aboutsummaryrefslogtreecommitdiff
path: root/modules
Commit message (Collapse)AuthorAgeFilesLines
* refactor(ui): remove buffer-state coloring of the buffer nameCraig Jennings44 hours3-51/+5
| | | | The modeline colored the buffer name by write and modification state (red read-only, green modified, gold overwrite) through a classifier in user-constants.el. I removed it, the same way the matching cursor coloring was removed earlier for being more confusing than useful. The classifier had no other live user, so it and its two test files go with it. The buffer name now renders in the normal mode-line color.
* fix(term): pin ghostel to pre-regression 0.33.0 to stop the crashCraig Jennings2 days1-0/+15
| | | | | | | | ghostel 0.35.0-0.35.2 hard-crash the whole Emacs process when a terminal buffer is displayed. The native PTY path was reworked to spawn worker threads. On Linux/glibc a SIGSETXID handler then calls malloc while the main thread holds the arena lock, so opening an agent terminal takes the whole editor down (upstream #422, with #423 the macOS recursive-lock variant). Reproduced down to a plain M-x ghostel in a GUI frame, so it's not ai-term-specific. Hold ghostel at the 0.33.0 build (ghostel-20260604.2049), which predates that rework. :ensure is satisfied by the installed 0.33.0 directory and won't upgrade it. The use-package block carries the rationale and a do-not-upgrade guard so it isn't bumped back into the crash before upstream ships a fix. Also set ghostel-module-auto-install to download so the native module installs without the interactive prompt.
* feat(ui-navigation): split from the dashboard opens scratch, not the dashboardCraig Jennings4 days1-4/+19
| | | | C-x 2 / C-x 3 already show the dashboard in the new window while point stays put. The one dead spot was splitting from the dashboard itself, which put the dashboard in both panes. Now the new window shows *scratch* when the current buffer is the dashboard, and the dashboard everywhere else. I pulled the choice into a pure predicate (cj/--split-from-dashboard-p) and a companion helper, both tested.
* refactor(system-utils): remove the *scratch* background tintCraig Jennings4 days1-29/+2
| | | | I dropped the buffer-local face remap that lightened the *scratch* background 5% above the theme default. Scratch now uses the plain theme background like every other buffer. The startup hook still moves the cursor to end-of-scratch. Only the tint call, its two helpers, the defcustom, the color require, and the now-orphaned test go.
* feat(keybindings): mirror the C-; command family under C-c ;Craig Jennings4 days1-0/+4
| | | | C-; is GUI-only: terminals can't encode Control-semicolon, so the whole custom command family (calendar, AI, Slack, org, pearl, jump, and the rest) was unreachable in a terminal frame (emacs -nw, emacsclient -nw, or Emacs inside vterm/tmux). I bound the single cj/custom-keymap under C-c ; alongside C-;, so the same leaf keys reach the identical map in both GUI and TTY with no relearning and no per-module edits. C-c is the standard user prefix and always TTY-encodable. I audited every leaf key in the family and they're all TTY-safe (letters, digits, punctuation, SPC, and arrow keys), so nothing needed remapping.
* fix(mail): theme mu4e buffers via a shared font-lock exclusionCraig Jennings4 days3-1/+31
| | | | mu4e paints its header lines, main menu, and view headers with manual `face' text properties. Global font-lock stripped them, so the buffers rendered unthemed, the same failure the dashboard hit. I extracted the dashboard's one-off exclusion into a shared, additive cj/exclude-from-global-font-lock helper in system-lib and repointed the dashboard to it. Then I excluded mu4e-headers-mode, mu4e-main-mode, and mu4e-view-mode. The view body renders through gnus's own washing rather than font-lock, so excluding it is safe.
* feat(dashboard): per-filetype icons on the list itemsCraig Jennings5 days1-0/+1
| | | | I enabled dashboard-set-file-icons so the project, bookmark, and recent-file rows show nerd-icons file icons, colored per filetype. They render now that the dashboard is out of global font-lock, which was stripping the icon faces too. Bookmarks fall back to a generic icon since there's no filetype to map.
