchor(input));
+    errors.push(...checkWordGdocsPasteArtifacts(input));
+    errors.push(...checkHtmlCommentLeak(input));
+    errors.push(...checkRawBlockHtml(input));
+    errors.push(...checkDoubleEncodedEntities(input));
+    errors.push(...checkInvisibleControlChars(input));
+    errors.push(...checkBareUrlNotAutolinked(input));
+    errors.push(...checkEmptyAltText(input));
+    errors.push(...checkIframeMissingDimsAndWrapper(input));
+    errors.push(...checkHeadingHierarchySkip(input));
+    errors.push(...checkPaywallMarkerLeak(input));
+    // Heuristic.
+    errors.push(...checkUnbalancedMarkdownEmphasis(input));
+    errors.push(...checkSingleSentenceParagraphsPattern(input, opts));
+    return errors.length === 0 ? { ok: true } : { ok: false, errors };
+}
+/**
+ * Filter a list of errors down to those with `block` severity. Callers
+ * use this to decide between 422 (block) and 200 + warnings (warn-only).
+ */
+export function blockErrors(errors) {
+    return errors.filter((e) => SEVERITY[e.code] === 'block');
+}
+export function warnErrors(errors) {
+    return errors.filter((e) => SEVERITY[e.code] === 'warn');
+}
+export function formatValidationErrors(errors) {
+    if (errors.length === 0)
+        return 'ok';
+    return errors.map((e) => `[${e.code}] ${e.message}`).join('\n');
+}
+export { DEFAULT_OPTIONS, SEVERITY };
+//# sourceMappingURL=index.js.map

package/dist/engine/validator/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../src/engine/validator/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;GAYG;AACH,OAAO,KAAK,OAAO,MAAM,qCAAqC,CAAA;AAC9D,OAAO,KAAK,mBAAmB,MAAM,mCAAmC,CAAA;AACxE,OAAO,KAAK,aAAa,MAAM,qCAAqC,CAAA;AACpE,OAAO,KAAK,QAAQ,MAAM,4BAA4B,CAAA;AACtD,OAAO,KAAK,gBAAgB,MAAM,oCAAoC,CAAA;AACtE,OAAO,KAAK,eAAe,MAAM,+BAA+B,CAAA;AAChE,OAAO,KAAK,iBAAiB,MAAM,6CAA6C,CAAA;AAChF,OAAO,KAAK,gBAAgB,MAAM,qCAAqC,CAAA;AACvE,OAAO,KAAK,aAAa,MAAM,iCAAiC,CAAA;AAChE,OAAO,KAAK,YAAY,MAAM,4BAA4B,CAAA;AAC1D,OAAO,KAAK,gBAAgB,MAAM,wCAAwC,CAAA;AAC1E,OAAO,KAAK,SAAS,MAAM,wCAAwC,CAAA;AAUnE,OAAO,EAAE,eAAe,EAAE,SAAS,EAAE,QAAQ,EAAE,MAAM,YAAY,CAAA;AAEjE,gFAAgF;AAChF,gFAAgF;AAChF,gFAAgF;AAEhF;;;;GAIG;AACH,MAAM,CAAC,MAAM,2BAA2B,GAAsB;IAC7D,wDAAwD;IACxD,2CAA2C;IAC3C,kFAAkF;IAClF,6BAA6B;CAC7B,CAAA;AAED,MAAM,kBAAkB,GAAG,iBAAiB,CAAA;AAE5C,SAAS,OAAO,CAAC,KAAoB;IACpC,OAAO,CAAC,KAAK,IAAI,KAAK,CAAC,IAAI,EAAE,CAAC,MAAM,KAAK,CAAC,CAAA;AAC3C,CAAC;AAED,MAAM,UAAU,oBAAoB,CAAC,KAAuB;IAC3D,MAAM,MAAM,GAAsB,EAAE,CAAA;IACpC,MAAM,OAAO,GAAG,mCAAmC,CAAA;IACnD,MAAM,UAAU,GAAG,CAAC,KAAoB,EAAE,SAAiB,EAAE,EAAE;QAC9D,IAAI,CAAC,KAAK;YAAE,OAAM;QAClB,MAAM,KAAK,GAAG,