[dependabot]

Bumps the docs-search group with 1 update in the / directory: pagefind.

Updates pagefind from 1.5.2 to 1.6.0

Release notes

Sourced from pagefind's releases.

v1.6.0

👋 A little overdue, but this release knocks out a healthy selection of bug reports, and includes a lot of work from the community. Thanks to everyone who contributed ❤️ - @​bglw

Multilingual Segmentation

Pagefind's extended build now segments Chinese, Japanese, and Thai pages with ICU4X, replacing the previous dependency (PR #1383). Alongside producing a smaller binary, ICU4X segments a lot closer to the Intl.Segmenter used in the browser on the search query (which is ICU4C). This closes out a mixture of latent and reported issues where the different segmentation between indexing and searching makes for iffy results. It's not perfect yet, but it's a darn sight closer.

Bonuses:

• Segmentation is smarter in languages such as Thai, and doesn't break inside a grapheme cluster
• Japanese katakana has special handling, but should still be improved over 1.5.2
• Japanese voiced and semi-voiced sound marks are no longer normalized into each other

Fewer Fragment Files

Pagefind now has an option to limit the total number of fragment files produced, for hosting providers that cap the number of uploaded files. Contributed in PR #1020 — thanks @​yakovsh !

This limit is set with --max-fragments / max_fragments, and applies per language. When applied, multiple fragments are bundled together and loaded as a set when required.

See 📘 Max fragments.

Explicit Sub-Result Titles

Added support for explicit sub-result titles. Any element with both an id and a data-pagefind-title attribute can be used as a sub-result (PR #1282 — thanks @​samuel-williams-shopify !)

See 📘 Defining a sub result explicitly.

Component UI

• Added a sort attribute to <pagefind-config>, to order results by a sort key (PR #1326)
• Added a max-sub-results attribute to <pagefind-searchbox> (PR #1324)
• Added a placeholder setter to <pagefind-searchbox> and <pagefind-modal-trigger> (PR #1321)
• configureInstance now applies to an instance that already exists but has not loaded yet (PR #1323)
• Fixed the UI language being resolved after the first render, which could flash the wrong translations (PR #1332)
• Fixed component cleanup after being disconnected (PR #1322)
• Fixed IME composition keystrokes support (PR #1284 — thanks @​oharu121 !)
• Fixed the modal staying open when clicking a result for the current page (PR #1327)
• Fixed the filter panel checkbox shrinking when its label wraps (PR #1166 — thanks @​mbrambilla !)
• Hid the keyboard shortcut badge on touch-only devices (PR #1325)

Search Results & Options

• Added an excerptEllipsis option to mark the truncated ends of excerpts (PR #1320)
  • See 📘 Excerpt ellipsis
• Fixed phrase search across diacritic variants, so searching "cafe latte" can match a page containing café latte (PR #1386 — thanks @​sean0407 !)
• Fixed filters not matching values containing escaped characters such as quotes or backslashes (PR #1392)
• Fixed compound words being counted twice in metadata coverage, which could overvalue a metadata match (PR #1319)
• Fixed a bug which could build excerpts from the wrong words (PR #1318)
• Fixed decoding for HTML entities in the automatic image and image_alt metadata (PR #1393 — thanks @​digilist !)

... (truncated)

Changelog

Sourced from pagefind's changelog.

v1.6.0 (November 1, 2026)

👋 A little overdue, but this release knocks out a healthy selection of bug reports, and includes a lot of work from the community. Thanks to everyone who contributed ❤️ - @​bglw

Multilingual Segmentation

Pagefind's extended build now segments Chinese, Japanese, and Thai pages with ICU4X, replacing the previous dependency (PR #1383). Alongside producing a smaller binary, ICU4X segments a lot closer to the Intl.Segmenter used in the browser on the search query (which is ICU4C). This closes out a mixture of latent and reported issues where the different segmentation between indexing and searching makes for iffy results. It's not perfect yet, but it's a darn sight closer.