* fix(dashboard): exclude dashboard-mode from global font-lockCraig Jennings5 days1-0/+9
| | | | | | The dashboard banner title and section headings render in the default face instead of their theme colors. The faces are defined correctly. Global font-lock fontifies the dashboard buffer and strips the `face' text properties dashboard sets by hand: it owns `face' and clears props it didn't apply. Items and the navigator survive because their color rides a dashboard-items-face overlay, which font-lock leaves alone. Excluding dashboard-mode from global font-lock keeps the banner and heading faces. I set it at top level so it applies regardless of the use-package body.
* fix(ai-config): intern model in gptel-switch-backend (modeline hang)Craig Jennings5 days1-1/+11
| | | | | | cj/gptel-switch-backend set gptel-model to the raw completing-read string. gptel's modeline code calls symbolp on gptel-model and signals wrong-type-argument on a string, which surfaces as a redisplay hang (reachable from C-; a B). The sibling command cj/gptel-change-model already interns. This one didn't. I added a pure cj/gptel--model-to-symbol helper (mirroring cj/gptel--model-to-string) and route the model through it before the setq. The existing switch-backend test asserted the buggy string value. It now asserts a symbol plus an explicit symbolp guard.
* feat(system-utils): tint the *scratch* background a shade lighterCraig Jennings5 days1-2/+29
| | | | A buffer-local face remap lightens the *scratch* default background by cj/scratch-background-lighten percent (default 5) so it reads as the scratch buffer, applied on emacs-startup-hook. The colour math is display-dependent (verified live); the pure helper's guard contract is unit-tested.
* fix(vertico): Page-Up/Down scroll the candidate list, not historyCraig Jennings5 days1-7/+11
| | | | <next>/<prior> weren't bound in vertico-map, so in a long completing-read they fell through and selected-then-dismissed the list. Bind them to vertico-scroll-up / vertico-scroll-down.
* fix(custom-comments): heavy-box interiors are valid commentsCraig Jennings5 days1-37/+47
| | | | The heavy-box empty and text lines began with a bare decoration char, so in line-comment languages (elisp, Python) C-; C h injected syntax-breaking lines. Prefix the interiors with the comment char and suffix them like cj/--comment-box does. Add the missing min-length guard so small or negative widths error cleanly instead of failing inside make-string. Updated the two characterization assertions to the corrected output.
* fix(dirvish): mark-all-visible no longer skips every other fileCraig Jennings5 days1-16/+8
| | | | dired-mark advances point itself, so the loop's extra forward-line skipped every other file (and could mark a directory). Use dired-get-filename + file-directory-p with an if/else: a marked file line advances once via dired-mark, non-file/directory lines advance manually. Replaces the regex line predicate (retired with its mock test) with a real-Dired marked-count test.
* refactor(ui-config): drop buffer-state cursor coloringCraig Jennings5 days1-45/+4
| | | | A cursor that changed color by buffer state was confusing. Remove cj/set-cursor-color-according-to-mode, its two cache defvars, and the post-command / server-after-make-frame hook registrations; the cursor now uses the theme cursor face. The cj/buffer-status-state / cj/buffer-status-color classifier stays in user-constants.el for the modeline buffer-name indicator. Delete the cursor-function integration test; keep the classifier tests.
* feat(dashboard): add a Signal launcherCraig Jennings5 days1-3/+4
| | | | Signal joins the dashboard launchers on key S (icon nf-md-message), opening cj/signel-message. Row sizes go 4-4-3-3 so Slack, Linear, and Signal share the last navigator row.
* feat(ai-term): dock the agent window by frame aspect ratioCraig Jennings5 days1-14/+22
| | | | The agent window now docks from whichever edge conserves more space, chosen at display time: right on a landscape frame, bottom on a square or portrait one, replacing the host (laptop/desktop) branch. cj/--ai-term-direction-for-aspect is the pure decision; default-size pairs the width or height fraction with it.
* fix(ui-navigation): undo-kill-buffer skips open files via equalCraig Jennings5 days1-1/+4
| | | | The visited-file filter used delq, comparing expand-file-name strings by eq, so an already-open file was never removed from the candidates (the skip logic was dead). Use delete (equal). Adds a test with the open file at the head of the list, where the eq/equal difference actually shows.
* feat(dashboard): g refreshes, Telegram moves to G, F1 refreshes on showCraig Jennings5 days1-1/+8
| | | | | - g on the dashboard now runs dashboard-refresh-buffer (the dired/magit convention) instead of opening Telegram. Telegram moves to G in the launcher table. - cj/dashboard-only (F1) refreshes before showing, so re-displaying the dashboard always lands on fresh content.
* feat(ui-navigation): C-x 2/3 show the dashboard in the new windowCraig Jennings5 days1-0/+28
| | | | Splitting with C-x 2 or C-x 3 now shows *dashboard* in the freshly created window and keeps point in the original, instead of mirroring the current buffer. cj/--split-show-buffer does the placement; cj/--dashboard-buffer fetches or opens the dashboard without disturbing windows.
* fix(eshell): flatten visual-commands, wire xterm-color correctlyCraig Jennings5 days1-14/+20
| | | | | - eshell-visual-commands is a flat string list, but add-to-list pushed the whole list as one element, so lf/ranger/htop/top never opened in a visual terminal. dolist the strings instead. (visual-subcommands/options are alists and were already correct.) - The xterm-color before-prompt hook was registered as eshell-before-prompt-hook, which use-package turned into eshell-before-prompt-hook-hook and never ran. Use the real hook name, and actually install xterm-color-filter into eshell-preoutput-filter-functions (dropping eshell-handle-ansi-color) so color output is interpreted.
* fix(prog-go): autoload gofmt so C-; f works under go-ts-modeCraig Jennings5 days1-0/+4
| | | | .go opens the built-in go-ts-mode, so nothing triggered the go-mode package: gofmt was never autoloaded (C-; f signalled void-function) and the :config setting goimports + exec-path never ran. Add :commands (gofmt) so the first format pulls go-mode and its config.
* fix(system-defaults): guard server-start and custom-file against batchCraig Jennings5 days1-3/+8
| | | | A raw module load under --batch started a server and dropped a throwaway custom-file temp on every run; the suite only passed because a testutil stubbed the server. Guard both with (unless noninteractive ...), the established pattern here. Adds a batch-load test for the custom-file guard.
* fix(auth-config): guard gpg-connect-agent, require user-constants at runtimeCraig Jennings5 days1-4/+6
| | | | | - The bare call-process to gpg-connect-agent in :config aborted init with file-missing on a machine without the binary. Guard it with cj/executable-find-or-warn. - user-constants was required only eval-when-compile, but authinfo-file is read at load time, so a standalone .elc load failed. Require it at runtime.
* fix(prog): keep electric-pair and line-number setup buffer-localCraig Jennings5 days4-5/+9
| | | | | | - Go/C/shell setup hooks called the global electric-pair-mode, so one prog buffer turned pairing on in org and text everywhere. Use electric-pair-local-mode. - prog-general set display-line-numbers-type inside the hook, after the mode turned on, so the first prog buffer of a session got absolute numbers. Set the type and width at top level instead. - Updated the go/c tests to stub the local mode.
* fix(eww): quick-add bookmarks to the default store, not a new dirCraig Jennings5 days1-5/+2
| | | | cj/eww-bookmark-quick-add let-bound eww-bookmarks-directory to a path and created a directory there, so B (eww-list-bookmarks) read an unreadable path and quick-added bookmarks vanished after restart. Use the default store both commands share.
