Drupal 9 release scenarios, the first beta deadline in 6 weeks and how you can help!

As Dries Buytaert explained in his Plan for Drupal 9 post at the end of 2018 (emphasis mine):

Drupal 8's biggest dependency is Symfony 3, which has an end-of-life date in November 2021. This means that after November 2021, security bugs in Symfony 3 will not get fixed. Therefore, we have to end-of-life Drupal 8 no later than November 2021. Or put differently, by November 2021, everyone should be on Drupal 9.

Working backwards from November 2021, we'd like to give site owners at least one year to upgrade from Drupal 8 to Drupal 9. While we could release Drupal 9 in December 2020, we decided it was better to try to release Drupal 9 on June 3, 2020. This gives site owners 18 months to upgrade. Plus, it also gives the Drupal core contributors an extra buffer in case we can't finish Drupal 9 in time for a summer release.

Here we are 14 months later and while most people took the June 3 release date and took it for granted, it is still not guaranteed! However you can help in various ways to make it much more likely!

Late last fall, we focused on defining what it means if we cannot make the June 3, 2020 release despite our best efforts and what is an early indicator that tells us we are going to miss it. First of all, that meant defining requirements for the first alpha release and requirements for the first beta (API complete) release. Also we needed to set some expectations as to when do we want to see the API-complete beta release to give enough time for the ecosystem to test it and find important problems in time. Based on how soon the beta requirements are met, there are three release scenarios and the best case ending in the June 3, 2020 release date for Drupal 9 has a beta deadline in 6 weeks! Yes, 42 days!

Alpha requirements simplified

The key requirements for the first alpha release are simple. We wanted to update the key dependencies: Symfony to version 4.4 and Twig to version 2, as well as remove frontend polyfills that were not needed and remove most of jQuery UI (which were already deprecated in Drupal 8). This gets our most important dependencies up to shape to what will be in Drupal 9. We also made it possible for contributed projects to depend on Drupal 8 and 9 at the same time, so they will not need to branch for Drupal 9 support.

There are two outstanding things for the Drupal 9 alpha:

  1. Drupal.org does not yet have an automated packaging pipeline that conforms to all the recent composer related improvements and therefore making core releases is error prone. I don't believe you can help with this at this time, the Drupal Association is hard at work on this.
  2. Drupal core should use a major version agnostic update feed for projects which is already being provided by Drupal.org but the core code to consume it is still in the works. While this is actively being worked on, reviews are always helpful. This will make sure Drupal 9's Update Status gets only contributed projects that are actually Drupal 9 compatible, while contributed modules will not need to establish a Drupal 9 branch.

Beta requirements simplified

The beta requirements are a bit more complicated and longer of course because we are looking at being API complete here. Once again, for the June 3, 2020 release date, we need these done in 6 weeks! The issue lists must haves and should haves, however the should have issues should be considered must-have for the June 3, 2020 release date and would only be reconsidered later if that date cannot be met. Here is a simplified rundown of the beta requirements:

  1. We want to keep dependencies up to date. There is no concrete pressing issue here at the moment that I know, but this really depends on how our dependencies evolve.
  2. We'd like to remove all the deprecated APIs themselves. Last year I built a graph to track this, and it shows nicely that we are down to half of them remaining (yay!), but still quite enough to deal with. There are various outstanding issues you can help with here.
  3. We want to make sure people can update to Drupal 9 from Drupal 8 by resolving critical upgrade path bugs. If you cannot update to a later version of Drupal 8 due to some critical bug, then you will be stuck on your version of Drupal 8. Not good. These include views, layout_discovery, taxonomy, menu_content, etc. related issues. All of them need help. If you are on an older version and can reproduce the problems, that is useful. If you have experience in these areas, your input would be useful.
  4. It will only be possible to update to Drupal 9 from Drupal 8.8 or 8.9, so all older update paths and their tests should be removed. Older versions of Drupal 8 will themselves be unsupported already at the time of Drupal 9's release. This issue is getting close but needs reviews.
  5. No new security or data integrity issues should be in Drupal 9. If there are any, they should be resolved. I don't know of any issues at the moment here.
  6. The API should be complete. There are no critical API additions or changes that I know of at the moment in this general category.
  7. We want to make sure people can migrate from Drupal 6/7 to Drupal 9. This needs the remaining multilingual migration paths to go stable. This is an area where we posted several call to actions, but still need your help. There are proposed migration paths for node translations and entity translations that respect revisions but they need at least code reviews to make sure they are good. Otherwise if you had content translations with revisions, the migration will not be correct. Without that, multilingual migrations will not go stable.
  8. PHP requirements should be finalised. It is likely at this time that Drupal 9 will require PHP 7.3 that is being worked on currently and could use a review.
  9. Database requirements should be finalised in terms of MySQL/MariaDB/Percona and PostgreSQL. Both issues need data as to which distributions and hosts support certain versions.
  10. The right security update information should be provided for users taking the one year support cycle and long term support of the last Drupal 8 release. This could also use reviews.
  11. We should put Drupal's base theme on a track so that it can evolve in Drupal 9 finally. This involves creating a new stable9 theme and decoupling the core themes from Classy. Various issues to help with here.
  12. Drupal.org should support multi-core compatibility eg. on project pages, localize.drupal.org, etc. This work is currently deprioritised by the Drupal Association due to the focus on the packaging pipeline blocker that I listed first. Once this gets attention, it will likely uncover core issues to resolve as well.

The two areas that receive the least attention at the moment are upgrade path blockers and the stability of the multilingual migration path, so those two are the most pressing where we need your help!

While the above is a complete rundown of the current beta requirements, it may change later on, so refer to the beta requirements issue later on for up to date information.

What happens if all the above are not done by end of February (in six weeks)

If all goes well, with your help, we'll be done with all the above in six weeks. If that does not work out, we have a plan B and a plan C. Here is how those options unfold. If beta requirements are only done two months later by end of April, then Drupal 9's first beta will be released on the first week of May and Drupal 9 is to be released on August 5, 2020. If the beta requirements are only done by end of August (four more months later), than the first beta will be released than with a Drupal 9 release date of December 2, 2020. In this case a Drupal 8.10 may also be released if needed. These dates are spelled out well in the Drupal core release cycle overview. I created this visual to help understand the alternate timelines:

Drupal 9 release scenarios visualised

While I think it is reassuring that we have plans for what happens if our initial date targets don't work out, unfortunately the end of life date at the end of next year for Drupal 8 is not movable because it is based off of Symfony 3's end of life. So the sooner we can make Drupal 9 happen (while meeting the upgrade and stability requirements) the better. What are you going to work on to help?

Helping with contributed projects

Based on the PHP deprecations our tools can identify, 43% of contributed projects would only need info.yml file updates to be Drupal 9 compatible. An additional 41% of the remaining projects have only issues that are resolvable now (even while keeping support for Drupal 8.7). Solving those anytime between now and Drupal 9's release (whenever it is) would put us to almost six thousand contributed projects compatible with Drupal 9 on day one! While that is a very idealistic number, helping with contributed projects is nonetheless a great avenue to contribute to Drupal 9 readiness. Help at the Drupal Global Contribution Weekend at end of next week or anytime before and after! I prepared a quickstart guide for this occasion.

Drupal 9 compatibility contribution quickstart guide for Drupal Global Contribution Weekend 2020

This year Drupal Global Contribution Weekend is on January 24-26, 2020 with such varied locations as Delhi, Novosibirsk, Ghent, Frankfurt, Milan, Zurich, Lutsk, London (on two continents!), Boston, Minneapolis, etc. Wow! It is truly a global gathering! With Drupal 9 planned to be released later this year, what better to focus on, than making drupal.org projects Drupal 9 ready?

To help you do that I went in and updated my open source State of Drupal 9 talk. People can use this to present at any location to get people up to speed about Drupal 9. If you need a video recording of it, there is one from DrupalCamp Belarus in May 2019. While the content got slightly updated since then, the recording should help get it.