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;QACjC,IAAI,CAAC,KAAK;YAAE,OAAM;QAClB,MAAM,CAAC,IAAI,CACV,SAAS,CACR,mBAAmB,EACnB,gEAAgE,SAAS,uCAAuC,EAChH,EAAE,KAAK,EAAE,KAAK,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAC1D,CACD,CAAA;IACF,CAAC,CAAA;IACD,UAAU,CAAC,KAAK,CAAC,aAAa,EAAE,eAAe,CAAC,CAAA;IAChD,UAAU,CAAC,KAAK,CAAC,SAAS,EAAE,WAAW,CAAC,CAAA;IACxC,OAAO,MAAM,CAAA;AACd,CAAC;AAED,MAAM,UAAU,4BAA4B,CAAC,KAAuB;IACnE,MAAM,MAAM,GAAsB,EAAE,CAAA;IACpC,MAAM,UAAU,GAAG,CAAC,KAAoB,EAAE,SAAiB,EAAE,EAAE;QAC9D,IAAI,CAAC,KAAK;YAAE,OAAM;QAClB,KAAK,MAAM,OAAO,IAAI,2BAA2B,EAAE,CAAC;YACnD,MAAM,KAAK,GAAG,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACjC,IAAI,CAAC,KAAK;gBAAE,SAAQ;YACpB,MAAM,CAAC,IAAI,CACV,SAAS,CACR,4BAA4B,EAC5B,mBAAmB,SAAS,2DAA2D,EACvF,EAAE,KAAK,EAAE,KAAK,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAC1D,CACD,CAAA;YACD,OAAM;QACP,CAAC;IACF,CAAC,CAAA;IACD,UAAU,CAAC,KAAK,CAAC,aAAa,EAAE,eAAe,CAAC,CAAA;IAChD,UAAU,CAAC,KAAK,CAAC,SAAS,EAAE,WAAW,CAAC,CAAA;IACxC,OAAO,MAAM,CAAA;AACd,CAAC;AAED,MAAM,UAAU,mBAAmB,CAAC,KAAuB;IAC1D,MAAM,MAAM,GAAsB,EAAE,CAAA;IACpC,IAAI,CAAC,KAAK,CAAC,aAAa;QAAE,OAAO,MAAM,CAAA;IACvC,MAAM,QAAQ,GAAuC;QACpD,EAAE,GAAG,EAAE,QAAQ,EAAE,EAAE,EAAE,kBAAkB,EAAE;QACzC,EAAE,GAAG,EAAE,KAAK,EAAE,EAAE,EAAE,eAAe,EAAE;KACnC,CAAA;IACD,KAAK,MAAM,EAAE,GAAG,EAAE,EAAE,EAAE,IAAI,QAAQ,EAAE,CAAC;QACpC,MAAM,KAAK,GAAG,EAAE,CAAC,IAAI,CAAC,KAAK,CAAC,aAAa,CAAC,CAAA;QAC1C,IAAI,CAAC,KAAK;YAAE,SAAQ;QACpB,MAAM,CAAC,IAAI,CACV,SAAS,CACR,kBAAkB,EAClB,uCAAuC,GAAG,iFAAiF,EAC3H,EAAE,KAAK,EAAE,KAAK,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAC1D,CACD,CAAA;IACF,CAAC;IACD,OAAO,MAAM,CAAA;AACd,CAAC;AAED,MAAM,UAAU,+BAA+B,CAAC,KAAuB;IACtE,IAAI,CAAC,KAAK,CAAC,aAAa;QAAE,OAAO,EAAE,CAAA;IACnC,MAAM,OAAO,GAAG,kDAAkD,CAAA;IAClE,MAAM,KAAK,GAAG,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,aAAa,CAAC,CAAA;IAC/C,IAAI,CAAC,KAAK;QAAE,OAAO,EAAE,CAAA;IACrB,OAAO;QACN,SAAS,CACR,+BAA+B,EAC/B,4HAA4H,EAC5H,EAAE,KAAK,EAAE,KAAK,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAC1D;KACD,CAAA;AACF,CAAC;AAED,MAAM,UAAU,+BAA+B,CAAC,KAAuB;IACtE,IAAI,CAAC,KAAK,CAAC,aAAa;QAAE,OAAO,EAAE,CAAA;IACnC,MAAM,MAAM,GAAsB,EAAE,CAAA;IACpC,MAAM,KAAK,GAAG,KAAK,CAAC,aAAa,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IAC7C,IAAI,SAAS,GAAG,CAAC,CAAA;IACjB,KAAK,MAAM,OAAO,IAAI,KAAK,EAAE,CAAC;QAC7B,MAAM,GAAG,GAAG,OAAO,CAAC,MAAM,GAAG,CAAC,CAAA;QAC9B,MAAM,aAAa,GAAG,SAAS,CAAA;QAC/B,SAAS,IAAI,GAAG,CAAA;QAChB,MAAM,QAAQ,GAAG,OAAO,CAAC,OAAO,CAAC,UAAU,EAAE,EAAE,CAAC,CAAA;QAChD,MAAM,QAAQ,GAAG,QAAQ,CAAC,OAAO,CAAC,OAAO,EAAE,EAAE,CAAC,CAAC,OAAO,CAAC,KAAK,EAAE,EAAE,CAAC,CAAA;QACjE,MAAM,SAAS,GAAG,QAAQ,CAAC,OAAO,CAAC,SAAS,EAAE,EAAE,CAAC,CAAA;QACjD,MAAM,KAAK,GAAG,CAAC,SAAS,CAAC,KAAK,CAAC,KAAK,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAA;QACnD,MAAM,MAAM,GAAG,CAAC,SAAS,CAAC,KAAK,CAAC,kCAAkC,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAA;QACjF,IAAI,KAAK,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC;YACrB,MAAM,GAAG,GAAG,SAAS,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YAClC,MAAM,CAAC,IAAI,CACV,SAAS,CAAC,8BAA8B,EAAE,6CAA6C,EAAE;gBACxF,KAAK,EAAE,aAAa,GAAG,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;gBAC3C,GAAG,EAAE,aAAa,GAAG,OAAO,CAAC,MAAM;aACnC,CAAC,CACF,CAAA;YACD,MAAK;QACN,CAAC;QACD,IAAI,MAAM,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC;YACtB,MAAM,GAAG,GAAG,SAAS,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YAClC,MAAM,CAAC,IAAI,CACV,SAAS,CAAC,8BAA8B,EAAE,6CAA6C,EAAE;gBACxF,KAAK,EAAE,aAAa,GAAG,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;gBAC3C,GAAG,EAAE,aAAa,GAAG,OAAO,CAAC,MAAM;aACnC,CAAC,CACF,CAAA;YACD,MAAK;QACN,CAAC;IACF,CAAC;IACD,OAAO,MAAM,CAAA;AACd,CAAC;AAED,MAAM,UAAU,oCAAoC,CACnD,KAAuB,EACvB,UAA4B,eAAe;IAE3C,IAAI,CAAC,KAAK,CAAC,aAAa;QAAE,OAAO,EAAE,CAAA;IACnC,MAAM,UAAU,GAAG,KAAK,CAAC,aAAa;SACpC,KAAK,CAAC,QAAQ,CAAC;SACf,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;SACpB,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC,CAAA;IAC7B,IAAI,UAAU,CAAC,MAAM,GAAG,OAAO,CAAC,0BAA0B;QAAE,OAAO,EAAE,CAAA;IACrE,IAAI,GAAG,GAAG,CAAC,CAAA;IACX,IAAI,aAAa,GAAG,CAAC,CAAC,CAAA;IACtB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,MAAM,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC;QAC/C,MAAM,IAAI,GAAG,UAAU,CAAC,CAAC,CAAC,CAAA;QAC1B,IAAI,sBAAsB,CAAC,IAAI,CAAC,IAAI,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;YAC/D,IAAI,GAAG,KAAK,CAAC;gBAAE,aAAa,GAAG,CAAC,CAAA;YAChC,GAAG,IAAI,CAAC,CAAA;YACR,IAAI,GAAG,IAAI,OAAO,CAAC,0BAA0B,EAAE,CAAC;gBAC/C,OAAO;oBACN,SAAS,CACR,oCAAoC,EACpC,SAAS,GAAG,kEAAkE,aAAa,GAAG,CAAC,gDAAgD,CAC/I;iBACD,CAAA;YACF,CAAC;QACF,CAAC;aAAM,CAAC;YACP,GAAG,GAAG,CAAC,CAAA;YACP,aAAa,GAAG,CAAC,CAAC,CAAA;QACnB,CAAC;IACF,CAAC;IACD,OAAO,EAAE,CAAA;AACV,CAAC;AAED,SAAS,sBAAsB,CAAC,SAAiB;IAChD,MAAM,OAAO,GAAG,SAAS,CAAC,IAAI,EAAE,CAAA;IAChC,IAAI,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC;QAAE,OAAO,KAAK,CAAA;IACzC,IAAI,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC;QAAE,OAAO,KAAK,CAAA;IACzC,IAAI,OAAO,CAAC,UAAU,CAAC,IAAI,CAAC,IAAI,UAAU,CAAC,IAAI,CAAC,OAAO,CAAC;QAAE,OAAO,KAAK,CAAA;IACtE,IAAI,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC;QAAE,OAAO,KAAK,CAAA;IAC3C,IAAI,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC,IAAI,OAAO,CAAC,QAAQ,CAAC,GAAG,CAAC;QAAE,OAAO,KAAK,CAAA;IAClE,IAAI,OAAO,CAAC,UAAU,CAAC,IAAI,CAAC,IAAI,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC;QAAE,OAAO,KAAK,CAAA;IACrE,IAAI,OAAO,CAAC,MAAM,GAAG,EAAE;QAAE,OAAO,KAAK,CAAA;IACrC,OAAO,IAAI,CAAA;AACZ,CAAC;AAED,SAAS,cAAc,CAAC,SAAiB;IACxC,MAAM,QAAQ,GAAG,SAAS;SACxB,OAAO,CAAC,UAAU,EAAE,EAAE,CAAC;SACvB,OAAO,CAAC,kBAAkB,EAAE,IAAI,CAAC;SACjC,OAAO,CAAC,cAAc,EAAE,IAAI,CAAC,CAAA;IAC/B,MAAM,OAAO,GAAG,QAAQ,CAAC,KAAK,CAAC,iBAAiB,CAAC,CAAA;IACjD,OAAO,OAAO,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,CAAA;AACpC,CAAC;AAED,MAAM,UAAU,wBAAwB,CAAC,KAAuB;IAC/D,IAAI,CAAC,KAAK,CAAC,aAAa;QAAE,OAAO,EAAE,CAAA;IACnC,MAAM,OAAO,GAAG,cAAc,CAAA;IAC9B,KAAK,MAAM,KAAK,IAAI,KAAK,CAAC,aAAa,CAAC,QAAQ,CAAC,OAAO,CAAC,EAAE,CAAC;QAC3D,MAAM,EAAE,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAA;QACzB,IAAI,kBAAkB,CAAC,QAAQ,CAAC,EAAE,CAAC;YAAE,SAAQ;QAC7C,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,IAAI,CAAC,CAAA;QAC9B,OAAO;YACN,SAAS,CACR,uBAAuB,EACvB,yDAAyD,EAAE,sEAAsE,EACjI,EAAE,KAAK,EAAE,GAAG,EAAE,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CACvC;SACD,CAAA;IACF,CAAC;IACD,OAAO,EAAE,CAAA;AACV,CAAC;AAED,MAAM,sBAAsB,GAAG,iCAAiC,CAAA;AAEhE,MAAM,UAAU,wBAAwB,CAAC,KAAuB;IAC/D,IAAI,CAAC,KAAK,CAAC,aAAa;QAAE,OAAO,EAAE,CAAA;IACnC,MAAM,KAAK,GAAG,KAAK,CAAC,aAAa,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IAC7C,IAAI,MAAM,GAAG,CAAC,CAAA;IACd,KAAK,MAAM,OAAO,IAAI,KAAK,EAAE,CAAC;QAC7B,MAAM,OAAO,GAAG,OAAO,CAAC,MAAM,GAAG,CAAC,CAAA;QAClC,MAAM,OAAO,GAAG,OAAO,CAAC,SAAS,EAAE,CAAA;QACnC,IAAI,YAAY,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC;YAChC,MAAM,WAAW,GAAG,OAAO,CAAC,OAAO,CAAC,YAAY,EAAE,EAAE,CAAC,CAAA;YACrD,MAAM,KAAK,GAAG,sBAAsB,CAAC,IAAI,CAAC,WAAW,CAAC,CAAA;YACtD,IAAI,KAAK,EAAE,CAAC;gBACX,MAAM,WAAW,GAChB,MAAM;oBACN,CAAC,OAAO,CAAC,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC;oBACjC,CAAC,OAAO,CAAC,MAAM,GAAG,WAAW,CAAC,MAAM,CAAC;oBACrC,CAAC,KAAK,CAAC,KAAK,IAAI,CAAC,CAAC,CAAA;gBACnB,OAAO;oBACN,SAAS,CACR,wBAAwB,EACxB,qCAAqC,KAAK,CAAC,CAAC,CAAC,8EAA8E,EAC3H,EAAE,KAAK,EAAE,WAAW,EAAE,GAAG,EAAE,WAAW,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAC1D;iBACD,CAAA;YACF,CAAC;QACF,CAAC;QACD,MAAM,IAAI,OAAO,CAAA;IAClB,CAAC;IACD,OAAO,EAAE,CAAA;AACV,CAAC;AAED,MAAM,UAAU,cAAc,CAAC,KAAuB;IACrD,IAAI,OAAO,CAAC,KAAK,CAAC,aAAa,CAAC,IAAI,OAAO,CAAC,KAAK,CAAC,SAAS,CAAC,EAAE,CAAC;QAC9D,OAAO;YACN,SAAS,CACR,YAAY,EACZ,8FAA8F,CAC9F;SACD,CAAA;IACF,CAAC;IACD,OAAO,EAAE,CAAA;AACV,CAAC;AAED,gFAAgF;AAChF,gFAAgF;AAChF,gFAAgF;AAEhF,MAAM,CAAC,MAAM,wBAAwB,GAAG,mBAAmB,CAAC,KAAK,CAAA;AACjE,MAAM,CAAC,MAAM,6BAA6B,GAAG,gBAAgB,CAAC,KAAK,CAAA;AACnE,MAAM,CAAC,MAAM,4BAA4B,GAAG,SAAS,CAAC,KAAK,CAAA;AAC3D,MAAM,CAAC,MAAM,oBAAoB,GAAG,eAAe,CAAC,KAAK,CAAA;AACzD,MAAM,CAAC,MAAM,iBAAiB,GAAG,YAAY,CAAC,KAAK,CAAA;AACnD,MAAM,CAAC,MAAM,0BAA0B,GAAG,aAAa,CAAC,KAAK,CAAA;AAC7D,MAAM,CAAC,MAAM,0BAA0B,GAAG,gBAAgB,CAAC,KAAK,CAAA;AAChE,MAAM,CAAC,MAAM,yBAAyB,GAAG,OAAO,CAAC,KAAK,CAAA;AACtD,MAAM,CAAC,MAAM,iBAAiB,GAAG,QAAQ,CAAC,KAAK,CAAA;AAC/C,MAAM,CAAC,MAAM,gCAAgC,GAAG,iBAAiB,CAAC,KAAK,CAAA;AACvE,MAAM,CAAC,MAAM,yBAAyB,GAAG,gBAAgB,CAAC,KAAK,CAAA;AAC/D,MAAM,CAAC,MAAM,sBAAsB,GAAG,aAAa,CAAC,KAAK,CAAA;AAUzD,MAAM,CAAC,MAAM,mBAAmB,GAAe;IAC9C,GAAG,eAAe;IAClB,WAAW,EAAE,IAAI;CACjB,CAAA;AAED;;;;;GAKG;AACH,MAAM,UAAU,mBAAmB,CAClC,KAAuB,EACvB,UAA+B,EAAE;IAEjC,MAAM,IAAI,GAAe,EAAE,GAAG,mBAAmB,EAAE,GAAG,OAAO,EAAE,CAAA;IAC/D,MAAM,MAAM,GAAsB,EAAE,CAAA;IAEpC,IAAI,IAAI,CAAC,WAAW;QAAE,MAAM,CAAC,IAAI,CAAC,GAAG,cAAc,CAAC,KAAK,CAAC,CAAC,CAAA;IAC3D,MAAM,CAAC,IAAI,CAAC,GAAG,4BAA4B,CAAC,KAAK,CAAC,CAAC,CAAA;IACnD,MAAM,CAAC,IAAI,CAAC,GAAG,+BAA+B,CAAC,KAAK,CAAC,CAAC,CAAA;IACtD,MAAM,CAAC,IAAI,CAAC,GAAG,oBAAoB,CAAC,KAAK,CAAC,CAAC,CAAA;IAC3C,MAAM,CAAC,IAAI,CAAC,GAAG,mBAAmB,CAAC,KAAK,CAAC,CAAC,CAAA;IAC1C,MAAM,CAAC,IAAI,CAAC,GAAG,wBAAwB,CAAC,KAAK,CAAC,CAAC,CAAA;IAC/C,MAAM,CAAC,IAAI,CAAC,GAAG,wBAAwB,CAAC,KAAK,CAAC,CAAC,CAAA;IAC/C,WAAW;IACX,MAAM,CAAC,IAAI,CAAC,GAAG,wBAAwB,CAAC,KAAK,CAAC,CAAC,CAAA;IAC/C,MAAM,CAAC,IAAI,CAAC,GAAG,6BAA6B,CAAC,KAAK,CAAC,CAAC,CAAA;IACpD,MAAM,CAAC,IAAI,CAAC,GAAG,4BAA4B,CAAC,KAAK,CAAC,CAAC,CAAA;IACnD,MAAM,CAAC,IAAI,CAAC,GAAG,oBAAoB,CAAC,KAAK,CAAC,CAAC,CAAA;IAC3C,MAAM,CAAC,IAAI,CAAC,GAAG,iBAAiB,CAAC,KAAK,CAAC,CAAC,CAAA;IACxC,MAAM,CAAC,IAAI,CAAC,GAAG,0BAA0B,CAAC,KAAK,CAAC,CAAC,CAAA;IACjD,MAAM,CAAC,IAAI,CAAC,GAAG,0BAA0B,CAAC,KAAK,CAAC,CAAC,CAAA;IACjD,MAAM,CAAC,IAAI,CAAC,GAAG,yBAAyB,CAAC,KAAK,CAAC,CAAC,CAAA;IAChD,MAAM,CAAC,IAAI,CAAC,GAAG,iBAAiB,CAAC,KAAK,CAAC,CAAC,CAAA;IACxC,MAAM,CAAC,IAAI,CAAC,GAAG,gCAAgC,CAAC,KAAK,CAAC,CAAC,CAAA;IACvD,MAAM,CAAC,IAAI,CAAC,GAAG,yBAAyB,CAAC,KAAK,CAAC,CAAC,CAAA;IAChD,MAAM,CAAC,IAAI,CAAC,GAAG,sBAAsB,CAAC,KAAK,CAAC,CAAC,CAAA;IAC7C,aAAa;IACb,MAAM,CAAC,IAAI,CAAC,GAAG,+BAA+B,CAAC,KAAK,CAAC,CAAC,CAAA;IACtD,MAAM,CAAC,IAAI,CAAC,GAAG,oCAAoC,CAAC,KAAK,EAAE,IAAI,CAAC,CAAC,CAAA;IAEjE,OAAO,MAAM,CAAC,MAAM,KAAK,CAAC,CAAC,CAAC,CAAC,EAAE,EAAE,EAAE,IAAI,EAAE,CAAC,CAAC,CAAC,EAAE,EAAE,EAAE,KAAK,EAAE,MAAM,EAAE,CAAA;AAClE,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,WAAW,CAAC,MAAyB;IACpD,OAAO,MAAM,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,QAAQ,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,OAAO,CAAC,CAAA;AAC1D,CAAC;AAED,MAAM,UAAU,UAAU,CAAC,MAAyB;IACnD,OAAO,MAAM,