* refactor(org-agenda): drop OVERDUE block, lead with the scheduleCraig Jennings5 days1-38/+9
| | | | | | The daily F8 agenda opened with an OVERDUE block that re-listed every past-due scheduled or deadline task. Those items already show in the SCHEDULE block on today's line, since org-scheduled-past-days and org-deadline-past-days are both 10000. The OVERDUE section was pure duplication. I removed it and moved SCHEDULE to the top so the calendar leads the view. The now-dead cj/org-agenda-skip-subtree-if-not-overdue helper, its header defvar, and the nine tests covering it are gone too.
* feat(face-diagnostic): bind cj/describe-face-at-point to C-h FCraig Jennings5 days1-1/+6
| | | | Bind the diagnostic in the help cluster on C-h F (Face), Craig's pick. It shadows helpful-function, which also sits on C-h F in this config. face-diagnostic is required after help-config, so this binding wins. A test asserts C-h F resolves to the command.
* feat(face-diagnostic): Phase 4 render, command, and init wiringCraig Jennings5 days1-2/+158
| | | | Phase 4 completes the face/font diagnostic. cj/describe-face-at-point renders cj/--face-diagnosis-at into a read-only *Face Diagnosis* buffer (cj/face-diagnostic-mode), with a region-scan mode over distinct face-runs (capped at 20) and an out-of-scope banner. It is required in init.el. The render is split into small section formatters tested on captured plists, and the command is smoke-tested and live-verified in the daemon, where it already surfaces the active auto-dim remaps. The command name is settled as cj/describe-face-at-point. The keybinding stays Craig's pick, and face-name buttons plus the module-header allowlist entry are filed as a follow-up. The spec is marked implemented and renamed to its lifecycle filename. 35 ERT tests, byte-compile clean.
* feat(face-diagnostic): Phase 3 per-face provenance traceCraig Jennings5 days1-4/+74
| | | | Add group 5 to the diagnostic core: per-face provenance. cj/--face-diag-provenance reports, for each named face in the stack, which themes set it (theme-face), whether config saved or customized it (saved-face / customized-face), its :inherit chain, and the attributes still unspecified after inherit-following (the ones that fall through to the default -- the direct read on the all-white-elfeed class of bug). The version-sensitive theme-face / saved-face internals sit behind small accessors that treat missing properties as absent rather than erroring. 30 ERT tests, byte-compile clean.
* feat(face-diagnostic): Phase 2 merged attributes and real fontCraig Jennings5 days1-3/+90
| | | | Extend the diagnostic core with the effective merged attributes and the real-font layer. cj/--face-diag-merged-attributes folds the ordered, remap-expanded spec stack (overlays over text-props over default), taking the first non-unspecified value per attribute, labeled "computed". cj/--face-diag-real-font reports font-at's font, or "unavailable" under batch and on terminals. cj/--face-diagnosis-at now returns groups 0-4. Settles spec decision #7 (the hand-fold approach), pinned by fixtures: overlay-over-text-prop, a default remap, a face-symbol attribute. 23 ERT tests, byte-compile clean.
* feat(face-diagnostic): Phase 1 pure read model for the face/font diagnosticCraig Jennings5 days1-0/+138
| | | | modules/face-diagnostic.el carries the Phase 1 core of the face-at-point diagnostic: cj/--face-diagnosis-at returns a plist with the buffer classification (theme-faced / terminal-ansi / document-shr / image-no-text), the character context (char, codepoint, Unicode name, script), and the face stack separated by source (text-property faces, overlays by priority, active face-remapping-alist entries, default). Built from small pure helpers, no display or prompts. 17 ERT tests cover Normal/Boundary/Error per helper. Not yet wired into init.el; the interactive command, rendering, and keybinding land in Phase 4. Spec: docs/specs/face-font-diagnostic-popup-spec.org.
* refactor(org-capture): single-Task desktop popup into the org-roam inboxCraig Jennings5 days1-55/+19
| | | | The Hyprland Super+Shift+N popup now goes straight to a single Task capture into the org-roam inbox (file+headline inbox-file "Inbox"), with no template menu. It drops Bug and Event from the popup, removes the now-pointless org-mks Customize-strip advice, and replaces the Task/Bug/Event subset filter with a one-template builder, cj/--quick-capture-template. The full org-capture menu in the daemon is unchanged. todo.org: cancelled the deferred Note/Recipe popup feature and replaced the old manual-verify checklist with one matching the simpler behavior.
* refactor(themes): retire dupre, fall back to modus-vivendiCraig Jennings5 days3-10/+9
| | | | WIP, the theme-studio export, is the active theme. dupre was only the fallback and a structural reference. Move the fallback to the built-in modus-vivendi, guaranteed present everywhere this config loads. Delete the three dupre files plus its test and palette assets, and fix the stale comments that pointed at dupre-faces.el for the auto-dim and org-keyword faces (those moved to org-faces-config.el). Repoint the dupre-clear-theme spec's palette reference to git history.
* docs: move specs to docs/specs/ with lifecycle-status filenamesCraig Jennings5 days5-5/+5
| | | | | | Separate the 27 formal specs from working notes. Specs move to docs/specs/, notes stay in docs/design/. Each spec carries its lifecycle in the filename (-spec, -spec-doing, -spec-implemented, -spec-superseded) plus an authoritative ID and STATUS property drawer. The status came from checking each spec against the code, not the doc's own field: 6 implemented, 8 in progress, 12 not started, 1 superseded. Inbound links become org-id links so future status renames don't break them; code-comment paths repoint to docs/specs/. Working notes, inventories, reviews, and brainstorms stay in docs/design/.
* feat(auto-dim): dim the org-faces header-row faces in unfocused windowsCraig Jennings6 days1-13/+15
| | | | auto-dim remapped the now-orphaned dupre-org-* keyword and priority faces to their -dim variants; this repoints it at the live org-faces-* layer (org-faces-config) so an unfocused window dims keywords and priorities to a darker shade of their own hue instead of leaving them at full color. Picks up DELEGATED and CANCELLED, which the dupre set never had. The dupre-org-* faces stay defined in dupre-faces.el but are no longer referenced here.
* feat(org): themeable agenda header-row faces via org-faces-configCraig Jennings6 days2-0/+195
| | | | | | Each TODO keyword and priority cookie gets its own named face instead of sharing org's built-in org-todo / org-done / org-priority. org-faces-config.el defines org-faces-<keyword> and org-faces-priority-a..d (plus -dim variants for auto-dim), each with a real default color, and wires them through org-todo-keyword-faces and org-priority-faces once org loads. The file is org-faces-config, not org-faces, because org ships its own org-faces feature that the bare name would shadow. This re-introduces the per-keyword/priority coloring that was stripped earlier, now as a named, theme-agnostic layer a theme can override. The design and the four resolved decisions are in docs/design/org-faces-spec.org; a theme-studio "org-faces" app and the auto-dim repoint follow in later phases.
* refactor(dashboard): drop the theme-test layout and tidy the configCraig Jennings6 days1-107/+30
| | | | | | I removed the theme-test layout: the configured/theme-test startupify variants and the cj/dashboard-show-theme-test-layout and restore commands I'd added to inspect faces live. Live theme testing now uses the loaded theme plus theme-studio's dashboard preview, so the extra layout isn't needed. While here I dropped the obsolete dashboard-set-footer and dashboard-set-navigator toggles (dashboard-startupify-list controls footer and navigator as of dashboard 1.9.0), normalized cj/dashboard-only's tabs to two spaces, and declared el (the variable dashboard binds for the bookmark-insertion override) so it stops byte-compiling as a free variable.
* feat(dashboard): theme-test layout for live face inspectionCraig Jennings6 days1-17/+100
| | | | | | The dashboard's startupify-list, items, and item-generators move from inline :custom values into named defconsts, so the layout-toggle commands can reuse them. cj/dashboard-show-theme-test-layout swaps in a layout that enables init-info, the footer, and a deliberately empty section, so the live dashboard exposes faces the normal layout hides, dashboard-no-items-face among them. cj/dashboard-restore-configured-layout returns to the normal layout. Both refresh the buffer in place.
* refactor(ui): color the cursor and modeline from theme faces, in syncCraig Jennings6 days3-52/+57
| | | | The cursor color and the modeline buffer-name indicator both hard-coded hex colors through cj/buffer-status-colors. Replaced that constant with cj/buffer-status-faces, mapping each buffer state to a theme face: read-only to error, overwrite and modified to warning, unmodified to success. A shared classifier (cj/buffer-status-state) and resolver (cj/buffer-status-color) in user-constants now drive both consumers, so the cursor and the modeline resolve their color from the active theme at use time and stay in sync, including the ghostel-terminal case the cursor already special-cased. Tests cover the map, the classifier, the resolver, and both integration paths.
* refactor(org): let the theme color todo keywords and prioritiesCraig Jennings6 days1-20/+3
| | | | org-todo-keyword-faces and org-priority-faces pinned the keywords and priority cookies to dupre-org-* faces, which only exist when the dupre theme is loaded. Under any other theme those faces are undefined, so org keywords and priorities fell back to the default face instead of the theme's org-todo / org-done / org-priority. Dropped both maps so org follows whatever theme is active. The TODO state sequence and the priority range stay.
* fix(slack): autoload w/@/# commands, guard close-all, register the prefixCraig Jennings6 days1-3/+7
| | | | Three lifecycle gaps that bit before slack loads. The w / @ / # keys bound slack-message-write-another-buffer, slack-message-embed-mention, and slack-message-embed-channel, none autoloaded or in :commands, so they void-function'd before slack started; added them to :commands. cj/slack-close-all-buffers read slack-current-buffer via buffer-local-value on every buffer, which signals void-variable on buffers without the local binding; it now guards with buffer-local-boundp like its sibling. And C-; S was bound with a raw global-set-key, invisible to the keybindings registry; it now registers through cj/register-prefix-map like the signal and erc prefixes.
* fix(erc): one mention notification, real server list, runtime requireCraig Jennings6 days1-4/+5
| | | | Three audit defects. erc-modules carried the built-in notifications module while :config also added cj/erc-notify-on-mention to the same erc-text-matched-hook, so every mention popped two desktop notifications. Dropped notifications from erc-modules and kept the custom one. cj/erc-connected-servers compared a buffer's erc-server-process to itself inside with-current-buffer (always true), so it returned every ERC buffer; it now filters on erc-server-buffer-p and erc-server-process-alive. And user-constants moved from an eval-when-compile-only require to a runtime require, since user-whole-name is read at load time for erc-user-full-name.
* refactor(faces): let the theme own face appearanceCraig Jennings6 days5-90/+6
| | | | Strip per-module face configuration so the active theme is the single source of face styling. Removed the dired text-greying set-face-attribute calls (kept the buffer-local face-remap), the org appearance set-face-attribute calls, the highlight-indent-guides colors, the rainbow-delimiters depth colors (which were overriding the theme's blue with red), and the cj/music-* defface forms plus the emms-playlist custom-set-faces. Left nerd-icons' set-face-foreground alone, since it's runtime tint logic rather than static config. Each touched file is backed up beside it as <name>.el.faces.bak (untracked). The cj/music-* faces now come from the theme instead of the module, so until the theme provides them music rendering falls back to the default face.