After or instead of presenting that session, I thought a quickstart guide would be really useful to help people get started with contributing. While this looks like a colorful guide you would print, it is actually full of useful links (some to my earlier blog posts for details), so I suggest you use it in digital form.

What are you planning to do for Drupal Global Contribution Weekend this year?

Introducing the Drupal 9 deprecation status tool on dev.acquia.com

You've probably seen the stats Dries Buytaert cites in his State of Drupal keynotes, and some of my earlier blog posts where I call for deprecated API removal automation and fixing drupal_set_message() calls all around in particular. You might have been wondering where is all that data coming from and how can you get a better sense of where Drupal projects stand and how can you help make progress. Well, I had a lot of fun recently building a new tool (with help from Matthew Grasmick, Shannon Vettes and Angie Byron) for dev.acquia.com that does exactly that! Check out this demo and head over to dev.acquia.com/drupal9 to try it out.

Let's automate deprecated API use removal from Drupal projects!

In his State of Drupal keynote at DrupalCon Amsterdam, Dries Buytaert showed once again some tools to use to prepare for Drupal 9 including the Upgrade Status module. To me the process is even more interesting than the tools, because it is entirely different from the last upgrade. As I wrote last week, you now make a lot of incremental improvements on your existing Drupal 8 site that makes the gap to Drupal 9 significantly smaller.

It is a new mindset to look at your Drupal 8 site to improve in preparation for Drupal 9 and we have tools to identify those improvements to make. However Dries also mentioned that we are not quite there to automate those improvements. Back in May Dezső Biczó of Pronovix built out a proof of concept integration with Rector that implements a sample set of refactors, including changes for global constants and best effort solutions to some EntityManager uses, a UrlGeneratorTrait and a drupal_set_message() rector. While the extent of impact of the global constant rectors are not yet known due to limitations in our tools not finding them yet, the rest of the implemented rectors are definitely tapping into the top list of deprecated APIs.

Unfortunately slightly after he posted his blog post about the proof of concept, Dezső did not have time for this project anymore. I think this tool could be of huge help in removing deprecated API use in your projects and thus making them more modern Drupal 8 projects (while being more if not already entirely compatible with Drupal 9). However, we need contributors to cover more of the deprecated APIs, especially around file_*() and db_*() functions. If we can figure out rectors for most of the top 15 errors (and Dezső already did some of them), we could cover over half of all deprecated API use in contributed projects:

Donut chart of top 15 usages of deprecated APIs on drupal.org

To top that off, I also think a simplytest.me style online service would be very useful to run drupal8-rector on your drupal.org project and suggest a patch to apply to remove deprecated APIs. Even better if it allows custom code to be uploaded so people can use it for that too. The more we can automate these tasks the easier the transition of Drupal 8 to 9 will be.

Submit issues and pull requests in the drupal8-rector project to improve API coverage. Look for me on Drupal slack if you get stuck and I'll try to help. I'd also love to talk to you if you want to set up an automated service. Let's do this!

The dramatic shift in how a Drupal upgrade is now done in Drupal 8 for Drupal 9

I've had various deep discussions with contributed module maintainers recently about their process to update code to Drupal 9 and one point struck me. We are so attached to "Make it ready for Drupal 9" that a key point of the message may be lost. Check out this section of the State of Drupal keynote from DrupalCon Amsterdam 2019 where Dries Buytaert showcases Johanna's relatively simple site that she prepares for the Drupal 9 upgrade entirely in Drupal 8. Notice that she does all the steps in Drupal 8 other than the final Drupal 9 upgrade itself:

This is the base principle of the process towards Drupal 9, making your Drupal 8 site better and more prepared, so the move to Drupal 9 itself at the end is a relatively small step and you got a better Drupal 8 site in the meantime. You are not jumping over the fence all at once, but in gradual steps. I thought a comparison with Drupal 6 to 7, 7 to 8 and 8 to 9 would help, since people may have assumptions or prior experiences with those, so its worth looking at how our new process compares to the two previous transitions.