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,QAAQ,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,MAAM,CAAC,CAAA;AACzD,CAAC;AAED,MAAM,UAAU,sBAAsB,CAAC,MAAyB;IAC/D,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC;QAAE,OAAO,IAAI,CAAA;IACpC,OAAO,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,IAAI,CAAC,CAAC,IAAI,KAAK,CAAC,CAAC,OAAO,EAAE,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;AAChE,CAAC;AAUD,OAAO,EAAE,eAAe,EAAE,QAAQ,EAAE,CAAA"}

package/dist/engine/validator/scan.d.ts

@@ -0,0 +1,28 @@
+/**
+ * Shared markdown-aware scanning helpers used by checks that need to
+ * skip fenced code, inline code, HTML attributes, and existing links.
+ *
+ * Ported verbatim from fulcrum-labs/fronts
+ * src/lib/article-body-validator/scan.ts (inventory §1: extract).
+ *
+ * The validator runs on PRE-strip source, so these helpers must match
+ * the textual reality of stored markdown — not the cleaned form that
+ * ships to the renderer. Each helper is pure and operates over the raw
+ * `body_markdown`. We intentionally avoid pulling in a full CommonMark
+ * AST library: the checks are conservative regex-based scans that
+ * classify each character span as "inside a protected region"
+ * (code/link/attr) vs "scannable prose".
+ */
+/**
+ * Mark which character offsets in the source are inside fenced code
+ * blocks (` ```...``` `), inline code spans (`` `…` ``), markdown link
+ * targets `[…](url)`, autolinks `<…>` , and HTML attribute values
+ * (`src="…"`). Each offset is `true` if the character is protected.
+ */
+export declare function buildProtectedMask(source: string): boolean[];
+/**
+ * Returns true if any character in the [start, end) range is inside a
+ * protected region.
+ */
+export declare function isProtectedSpan(mask: boolean[], start: number, end: number): boolean;
+//# sourceMappingURL=scan.d.ts.map

package/dist/engine/validator/scan.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"scan.d.ts","sourceRoot":"","sources":["../../../src/engine/validator/scan.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;GAcG;AAEH;;;;;GAKG;AACH,wBAAgB,kBAAkB,CAAC,MAAM,EAAE,MAAM,GAAG,OAAO,EAAE,CA+D5D;AAQD;;;GAGG;AACH,wBAAgB,eAAe,CAAC,IAAI,EAAE,OAAO,EAAE,EAAE,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,GAAG,OAAO,CAKpF"}

package/dist/engine/validator/scan.js

@@ -0,0 +1,97 @@
+/**
+ * Shared markdown-aware scanning helpers used by checks that need to
+ * skip fenced code, inline code, HTML attributes, and existing links.
+ *
+ * Ported verbatim from fulcrum-labs/fronts
+ * src/lib/article-body-validator/scan.ts (inventory §1: extract).
