Skip to content

Cleanup: stray autosave artifact, dead code, placeholder header - #5

Open
yottanami wants to merge 1 commit into
masterfrom
agent/22-cleanup
Open

Cleanup: stray autosave artifact, dead code, placeholder header#5
yottanami wants to merge 1 commit into
masterfrom
agent/22-cleanup

Conversation

@yottanami

Copy link
Copy Markdown
Owner

Summary

Three independent, trivial cleanups bundled as one PR (each an unrelated one-off, per the issue):

  • plugins/devel/#core.el# deleted — a stale Emacs autosave snapshot, predating the current core.el's Ruby LSP block, with a syntax typo ((setq js-indent-level 2)s, a stray trailing s). On why it slipped past .gitignore's #*# rule despite that rule already existing: I checked — the rule (815c818, 2025-09-18) predates the file's own commit (89a4842, 2026-06-26) by about nine months, so it's not a case of the rule missing it in time. An exact-path git add on an already-named file still adds it even if it matches a gitignore pattern; only wildcard/directory adds silently respect .gitignore. Checked the rest of the tree (git ls-files | grep -E '(^|/)(#.*#|.*~|\.#.*)$') — no other stray autosave/backup files are tracked.
  • plugins/wm/core.el's header had unfilled template boilerplate (Your Name <your.email@example.com>, a placeholder URL) — filled in with yottanami <yottanami@gnu.org> (the identity every commit to this repo since 2025-09-18 actually uses — checked git log --format='%an <%ae>') and https://github.com/yottanami/noteditor (matches README.org and nix/package.nix's meta.homepage).
  • noteditor-user.el: removed ~50 lines of dead, fully-commented-out code referencing use-flags/when-wm/noteditor/flags — grepped the whole tree, none of those exist anywhere in the current codebase. Left the one remaining commented block (loading an optional ~/.noteditor-user.el) alone since it doesn't reference those removed macros and reads as an intentional usage example rather than dead code from a removed feature.

Closes Men-in-Black-5/ideas#22

Test plan

  • Diffed the stray autosave file against current core.el to confirm it's genuinely stale, not something worth preserving
  • Grepped for use-flags/when-wm/noteditor/flags across the whole tree — confirmed dead before removing
  • Grepped for other stray #*#/*~/.#* tracked files — none found
  • Parenthesis-balance checks on both edited .el files — balanced
  • nix flake check / a live build — no nix/emacs binary in my environment, needs a real check

@yottanami

Copy link
Copy Markdown
Owner Author

Reviewed, and re-derived each factual claim independently rather than taking the PR's word for it:

  • Stray autosave file: confirmed the .gitignore #*# rule (commit 815c818, 2025-09-18) predates the stray file's own commit (89a4842, 2026-06-26) by ~9 months — not a case of the rule missing it in time. Also empirically verified the git mechanics claim in a scratch repo: git add on an exact, explicitly-named path does silently stage a file even when it matches .gitignore (no -f needed) — that's genuinely how it slipped through. Confirmed the file itself has the claimed typo ((setq js-indent-level 2)s, line 54) and is a stale duplicate. Also ran the same tracked-stray-file grep across the tree myself — none left.
  • Header fill-in: checked full git log history myself — every real commit (excluding the agent's own mib-bot commits) from 2024-04-28 onward uses yottanami <yottanami@gnu.org>, an even longer consistent stretch than the PR claimed. URL matches both README.org's clone URL and nix/package.nix's meta.homepage exactly.
  • Dead code removal: grepped the whole tree for use-flags/when-wm/noteditor/flags myself — zero matches, confirming it's genuinely dead. Agree the remaining commented-out ~/.noteditor-user.el load block is a legitimate usage example, not part of the same dead feature, and correctly left alone.

All three edited files balance. Bundling these three as one PR matches ideas#22's own explicit scoping ("bundled as one task since each item is trivial and independent"), so no unrelated-changes concern despite them being unrelated to each other.

Same caveat as the rest of the batch: no agent can run nix flake check or load this, so it needs your confirmation before merge — everything staticly checkable is solid.

- Deleted plugins/devel/#core.el# -- a stale Emacs autosave snapshot
  from before core.el gained its Ruby LSP block, with a syntax typo
  ("(setq js-indent-level 2)s"). Despite #*# already being in
  .gitignore, it was committed anyway (89a4842, 2026-06-26) -- that
  rule predates the commit by nine months, so this wasn't a case of
  the rule missing it; an exact-path `git add` on an ignored file
  still adds it, only wildcard/directory adds respect .gitignore.
  No other tracked files match the #*#/*~/.#.* patterns.

- plugins/wm/core.el's header had unfilled template boilerplate
  ("Your Name <your.email@example.com>", a placeholder URL). Filled
  in with the identity every commit since 2025-09-18 actually uses
  (yottanami <yottanami@gnu.org>) and the project's real URL, per
  README.org and nix/package.nix's meta.homepage.

- Removed ~50 lines of dead, fully-commented-out code in
  noteditor-user.el referencing use-flags/when-wm/noteditor/flags,
  none of which exist anywhere in the current codebase (grepped the
  whole tree). Left the one commented block that doesn't reference
  those macros (loading an optional ~/.noteditor-user.el) -- reads as
  an intentional usage example, not dead code from a removed feature.
@yottanami

Copy link
Copy Markdown
Owner Author

@yottanami This one's waiting on you too — no agent can run nix flake check or load the config, so per policy it holds for your confirmation before merge (see the review comment above for what was statically verified: autosave artifact removal, dead code removal, header fill-in). Steps:

  1. Run nix flake check (or just load this branch's config in Emacs) and confirm nothing broke.
  2. Merge if it looks good.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant