A week of ADHD-driven running around changelog and other new Ruby release-related tasks!

The plan for today:

  • finalize adjusting WeakMap/WeakKeyMap docs and create a PR;
  • put what was discovered/understood into the corresponding changelog entries (that’s the main driver of this adventure, after all!)
  • publish a blog post/email summarizing the first seven days and linking to this rough diary;
  • (maybe) also handle Module#set_temporary_name docs/changelog entry, while my memory of the investigation is still fresh.

Also, before we start, there is some new information that to some extent changes the scope of my work:

  • today, a development meeting discussion was published, leading to adjustments in some outstanding tickets on the tracker; some of the messages are related to those “new syntaxes” I described in the recent series:
    • anonymous block parameters: after a lot of discussions and pressure, Matz finally agreed to have something more nicely-looking than _1: the name it would be warned in corresponding context in Ruby 3.3, and introduced as an alternative name for anonymous block parameter in 3.4 (it would be done in a gentle way, so, say, RSpec’s API wouldn’t be broken);
    • endless methods: the core team is not persuaded that the “quirks” with parsing order are deserving to be fixed, the ticket is closed; I am still not giving up hope and will try to reiterate on the problem, but only after 3.3 release;
  • A lot of new things have happened in NEWS.md since I initiated the changelog, so I’ll need to add more entries into my files— but not today.

So… Here we go at WeakKeyMap docs:

As I explained yesterday, to check if references to other docs—here, Object#eql?—rendered correctly, I need to include more files in rdoc command, namely

rdoc weakmap.c object.c -o tmp/doc

And finally… Here we go.

A half-an-hour later, a small commit to the changelog is ready, too: c716b8e. The texts are very rough for now, but at least, I now can put a mental checkbox that some features alreadys have drafty coverage. Once they all would have at least that, I would be able to move to reorganizing and polishing.

After that, I am switching to house-keeping for my “self-publishing”: move rough day**.md files into a separate folder, add some navigatin to them, and write a blog post covering this first week.

I would probably be more efficient working on changelog in silence, as I do every year… Or maybe not, because commitment to everyday diary makes me work on it every single day instead of depending on the mood, physical and mental state. It is nice to have something organizing when no external factors push you to work on a hobby project (last year, I finally made myself finish it by February!)