+ *
+ * The validator runs on PRE-strip source, so these helpers must match
+ * the textual reality of stored markdown — not the cleaned form that
+ * ships to the renderer. Each helper is pure and operates over the raw
+ * `body_markdown`. We intentionally avoid pulling in a full CommonMark
+ * AST library: the checks are conservative regex-based scans that
+ * classify each character span as "inside a protected region"
+ * (code/link/attr) vs "scannable prose".
+ */
+/**
+ * Mark which character offsets in the source are inside fenced code
+ * blocks (` ```...``` `), inline code spans (`` `…` ``), markdown link
+ * targets `[…](url)`, autolinks `<…>` , and HTML attribute values
+ * (`src="…"`). Each offset is `true` if the character is protected.
+ */
+export function buildProtectedMask(source) {
+    const len = source.length;
+    const mask = new Array(len).fill(false);
+    // 1) Fenced code blocks. Match either ``` or ~~~ fences; cover both
+    //    the fence lines and the body between them.
+    const fenceRe = /(^|\n)([ \t]*)(`{3,}|~{3,})[^\n]*\n([\s\S]*?)\n\2\3/g;
+    for (const match of source.matchAll(fenceRe)) {
+        const start = match.index ?? 0;
+        const end = start + match[0].length;
+        fill(mask, start, end);
+    }
+    // 2) Inline code spans. We deliberately accept single-backtick spans
+    //    only — multi-backtick spans are rare and the cost of missing
+    //    them is one extra warning on an unusual page, not a render bug.
+    const inlineCodeRe = /`[^`\n]+`/g;
+    for (const match of source.matchAll(inlineCodeRe)) {
+        const start = match.index ?? 0;
+        const end = start + match[0].length;
+        // Skip if we're already inside a protected (fenced) region.
+        if (mask[start])
+            continue;
+        fill(mask, start, end);
+    }
+    // 3) Markdown links — the URL portion of `[label](url)` and the
+    //    label of reference links `[label][id]` plus the ref target
+    //    `[id]: url`. We protect the URL only so e.g. label text still
+    //    scans for normal lint codes.
+    const linkUrlRe = /\]\(([^)\s]+)(?:\s+"[^"]*")?\)/g;
+    for (const match of source.matchAll(linkUrlRe)) {
+        const start = (match.index ?? 0) + 2; // skip past `](`
+        const end = start + (match[1]?.length ?? 0);
+        fill(mask, start, end);
+    }
+    const refDefRe = /^\s*\[[^\]]+\]:\s*\S+/gm;
+    for (const match of source.matchAll(refDefRe)) {
+        const start = match.index ?? 0;
+        const end = start + match[0].length;
+        fill(mask, start, end);
+    }
+    // 4) Autolinks <https://…>.
+    const autolinkRe = /<https?:\/\/[^>\s]+>/g;
+    for (const match of source.matchAll(autolinkRe)) {
+        const start = match.index ?? 0;
+        const end = start + match[0].length;
+        fill(mask, start, end);
+    }
+    // 5) HTML attribute values (single or double quoted). This catches
+    //    src="https://…", href="…", style attributes, etc. We do this
+    //    after fenced/inline code so we don't reprotect those regions.
+    const attrRe = /\s[a-zA-Z][\w-]*\s*=\s*("[^"]*"|'[^']*')/g;
+    for (const match of source.matchAll(attrRe)) {
+        const valueStart = (match.index ?? 0) + match[0].indexOf(match[1] ?? '');
+        if (valueStart < 0)
+            continue;
+        const end = valueStart + (match[1]?.length ?? 0);
+        fill(mask, valueStart, end);
+    }
+    return mask;
+}
+function fill(mask, start, end) {
+    for (let i = start; i < end && i < mask.length; i += 1) {
+        mask[i] = true;
+    }
+}
+/**
+ * Returns true if any character in the [start, end) range is inside a
+ * protected region.