* fix(modeline): drop per-render truename, guard vc fetch against signalsCraig Jennings6 days1-18/+25
| | | | The VC modeline cache rebuilt its key every render, and the key included file-truename, so a stat ran on every redisplay rather than once per refresh as the comment claimed. Now it keys on (file show-remote). A moved symlink target is caught at the next TTL refresh, when vc-backend resolves the link fresh. And cj/modeline-vc-fetch is wrapped in condition-case returning nil, so a git signal on a slow or unmounted filesystem degrades to no-VC-info instead of breaking all redisplay.
* fix(help-config): non-destructive info open, clean cancel, drop dead configCraig Jennings7 days1-22/+27
| | | | Three audit defects in one file. cj/open-with-info-mode used cl-return-from inside a plain defun, so declining the save prompt threw "No catch for tag" instead of cancelling. The decision is now a pure cj/--info-open-plan and the command routes through it. A dead :hook (info-mode . info-persist-history-mode, which names a non-existent mode on the wrong hook) and an empty :preface are gone. The auto-mode-alist entry that mapped .info to that interactive, buffer-killing command is dropped, so find-file-noselect of a .info no longer destroys buffers. cj/open-with-info-mode stays an M-x command and C-h i still browses info files.
* fix(dwim-shell): valid zip name, real backup timestamp, reachable menu keyCraig Jennings7 days2-5/+17
| | | | Three audit fixes. Single-file zip named the archive after the input file (<<fne>>.<<e>>), making invalid archives and a "foo." name for directories. It now builds <fne>.zip. The dated backup single-quoted $(date ...), so the stamp sat literal in the filename. Now the timestamp is interpolated in Elisp with format-time-string. The dired menu used "M-S-d", which Meta-Shift-d never emits (it sends M-D), so the menu was unreachable in plain dired. It now binds M-D, matching the dirvish sibling. Both command strings moved to top-level builders so they're unit-testable without loading the dwim-shell-command package.
* fix(markdown): restore the live preview (rename, guard, dead setq)Craig Jennings7 days1-6/+4
| | | | F2 ran markdown-mode's own markdown-preview instead of the custom one, so the impatient-mode strapdown preview was dead. I renamed the custom command to cj/markdown-preview and rebound F2 so markdown-mode no longer shadows it. The server guard now uses httpd-running-p (httpd-process isn't a variable), and a dead (setq imp-set-user-filter 'markdown-html) that named neither a real variable nor a real filter is gone.
* fix(completion): let orderless match, prescient sortCraig Jennings7 days1-0/+5
| | | | vertico-prescient-mode defaulted vertico-prescient-enable-filtering to t, which overrode completion-styles to prescient inside vertico sessions and left the orderless config dead where it mattered. Setting it nil restores orderless space-separated out-of-order matching while prescient keeps doing the sorting.
* fix(keys): destructive delete on capital D, diff on lowercase dCraig Jennings7 days1-2/+2
| | | | C-; b d now runs cj/diff-buffer-with-file (the op I hit most, comparing a buffer against the saved file) and C-; b D runs cj/delete-buffer-and-file. The destructive command sat on the easy lowercase key and diff on the capital. A keymap-lookup test guards the swap.
* fix: vertico-repeat C-s and dotted-repo discovery (audit bugs)Craig Jennings7 days2-1/+9
| | | | | | | | | Two bugs from the 2026-06 config audit. - C-s C-s never repeated a search. cj/consult-line-or-repeat calls vertico-repeat on the second press, but vertico-repeat-save was never on minibuffer-setup-hook, so it always signalled "No Vertico session". I hooked vertico-repeat-save next to the vertico block. It's autoloaded, so the load defers to the first minibuffer. - reconcile-open-repos skipped any repo with a dot in its name. cj/find-git-repos filtered children through "^[^.]+$", which matches only dot-free names, so mcp.el, capture.el, and google-contacts.el were never reconciled while M-P still reported "Complete". It now filters through directory-files-no-dot-files-regexp plus a hidden-dir guard, so dotted repos pass and .git stays out. Each fix is test-first: a failing assertion (hook membership, three dotted repos found) precedes the change.