The comment property on the node property is being refactored into a field. Let's pretend -- just because it's a great example not because we want to postpone it -- this happens during Drupal 9! Modules using $node->comment break but modules using $node->get('comment') don't. Modules running SQL queries against {node}.comment break but modules running entity queries with entity_query('node')->condition('comment', 1) do not.

The Dependency Injection Container is now written to disk. When there is no container file available then the DIC is compiled and if the storage backend is writeable then it is written to disk. By default this is written in a tricky (slightly slower but more secure) way into the public files directory.

The EntityFieldQuery rewrite is progressing well, finally had a good review from fago. The following patch had a 36KB interdiff on a 120KB patch. Such is life in Drupal core: you try to write something to the best of your abilities, someone smarter comes along, you adjust your code. Now your code and yourself is better if you take the effort to learn from it.

Closures seem like big magic but in PHP they are just syntactic sugar over the following construct:

Now, closures use the following syntax:
<?php
function get_add_printer($add) {
return function ($x) use ($add) {
print $x + $add;
print "\n";
};
}

The system table removal patch went through some significant changes thanks to alexpott and sun and is now ready. That's great! I got busy with creating a proper version for the temporary key-value store patch: this is a requirement for Views in Core and also would allow for a nice entity preview workflow if someone would write it. EntityFieldQuery has been rewritten, there are a six more tests to fix, doesn't seem like a big problem, it still needs a lot of documentation and new tests but bojanz is back from vacation and promised tests.

Google Code-In (GCI) is an incredible program in which junior high and high school students are encouraged to participate in an exciting open source project, like Drupal! Open source projects participating in GCI create a list of short-term, high-impact tasks for students to work on.

Over the past few years that we participated in GCI we attracted a number of new, talented young developers to the Drupal project who are still active community members today.

As promised I am trying to keep a weekly schedule with these posts. The system table removal issue with the help of Alex Pott is almost ready although it has been pushed into process hell just before I started writing this so I am again worried about it given how tight our schedule is now. The date table removal issue got really far thanks to cosmicdreams and KarenS, I will finish if necessary but it's also relatively close. The code around it is not the prettiest to say the least.

I will try to keep a weekly schedule but nearly two weeks passed since the previous post. These days I am still working on the system table removal. The key-value store patch it relies on is now fully ready (edit: it just got committed) and the system table to CMI conversion patch is now at its three major revision.

I can't let the Drupal 7 User IDs - Not Consecutive post alone just like that. A couple of emphasized "they" referring to some core contributors is really deeply disturbing because it furthers a false argument: there are "core developers" and there are "others". Everyone can write core patches and you are more than welcome to! As for this issue, it's not about liking or disliking auto-increment, it's about real problems with auto-increment, one of the biggest is not even listed because it's implied to be known: in D6 the presence of a 0 in an autoincrement column (the user id) causing massive import headaches. Oh and please don't use the "solution" in that blog post, it has a lot of race conditions... and why I didn't comment there? Because comments are Facebook only and I don't use Facebook.

It was more than two years ago, in the predictions for 2010 thread that I crossed words with kaakuu over Drupal and unicode text search. I told kaakuu to file a usable bug report -- and to be honest I didn't expect one. Much to my surprise he did participate in an issue and filed an exceptionally useful comment. Eight days later, the fix I written got committed, fixing search for 200 million+ users or who knows how many more were affected. And I learned a lot about Unicode. Why I am writing only now, I do not know but I have told this story over and over and such helpful bug reports are still very rare. Will you file the next one I will talk about for years?