Bonuses:

• Segmentation is smarter in languages such as Thai, and doesn't break inside a grapheme cluster
• Japanese katakana has special handling, but should still be improved over 1.5.2
• Japanese voiced and semi-voiced sound marks are no longer normalized into each other

Fewer Fragment Files

Pagefind now has an option to limit the total number of fragment files produced, for hosting providers that cap the number of uploaded files. Contributed in PR #1020 — thanks @​yakovsh !

This limit is set with --max-fragments / max_fragments, and applies per language. When applied, multiple fragments are bundled together and loaded as a set when required.

See 📘 Max fragments.

Explicit Sub-Result Titles

Added support for explicit sub-result titles. Any element with both an id and a data-pagefind-title attribute can be used as a sub-result (PR #1282 — thanks @​samuel-williams-shopify !)

See 📘 Defining a sub result explicitly.

Component UI

• Added a sort attribute to <pagefind-config>, to order results by a sort key (PR #1326)
• Added a max-sub-results attribute to <pagefind-searchbox> (PR #1324)
• Added a placeholder setter to <pagefind-searchbox> and <pagefind-modal-trigger> (PR #1321)
• configureInstance now applies to an instance that already exists but has not loaded yet (PR #1323)
• Fixed the UI language being resolved after the first render, which could flash the wrong translations (PR #1332)
• Fixed component cleanup after being disconnected (PR #1322)
• Fixed IME composition keystrokes support (PR #1284 — thanks @​oharu121 !)
• Fixed the modal staying open when clicking a result for the current page (PR #1327)
• Fixed the filter panel checkbox shrinking when its label wraps (PR #1166 — thanks @​mbrambilla !)
• Hid the keyboard shortcut badge on touch-only devices (PR #1325)

Search Results & Options

• Added an excerptEllipsis option to mark the truncated ends of excerpts (PR #1320)
  • See 📘 Excerpt ellipsis
• Fixed phrase search across diacritic variants, so searching "cafe latte" can match a page containing café latte (PR #1386 — thanks @​sean0407 !)
• Fixed filters not matching values containing escaped characters such as quotes or backslashes (PR #1392)
• Fixed compound words being counted twice in metadata coverage, which could overvalue a metadata match (PR #1319)
• Fixed a bug which could build excerpts from the wrong words (PR #1318)
• Fixed decoding for HTML entities in the automatic image and image_alt metadata (PR #1393 — thanks @​digilist !)

... (truncated)

Commits

• 860a4ea Changelog for 1.6.0
• deea3c9 Changelog (#1403)
• 13ebd8e Update multilignual docs (#1402)
• 0544804 Adding support for fewer number of fragment files (#74 and #739) (#1020)
• 9b465ea fix: percent-encode page URLs built from file paths (#1335)
• fcf939c chore(deps): bump certifi (#1385)
• 124b73c fix: EACCES during playground build (#1371)
• f537e50 Fix exact phrase matching across diacritic variants (#1386)
• a0d31b7 Decode HTML entities in the automatic image metadata (#1393)
• c9d4fda Perf: group matching words by page before scoring (#1334)
• Additional commits viewable in compare view

[cloudflare-workers-and-pages]

🚀 Deploying Preview to Cloudflare 🚀

Preview URL: https://dependabot-npm-and-yarn-docs-search-f2267c0-developer-sumup-com.sumup-limited.workers.dev (commit 6c9c49f)

This URL reflects your latest Preview deployment

Preview Deployments by commit

Status	Deployment URL	Commit	Updated (UTC)	See this deployment's details
Build: Success ✅ Deployment: Success ✅ View logs ↗	https://e31149aa-developer-sumup-com.sumup-limited.workers.dev	6c9c49f	2026-11-06T21:29:19.128Z	Visit the dashboard ↗
Build: Success ✅ Deployment: Success ✅ View logs ↗	https://59c0bcdc-developer-sumup-com.sumup-limited.workers.dev	a28fae0	2026-11-06T03:15:48.424Z	Visit the dashboard ↗
Build: Success ✅ Deployment: Success ✅ View logs ↗	https://a6943cb4-developer-sumup-com.sumup-limited.workers.dev	67a2382	2026-11-05T03:16:03.031Z	Visit the dashboard ↗