Open Source Update / April '25
2025-04-06, originally sent directly to open-source sponsors. See previous update.
The last month has been pretty busy, and I haven't had as much time as I wanted to invest in open source. Some notable events in my personal life: my daughter got her driver's license 😅 and I hit the six-month mark at 37 signals 😲!
Despite being a bit busy, I feel like I still got quite a bit done Here's an overview of what I shipped in the last month …
Nokogiri
Releases
- Release v1.18.4 / 2025-03-14 · sparklemotion/nokogiri
- and wrote the accompanying GHSA Nokogiri updates packaged libxslt to v1.1.43 to resolve multiple CVEs · Advisory · sparklemotion/nokogiri
- Release v1.18.5 / 2025-03-19 · sparklemotion/nokogiri to unblock a major ruby-saml release.
- Release v1.18.6 / 2025-03-24 · sparklemotion/nokogiri
- Release v1.18.7 / 2025-03-31 · sparklemotion/nokogiri
Contributions
- Fixed a JRuby namespace bug when creating a new element: fix(jruby): create_element should not set the doc's default ns by flavorjones · Pull Request #3463 · sparklemotion/nokogiri
- Removed the JRuby implementation's namespace cache, which was causing subtle issues: fix(java): stop using the cache for Node#namespace_definitions by flavorjones · Pull Request #3460 · sparklemotion/nokogiri
- Fixed a Builder namespace issue: fix: XML::Builder namespace stitching by flavorjones · Pull Request #3461 · sparklemotion/nokogiri
- Added downstream ruby-saml testing for JRuby: ci: add jruby coverage for ruby-saml by flavorjones · Pull Request #3465 · sparklemotion/nokogiri
- Updated vendored libxslt to 1.1.43: dep: update libxslt to 1.1.43 (v1.18.x branch) by flavorjones · Pull Request #3467 · sparklemotion/nokogiri
- Updated vendored libxml to v2.13.7: dep: bump libxml2 to 2.13.7 by flavorjones · Pull Request #3494 · sparklemotion/nokogiri
- Fixed a JRuby regression fix(jruby): Node#attribute in HTML documents by flavorjones · Pull Request #3490 · sparklemotion/nokogiri.
- Work in progress:
- Another subtle namespace issue, which needs more work to not break downstream projects: fix(cruby): reconciliation of ns on unparented nodes by flavorjones · Pull Request #3462 · sparklemotion/nokogiri
- Simplifying library linking on windows: IN_LIBXML no longer needed for windows linkage by flavorjones · Pull Request #3475 · sparklemotion/nokogiri
- An API improvement: feat: Implement Document#parent, which always returns nil by flavorjones · Pull Request #3481 · sparklemotion/nokogiri
- Updating vendored libxml to v2.14.1, which promises to be a nice performance improvement, particularly for XPath queries: dep: update packaged libxml2 to v2.14.1 by flavorjones · Pull Request #3499 · sparklemotion/nokogiri
Reviews and user help
- Reviewed some great work by Johnny Shields to fix a JRuby issue that was blocking a ruby-saml release Fix MRI Ruby vs. JRuby XML child namespace output differences by johnnyshields · Pull Request #3456 · sparklemotion/nokogiri
- Charlie Nutter contributed:
- Documentation improvements from Burdette and Anton:
- Doc for Node#add_child by BurdetteLamar · Pull Request #3453 · sparklemotion/nokogiri
- Tweaks for Node subclass list by BurdetteLamar · Pull Request #3471 · sparklemotion/nokogiri
- doc: Nokogiri::XML::Node.traverse is recursively called on children first by AntonKhorev · Pull Request #3480 · sparklemotion/nokogiri
tailwindcss-rails
Releases
- Release v4.2.1 / 2025-03-19 · rails/tailwindcss-rails
- Release v4.2.2 / 2025-04-05 · rails/tailwindcss-rails
Contributions
- Fixing failing upgrade tests in CI. After two dead ends, finally figured it out in ci: use the latest @tailwindcss/upgrade package by flavorjones · Pull Request #529 · rails/tailwindcss-rails.
- Improve troubleshooting when things go wrong: Improve the state of debugging when things go wrong by flavorjones · Pull Request #530 · rails/tailwindcss-rails
Reviews and user help
- Some more contributions from Patricio to improve the scaffold templates:
- Fix hardcoded attribute in form by patriciomacadden · Pull Request #513 · rails/tailwindcss-rails
- add accept_confirm if turbo is available so system tests don't fail. fixes #519 by patriciomacadden · Pull Request #520 · rails/tailwindcss-rails
- Simplify markup by patriciomacadden · Pull Request #523 · rails/tailwindcss-rails
- Spent time focusing on some user issues, including:
- Reviewed a proposal for engine support: Implement Engines support by bopm · Pull Request #484 · rails/tailwindcss-rails
tailwindcss-ruby
Releases
- Release v4.0.14 · flavorjones/tailwindcss-ruby
- Release v4.0.15 · flavorjones/tailwindcss-ruby
- Release v4.0.16 · flavorjones/tailwindcss-ruby
- Release v4.0.17 · flavorjones/tailwindcss-ruby
- Release v4.1.3 · flavorjones/tailwindcss-ruby
Contributions
- Cleaning up the CI workflows:
sqlite3-ruby
Reviews and user help
- Investigated a report of a performance regression, which I narrowed down to upstream changes in sqlite v3.46.0. Unexpected regression in query plan between 1.7.3 and 2.0.0 · Issue #619 · sparklemotion/sqlite3-ruby
Rails
Contributions
- Fixed a longstanding issue with the sqlite3 adapter's handling of "Infinity": fix: sqlite3 adapter should quote Infinity and NaN by flavorjones · Pull Request #54735 · rails/rails
mini_portile
Contributions
- Fixing Windows CI failures: ci: windows images no longer support older cmake versions by flavorjones · Pull Request #157 · flavorjones/mini_portile
Thank you for reading and thank you for your support!
❤️❤️❤️,
-mike