Status report

[Edward K. Ream]

PR #4183 contains a few final tweaks to Leo's colorizer. It is now in the "devel" branch.

Issue #4161 resulted in a thorough review of all of Leo's documentation! The revised documents are live on Leo's website. LeoDocs.leo is up to date on "devel".

Issue #4166 (rewrite the syntax coloring docs) remains open. It must document the changes made by PR #4182.

Summary

I plan no more changes to Leo's colorizer. But please continue testing!

I'll soon finish updating Leo's documentation. These changes will immediately become live on Leo's website.

Edward

[Edward K. Ream]

On Saturday, November 16, 2024 at 8:08:29 AM UTC-6 Edward K. Ream wrote:

I'll soon finish updating Leo's documentation. These changes will immediately become live on Leo's website.

Done. See this page in particular.

The key new constraint on the colorizer is this:

     Leo’s core must never call redraw!

For the first time ever, Leo's colorizer satisfies this constraint. This is an epic milestone.

Edward

[Edward K. Ream]

The key new constraint on the colorizer is this:

     Leo’s core must never call recolor!

Oops, I actually said "redraw" instead of "recolor"

The first comment of PR #4182 now mentions this constraint.

This insight resolves (as much as possible) the mystery surrounding the bug.

We'll never know in detail why the qsh hid lines, but we now know the root cause:

Leo's core called jedit.colorizer, which in turn called jedit.recolor.

Oops! Leo's core called jedit.recolor indirectly!

Edward