Merge pull request #11191 from quarto-dev/bugfix/10285

Index pre-section contents of book chapters and `.quarto-include-in-search-index` elements

Author: cscheid

news/changelog-1.6.md

@@ -78,6 +78,7 @@ All changes included in 1.6:
 ## Projects
 
 - ([#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.
 - ([#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

@@ -326,19 +326,31 @@ export async function updateSearchIndex(
 
         // If there are any paragraphs residing outside a section, just
         // include that in the document entry
-        const pararaphNodes = doc.querySelectorAll(
-          `${mainSelector} > p, ${mainSelector} > div.cell`,
+        const paragraphNodes = doc.querySelectorAll(
+          `${mainSelector} > p, ${mainSelector} > div.cell, main > p`,
         );
 
-        for (const paragraphNode of pararaphNodes) {
+        const addToIndex = (paragraphNode: Element) => {
           const text = paragraphNode.textContent.trim();
           if (text) {
             pageText.push(text);
           }
 
           // Since these are already indexed with the main entry, remove them
           // so they are not indexed again
-          (paragraphNode as Element).remove();
+          paragraphNode.remove();
+        };
+
+        for (const paragraphNode of paragraphNodes) {
+          addToIndex(paragraphNode as Element);
+        }
+
+        // Add forced inclusions to the index
+        const forcedInclusions = doc.querySelectorAll(
+          `.quarto-include-in-search-index`,
+        );
+        for (const forcedInclusion of forcedInclusions) {
+          addToIndex(forcedInclusion as Element);
         }
 
         if (pageText.length > 0) {

tests/docs/search/issue-10285/.gitignore

@@ -0,0 +1 @@
+/.quarto/

tests/docs/search/issue-10285/_quarto.yml

@@ -0,0 +1,25 @@
+project:
+  type: book
+  post-render:
+    - check-index.ts
+
+book:
+  title: "issue-10285"
+  author: "Norah Jones"
+  date: "10/24/2024"
+  chapters:
+    - index.qmd
+    - intro.qmd
+    - search-test.qmd
+    - search-test-2.qmd
+    - search-test-3.qmd
+    - summary.qmd
+    - references.qmd
+
+bibliography: references.bib
+
+format:
+  html:
+    theme: cosmo
+  pdf:
+    documentclass: scrreprt

tests/docs/search/issue-10285/check-index.ts

@@ -0,0 +1,9 @@
+const json = JSON.parse(Deno.readTextFileSync("_book/search.json"));
+
+const obj = Object.fromEntries(json.map((x: any) => [x.objectID, x]));
+
+for (const file of ["search-test.html", "search-test-2.html", "search-test-3.html"]) {
+  if (obj[file].text.match("Please find me.") === null) {
+    throw new Error("missing pre-section content in " + file);
+  };
+}

tests/docs/search/issue-10285/cover.png

@@ -0,0 +1,5 @@
+# Preface {.unnumbered}
+
+This is a Quarto book.
+
+To learn more about Quarto books visit <https://quarto.org/docs/books>.

tests/docs/search/issue-10285/index.qmd

@@ -0,0 +1,5 @@
+# Introduction
+
+This is a book created from markdown and executable code.
+
+See @knuth84 for additional discussion of literate programming.

tests/docs/search/issue-10285/intro.qmd

@@ -0,0 +1,19 @@
+@article{knuth84,
+  author = {Knuth, Donald E.},
+  title = {Literate Programming},
+  year = {1984},
+  issue_date = {May 1984},
+  publisher = {Oxford University Press, Inc.},
+  address = {USA},
+  volume = {27},
+  number = {2},
+  issn = {0010-4620},
+  url = {https://doi.org/10.1093/comjnl/27.2.97},
+  doi = {10.1093/comjnl/27.2.97},
+  journal = {Comput. J.},
+  month = may,
+  pages = {97–111},
+  numpages = {15}
+}
+
+

tests/docs/search/issue-10285/references.bib

@@ -0,0 +1,4 @@
+# References {.unnumbered}
+
+::: {#refs}
+:::