| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
| |
The race detector found my own new flag: Apply writes it on a worker
and Close reads it from the main loop. The test was polling a plain
field too, where the real window hears about the apply on the main
loop, which orders the writes.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Five of the Settings window's controls - the sort order, the three
column toggles and the preview height - were read when it built its new
preferences but never connected to anything, so changing them did
nothing until some other control happened to fire, and then they all
landed at once out of nowhere. Every control is connected now.
The same closure composed a whole Prefs from its own widgets, which
wrote the remembered divider positions back as zeros: dragging the
panes to taste and then ticking any checkbox threw them away. A
settings change is now applied over the preferences as they are, by
model.Prefs.WithDisplay - which is where it can be tested, and is.
"Keep this copy, replace the other" wrote a Displaces straight into the
chain. internal/plan refuses to build two steps that displace one path,
because the second destroys what the first put there; the window went
round that code, so choosing it for two duplicates of one file left
both rows saying "done" with the first file in the Trash. It now
refuses the second, naming the file that has the place.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Close releases the directory lock and closes the log. The window could
reach it while an apply was still running - saving rules, saving
settings, adding a directory, or closing the window - and the engine
then went on moving files with the log shut underneath: a file moved
that no krino undo can see, the rest of the plan silently abandoned,
and the directory unlocked while krino was still working in it.
PlanTab and UndoTab refuse to close while their apply is in flight
(model.ErrApplying), the window's close request and reloadEngine honour
the refusal instead of ignoring it, and the tabs and Settings are greyed
out for the duration so a button that cannot work says so by being
unavailable rather than by an error afterwards.
The test starts an apply, calls Close from another goroutine while it
is in flight, and requires the refusal.
|
| |
|
|
|
|
|
|
|
|
|
| |
So that go install can find it. cgit serves no go-import meta tag, so
nginx answers a ?go-get=1 request for /NAME with one pointing at
https://git.labunix.xyz/NAME.git; the alternative was carrying a .git
suffix through every import line forever.
One sed over the imports, both go.mod files, and the dependency gate's
whitelist. Nothing else depends on the path. go install works from the
next tag, the first release whose go.mod carries it.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
About 340 comments cited the development process: task and plan
numbers, fix waves, rulings, reviewers, and the author in the third
person with a date. None of that exists outside the work itself, so to
a reader it pointed at nothing. Each one now states the engineering
reason it was standing in front of; where a comment was provenance and
nothing else, it is gone.
References to docs/design.md and docs/gui-design.md by section stay:
both ship with the repository. The design documents lose their
amendment diaries - CHANGELOG.md is that record - and the GUI's says
plainly that the window has gone further than the document.
Only comments changed. Every .go file was parsed and its code printed
with comments stripped, before and after: the two hashes are identical
across all 175 files.
|
| | |
|
| |
|
|
| |
explanation
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
