remove some markdown & html syntax in feed and metas plugins

CHANGELOG.md

@@ -34,6 +34,7 @@ Go to the `v1` branch to see the changelog of Lume 1.
 - `theme.path` option of `code_highlight` plugin.
 
 ### Fixed
+- Removed markdown syntax for some values in `metas`, `feed` and `json_ld`.
 - `relative_urls` plugin with no pretty urls [#711].
 - Updated deps: `sass`, `preact`, `xml`, `katex`, `jsr`, `pagefind`, `highlight.js`, `tailwindcss`, `magic-string`, `std`, `esbuild`, `liquid`, `markdown-it-attrs`, `unocss`, `vento`, `deno_dom`, `unocss`, `satori` and some icons.
 - Several changes on `esbuild` plugin:

core/utils/data_values.ts

@@ -1,3 +1,5 @@
+import { plainText } from "../../deps/remove-markdown.ts";
+
 import type { Data } from "../file.ts";
 
 /**
@@ -18,6 +20,16 @@ export function getDataValue(data: Partial<Data>, value?: unknown) {
   return value;
 }
 
+export function getPlainDataValue(data: Partial<Data>, value?: unknown) {
+  const val = getDataValue(data, value);
+
+  if (typeof val === "string") {
+    return plainText(val);
+  }
+
+  return val;
+}
+
 function searchValue(data: Partial<Data>, value: string): unknown {
   if (!value) {
     return;

deps/remove-markdown.ts

@@ -0,0 +1,25 @@
+import removeMarkdown from "npm:[email protected]";
+
+export function plainText(md: string, options?: RemoveMarkdownOptions): string {
+  return removeMarkdown(md.replaceAll(/\s+/g, " ").trim(), options);
+}
+
+export interface RemoveMarkdownOptions {
+  /** strip list leaders (default: true) */
+  stripListLeaders?: boolean;
+
+  /** char to insert instead of stripped list leaders (default: '') */
+  listUnicodeChar?: string;
+
+  /** support GitHub-Flavored Markdown (default: true) */
+  gfm?: boolean;
+
+  /** replace images with alt-text, if present (default: true) */
+  useImgAltText: boolean;
+
+  abbr?: boolean;
+
+  /** replace links with URLs instead anchor text (default: false) */
+  replaceLinksWithURL?: boolean;
+  htmlTagsToSkip?: string[];
+}

plugins/feed.ts

@@ -1,7 +1,7 @@
 import { getExtension } from "../core/utils/path.ts";
 import { isPlainObject, merge } from "../core/utils/object.ts";
 import { getCurrentVersion } from "../core/utils/lume_version.ts";
-import { getDataValue } from "../core/utils/data_values.ts";
+import { getDataValue, getPlainDataValue } from "../core/utils/data_values.ts";
 import { cdata, stringify } from "../deps/xml.ts";
 import { Page } from "../core/file.ts";
 import { parseDate } from "../core/utils/date.ts";
@@ -170,8 +170,8 @@ export function feed(userOptions?: Options) {
       const rootData = site.source.data.get("/") || {};
 
       const feed: FeedData = {
-        title: getDataValue(rootData, info.title),
-        description: getDataValue(rootData, info.description),
+        title: getPlainDataValue(rootData, info.title),
+        description: getPlainDataValue(rootData, info.description),
         published: getDataValue(rootData, info.published),
         lang: getDataValue(rootData, info.lang),
         url: site.url("", true),
@@ -189,9 +189,9 @@ export function feed(userOptions?: Options) {
             : undefined;
 
           return {
-            title: getDataValue(data, items.title),
+            title: getPlainDataValue(data, items.title),
             url: site.url(data.url, true),
-            description: getDataValue(data, items.description),
+            description: getPlainDataValue(data, items.description),
             author: getAuthor(data, items),
             published: toDate(getDataValue(data, items.published)) ||
               new Date(),
@@ -234,7 +234,7 @@ function getAuthor(
   data: Partial<Data>,
   info: FeedInfoOptions | FeedItemOptions,
 ): Author | undefined {
-  const name = getDataValue(data, info.authorName);
+  const name = getPlainDataValue(data, info.authorName);
   const url = getDataValue(data, info.authorUrl);
 
   if (name || url) {

plugins/json_ld.ts

@@ -1,5 +1,5 @@
 import { isPlainObject, merge } from "../core/utils/object.ts";
-import { getDataValue } from "../core/utils/data_values.ts";
+import { getPlainDataValue } from "../core/utils/data_values.ts";
 
 import type Site from "../core/site.ts";
 import type { Page } from "../core/file.ts";
@@ -205,7 +205,7 @@ export function jsonLd(userOptions?: Options) {
       // Recursive function to traverse and process JSON-LD data
       function traverse(key: string | undefined, value: unknown): unknown {
         if (typeof value === "string") {
-          const dataValue = getDataValue(data, value);
+          const dataValue = getPlainDataValue(data, value);
           // Check if the value is a URL or ID that needs to be processed
           if (
             key &&

plugins/metas.ts

@@ -1,6 +1,6 @@
 import { merge } from "../core/utils/object.ts";
 import { getCurrentVersion } from "../core/utils/lume_version.ts";
-import { getDataValue } from "../core/utils/data_values.ts";
+import { getDataValue, getPlainDataValue } from "../core/utils/data_values.ts";
 
 import type Site from "../core/site.ts";
 import type { Data, Page } from "../core/file.ts";
@@ -96,12 +96,12 @@ export function metas(userOptions?: Options) {
         : undefined;
 
       const type = getDataValue(data, main["type"]);
-      const site_name = getDataValue(data, main["site"]);
+      const site_name = getPlainDataValue(data, main["site"]);
       const lang = getDataValue(data, main["lang"]);
-      const title = getDataValue(data, main["title"]);
-      const description = getDataValue(data, main["description"]);
-      const twitter = getDataValue(data, main["twitter"]);
-      const fediverse = getDataValue(data, main["fediverse"]);
+      const title = getPlainDataValue(data, main["title"]);
+      const description = getPlainDataValue(data, main["description"]);
+      const twitter = getPlainDataValue(data, main["twitter"]);
+      const fediverse = getPlainDataValue(data, main["fediverse"]);
       const keywords = getDataValue(data, main["keywords"]);
       const robots = getDataValue(data, main["robots"]);
       const color = getDataValue(data, main["color"]);
@@ -181,10 +181,6 @@ function addMeta(
   if (!content) {
     return;
   }
-  content = content
-    .replaceAll(/<[^>]*>/g, "")
-    .replaceAll(/\s+/g, " ")
-    .trim();
 
   if (limit && content.length > limit) {
     content = content.substring(0, limit - 1).trimEnd() + "…";

tests/__snapshots__/feed.test.ts.snap

@@ -259,7 +259,7 @@ snapshot[`RSS plugin 3`] = `
       published: "1979-06-21T23:45:00.000Z",
       search: [],
       tags: "Array(0)",
-      title: "Page 5",
+      title: "Page **5**",
       url: "/page5/",
     },
     src: {

tests/assets/feed/page5.yaml

@@ -1,4 +1,4 @@
-title: Page 5
+title: Page **5**
 content: Content of Page 5
 date: 1979-06-21T23:45:00.000Z
 published: "1979-06-21T23:45:00.000Z"

tests/assets/metas/page-2.vto

@@ -1,6 +1,6 @@
 ---
 metas:
-  title: Relative paths
+  title: Relative **paths**
   description: Tests the use of relative path (to page.data.url) when filling out the og:image or og:icon URL
   image: ./my-image.png
   icon: ./my-icon.png