Reword pywal mentions across remaining bread apps and docs #34
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Comment/doc-only references that don't affect behaviour (none of these read ~/.cache/wal directly). Do opportunistically alongside each app's next bread-theme bump:
Part of epic PYWAL.
All five wording PRs are open, one per repo:
Skipped per scope: bread-ecosystem (explicitly out of scope, handled by the bread-theme agent). breadway.dev checked and skipped: it is a superseded old Zola site (repo has an existing dirty working tree with unrelated uncommitted changes too), the live homepage is ~/Projects/Portfolio, not this repo.
breadhelp's packaging/src copy mentioned in the issue is gitignored build output, not tracked in git, so nothing to change there beyond src/theme.rs.
All changes are comment/doc wording only. Verified with cargo build (all succeed) and cargo fmt --check (each repo has pre-existing formatting diffs in files I did not touch, left alone). Worktrees left in place at ~/Projects/-pywal for review.
Review of the five wording PRs:
cargo fmt --checkfailure at breadlock/src/state.rs:251, and origin/main has the same diff. The wording is fine and it can merge once main's fmt is fixed.The bread-ecosystem parts of the issue (README, upgrade.md, registry) are in bread-ecosystem#36.
Moved to todo for the breadbar fix. Rework happens on the existing fix/pywal-wording branches, and the PRs stay open. Do not open second PRs.
Updated all 3 remaining open wording PRs per review:
Note: both bread's and breadlock's CI runners are still showing "Failing after Xs" (single-digit to ~20s) after these pushes, too fast to be a real clippy/test run, consistent with the open hestia CI-runner issue rather than anything in these commits. Local verification is clean and matches (or exceeds, for breadlock which actually runs fmt --check in CI) what each repo's CI runs.
breadbox#10 and breadhelp#3 are already merged; removed their worktrees (~/Projects/breadbox-pywal, ~/Projects/breadhelp-pywal) and deleted the local fix/pywal-wording branches in both repos.
PRs: Breadway/breadbar#25, Breadway/bread#6, Breadway/breadlock#9
Second-round review:
breadd/tests/ipc_integration.rstests cannot findbread-module-host, becauseresolve_module_host_binary(breadd/src/module_host.rs:377-387) looks next totarget/debug/deps/, nottarget/debug/. It fails the same way on the unrelated run 95. bread is also release-sensitive, so the user lands it.pacman: target not found: clippy(.forgejo/workflows/ci.yml:21; Arch folded clippy/rustfmt intorust). main is red the same way.The earlier "hestia runner" diagnosis does not hold for either repo. Both need a small separate CI fix in their own repo, and then these PRs can merge. ECO-34 stays in review until they do.
Round-3 CI fixes pushed to unblock the two remaining wording PRs:
cargo build -p bread-module-host --lockedonce before spawning anything. Reproduced this repo's actual CI container to verify. CI is now green: https://git.breadway.dev/Breadway/bread/actions/runs/99Both are release-sensitive repos (bread, breadlock), so merge stays with the user. Staying in review.
Post-merge review of the round-3 CI fixes. bread#6: acceptable, but a follow-up fix PR is recommended (Breadway/bread#6 (comment)). module_host_sandbox.rs never calls ensure_module_host_built and only passes because of test ordering. Run alone on a fresh target dir, it fails and falls back to the installed ~/.local/bin/bread-module-host through PATH. The nested build also ignores --release, --target-dir and --target. breadlock#9: fine, since as_chunks_mut is stable since 1.88 and all toolchains in use are newer (Breadway/breadlock#9 (comment)). Optional follow-up: add rust-version = "1.88".