+ */
+export function isProtectedSpan(mask, start, end) {
+    for (let i = start; i < end && i < mask.length; i += 1) {
+        if (mask[i])
+            return true;
+    }
+    return false;
+}
+//# sourceMappingURL=scan.js.map

package/dist/engine/validator/scan.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"scan.js","sourceRoot":"","sources":["../../../src/engine/validator/scan.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;GAcG;AAEH;;;;;GAKG;AACH,MAAM,UAAU,kBAAkB,CAAC,MAAc;IAChD,MAAM,GAAG,GAAG,MAAM,CAAC,MAAM,CAAA;IACzB,MAAM,IAAI,GAAG,IAAI,KAAK,CAAU,GAAG,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;IAEhD,oEAAoE;IACpE,gDAAgD;IAChD,MAAM,OAAO,GAAG,sDAAsD,CAAA;IACtE,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,QAAQ,CAAC,OAAO,CAAC,EAAE,CAAC;QAC9C,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,IAAI,CAAC,CAAA;QAC9B,MAAM,GAAG,GAAG,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,CAAA;QACnC,IAAI,CAAC,IAAI,EAAE,KAAK,EAAE,GAAG,CAAC,CAAA;IACvB,CAAC;IAED,qEAAqE;IACrE,kEAAkE;IAClE,qEAAqE;IACrE,MAAM,YAAY,GAAG,YAAY,CAAA;IACjC,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,QAAQ,CAAC,YAAY,CAAC,EAAE,CAAC;QACnD,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,IAAI,CAAC,CAAA;QAC9B,MAAM,GAAG,GAAG,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,CAAA;QACnC,4DAA4D;QAC5D,IAAI,IAAI,CAAC,KAAK,CAAC;YAAE,SAAQ;QACzB,IAAI,CAAC,IAAI,EAAE,KAAK,EAAE,GAAG,CAAC,CAAA;IACvB,CAAC;IAED,gEAAgE;IAChE,gEAAgE;IAChE,mEAAmE;IACnE,kCAAkC;IAClC,MAAM,SAAS,GAAG,iCAAiC,CAAA;IACnD,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,QAAQ,CAAC,SAAS,CAAC,EAAE,CAAC;QAChD,MAAM,KAAK,GAAG,CAAC,KAAK,CAAC,KAAK,IAAI,CAAC,CAAC,GAAG,CAAC,CAAA,CAAC,iBAAiB;QACtD,MAAM,GAAG,GAAG,KAAK,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,MAAM,IAAI,CAAC,CAAC,CAAA;QAC3C,IAAI,CAAC,IAAI,EAAE,KAAK,EAAE,GAAG,CAAC,CAAA;IACvB,CAAC;IAED,MAAM,QAAQ,GAAG,yBAAyB,CAAA;IAC1C,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,QAAQ,CAAC,QAAQ,CAAC,EAAE,CAAC;QAC/C,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,IAAI,CAAC,CAAA;QAC9B,MAAM,GAAG,GAAG,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,CAAA;QACnC,IAAI,CAAC,IAAI,EAAE,KAAK,EAAE,GAAG,CAAC,CAAA;IACvB,CAAC;IAED,4BAA4B;IAC5B,MAAM,UAAU,GAAG,uBAAuB,CAAA;IAC1C,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,QAAQ,CAAC,UAAU,CAAC,EAAE,CAAC;QACjD,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,IAAI,CAAC,CAAA;QAC9B,MAAM,GAAG,GAAG,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,CAAA;QACnC,IAAI,CAAC,IAAI,EAAE,KAAK,EAAE,GAAG,CAAC,CAAA;IACvB,CAAC;IAED,mEAAmE;IACnE,kEAAkE;IAClE,mEAAmE;IACnE,MAAM,MAAM,GAAG,2CAA2C,CAAA;IAC1D,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,QAAQ,CAAC,MAAM,CAAC,EAAE,CAAC;QAC7C,MAAM,UAAU,GAAG,CAAC,KAAK,CAAC,KAAK,IAAI,CAAC,CAAC,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAA;QACxE,IAAI,UAAU,GAAG,CAAC;YAAE,SAAQ;QAC5B,MAAM,GAAG,GAAG,UAAU,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,MAAM,IAAI,CAAC,CAAC,CAAA;QAChD,IAAI,CAAC,IAAI,EAAE,UAAU,EAAE,GAAG,CAAC,CAAA;IAC5B,CAAC;IAED,OAAO,IAAI,CAAA;AACZ,CAAC;AAED,SAAS,IAAI,CAAC,IAAe,EAAE,KAAa,EAAE,GAAW;IACxD,KAAK,IAAI,CAAC,GAAG,KAAK,EAAE,CAAC,GAAG,GAAG,IAAI,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC;QACxD,IAAI,CAAC,CAAC,CAAC,GAAG,IAAI,CAAA;IACf,CAAC;AACF,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,eAAe,CAAC,IAAe,EAAE,KAAa,EAAE,GAAW;IAC1E,KAAK,IAAI,CAAC,GAAG,KAAK,EAAE,CAAC,GAAG,GAAG,IAAI,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC;QACxD,IAAI,IAAI,CAAC,CAAC,CAAC;YAAE,OAAO,IAAI,CAAA;IACzB,CAAC;IACD,OAAO,KAAK,CAAA;AACb,CAAC"}