Fixes quarto-dev#11189

- Users will now be able to create `div`s and `span`s that exclude
certain page elements from the search index.

news/changelog-1.6.md

@@ -79,6 +79,7 @@ All changes included in 1.6:
 
 - ([#10268](https://github.com/quarto-dev/quarto-cli/issues/10268)): `quarto create` supports opening project in Positron, in addition to VS Code and RStudio IDE.
 - ([#10285](https://github.com/quarto-dev/quarto-cli/issues/10285)): Include text from before the first chapter sections in search indices. In addition, include text of every element with `.quarto-include-in-search-index` class in search indices.
+- ([#11189](https://github.com/quarto-dev/quarto-cli/issues/11189)): Exclude text of every element with the `.quarto-exclude-from-search-index` class from search indices.
 - ([#10566](https://github.com/quarto-dev/quarto-cli/issues/10566)): Ensure that `quarto run` outputs `stdout` and `stderr` to the correct streams.
 
 ### Websites

src/project/types/website/website-search.ts

@@ -291,6 +291,12 @@ export async function updateSearchIndex(
         }
       });
 
+      // Remove all page elements that should be excluded from the search index
+      const exclusions = doc.querySelectorAll(".quarto-exclude-from-search-index");
+      for (const exclusion of exclusions) {
+        exclusion._remove();
+      }
+
       // We always take the first child of the main region (whether that is a p or section)
       // and create an index entry for the page itself (with no hash). If there is other
       // 'unsectioned' content on the page, we include that as well.

tests/docs/search/issue-11189/.gitignore

@@ -0,0 +1,6 @@
+/.quarto/
+*.html
+*.pdf
+*_files/
+/_site_with_filter/
+/_site_without_filter/

tests/docs/search/issue-11189/_extensions/remove-class/_extension.yml

@@ -0,0 +1,8 @@
+title: Remove-class
+author: Nick Vigilante
+version: 1.0.0
+quarto-required: ">=99.9.0"
+contributes:
+  filters:
+    - remove-class.lua
+

tests/docs/search/issue-11189/_extensions/remove-class/remove-class.lua

@@ -0,0 +1,8 @@
+
+-- Reformat all heading text 
+function RemoveClass(el)
+  class = "quarto-exclude-from-search-index"
+  if el.classes:includes(class) then
+    el.classes:remove(class)
+  end
+end

tests/docs/search/issue-11189/_quarto-with-filter.yml

@@ -0,0 +1,4 @@
+project:
+  output-dir: _site_with_filter
+  post-render:
+    - check-index-with-filter.ts

tests/docs/search/issue-11189/_quarto-without-filter.yml

@@ -0,0 +1,4 @@
+project:
+  output-dir: _site_without_filter
+  post-render:
+    - check-index-without-filter.ts

tests/docs/search/issue-11189/_quarto.yml

@@ -0,0 +1,19 @@
+project:
+  type: website
+
+profile:
+  default: without-filter
+
+website:
+  title: "issue-11189"
+  navbar:
+    left:
+      - href: index.qmd
+        text: Home
+      - about.qmd
+
+format:
+  html:
+    theme: cosmo
+    css: styles.css
+    toc: true

tests/docs/search/issue-11189/about.qmd

@@ -0,0 +1,11 @@
+---
+title: "About"
+---
+
+About this site
+
+::: {.quarto-exclude-from-search-index}
+
+Please don't find me.
+
+:::

tests/docs/search/issue-11189/check-index-with-filter.ts

@@ -0,0 +1,9 @@
+const json = JSON.parse(Deno.readTextFileSync("_site_with_filter/search.json"));
+
+const obj = Object.fromEntries(json.map((x: any) => [x.objectID, x]));
+
+const file = "index.html";
+
+if (obj[file].text.match("Please find me.") === null) {
+  throw new Error("could not find text that should be shown in " + file);
+};

tests/docs/search/issue-11189/check-index-without-filter.ts

@@ -0,0 +1,9 @@
+const json = JSON.parse(Deno.readTextFileSync("_site_without_filter/search.json"));
+
+const obj = Object.fromEntries(json.map((x: any) => [x.objectID, x]));
+
+const file = "about.html";
+
+if (obj[file].text.match("Please don't find me.") !== null) {
+  throw new Error("found text that should be hidden in " + file);
+};

tests/docs/search/issue-11189/index.qmd

@@ -0,0 +1,15 @@
+---
+title: "issue-11189"
+filters:
+  - _extensions/remove-class/remove-class.lua
+---
+
+This is a Quarto website.
+
+To learn more about Quarto websites visit <https://quarto.org/docs/websites>.
+
+::: {.quarto-exclude-from-search-index}
+
+Please find me.
+
+:::

tests/docs/search/issue-11189/styles.css

@@ -0,0 +1 @@
+/* css styles */

tests/smoke/search/validate-search-index-exclusion.test.ts

@@ -0,0 +1,40 @@
+import { testQuartoCmd } from "../../test.ts";
+import { fileExists, noErrorsOrWarnings } from "../../verify.ts";
+
+import { existsSync } from "../../../src/deno_ral/fs.ts";
+import { join } from "../../../src/deno_ral/path.ts";
+import { docs } from "../../utils.ts";
+
+// Test a simple site
+const input = docs("search/issue-11189");
+const withFilterProfile = "--profile with-filter"
+
+testQuartoCmd(
+  "render",
+  [input],
+  [noErrorsOrWarnings],
+  {
+    name: "Test search exclusions without filter",
+    teardown: async () => {
+      const siteDir = join(input, "_site_without_filter");
+      if (existsSync(siteDir)) {
+        await Deno.remove(siteDir, { recursive: true });
+      }
+    },
+  },
+);
+
+testQuartoCmd(
+  "render",
+  [withFilterProfile, input],
+  [noErrorsOrWarnings],
+  {
+    name: "Test search exclusions with filter",
+    teardown: async () => {
+      const siteDir = join(input, "_site_with_filter");
+      if (existsSync(siteDir)) {
+        await Deno.remove(siteDir, { recursive: true });
+      }
+    },
+  },
+);