| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
| |
I replaced the legacy bold/italic/underline/strike booleans with the final model shape across both sides of the tool. weight (light/normal/medium/semibold/bold/heavy) and slant (normal/italic/oblique) replace the bold/italic flags, underline becomes {style: line|wave, color}, strike becomes {color}, and null means unset.
A single migration converts a legacy face on the way in, mirrored as migrateLegacyFace in app-core.js and migrate_legacy in face_specs.py so the JS and Python models can't drift. It runs on import (applyImported, mergePackagesInto) and on every seed that face_spec touches. The captured-snapshot seed (default_faces.seed) narrows the same way it did before. Only bold and italic survive, as weight "bold" and slant "italic", so the generated themes stay byte-identical.
The B/I/U/S toggle buttons keep working through a transitional bridge (legacyStyleOn / toggleLegacyStyle). The weight/slant dropdowns and underline/strike controls that replace them land next. The live previews read the new shape, with a weight name mapped to a numeric CSS font-weight.
The cutover is proven emit-neutral two ways. An ERT test asserts the migrated shapes emit the same attributes as the legacy booleans, and deep-migrating every face in dupre, distinguished, sterling, now, theme, and WIP then running build-theme yields byte-identical output. Full suite green: Python 59, Node 200, ERT 41, plus the browser hash gates.
|
| |
|
|
|
|
|
|
|
|
| |
This is Phase 2 of the face-attribute expansion. The model now carries distant-fg, family, overline, inverse, and extend in final shape across all three tiers, and inherit and height are no longer package-only (a ui or syntax face can set them too).
I kept bold/italic/underline/strike as the legacy booleans for now. The cutover to weight/slant and the underline/strike object forms lands in the next phase with the editor widgets that force it, so the representation and the controls that drive it move together.
face_specs.py holds the canonical defaults. In app-core.js, normalizePkgFace and packagesForExport carry and emit the new attrs: distant-fg resolves through the palette like fg/bg, and each attr exports only when set, so existing presets re-export unchanged. app.js syntaxBlank, uiFaceBlank, and seedFace match the shape.
Nothing changed shape, so dupre, distinguished, sterling, now, theme, and WIP all emit byte-identical themes. make check green: Python 58, Node 193, ERT 40.
|
| |
|
|
|
|
|
|
| |
I extended build-theme's emitter to the full attribute set: family, distant-foreground, a weight and slant range, structured underline (color and wave), overline, strike color, inverse-video, extend, and inherit/height on every tier. It still reads the legacy boolean bold/italic/underline/strike fields, so every committed preset round-trips unchanged.
The emitter is the first piece of widening the studio to all face attributes; the model and UI that produce these fields come next. To keep the change clean I refactored --attrs from nine positional arguments to a single face-spec object and lifted the accessor helpers above their callers.
Added 40 ERT tests covering legacy compatibility, each new attribute, the coercion helpers' edge cases, and an end-to-end round-trip that loads a theme and reads the attributes back off the faces. They run in the theme-studio suite as a new stage.
|
| |
|
|
| |
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.
|
| | |
|
| |
|
|
| |
A newly-loaded package (or a new built-in face) used to fall into emacs-core because grouping is by name-prefix and an unknown prefix matched nothing. Now the fallback routes by where the defface lives: an elpa face becomes its own package bucket, a built-in face a new emacs-general child. So loading a package and running make face-coverage surfaces it as a fresh TODO bucket instead of an orphan in core. Recognized faces still match their family first, and faces.el/frame.el faces stay in emacs-core.
|
| |
|
|
|
|
| |
face-coverage.org was rebuilt by a throwaway /tmp script each time. This makes it reproducible: face-coverage-dump.el dumps every face's name, docstring, and defface file from the live daemon (plus all group docs and package summaries), and face_coverage.py turns that into the tiered worklist (emacs-core / emacs-general / per-package), classifying each face by where its defface lives. make face-coverage regenerates the file; make face-coverage-diff reports the coverage delta against the committed copy.
The dump binds coding-system-for-write so writing the docstring JSON never drops into the interactive coding-system prompt. I validated the builder by regenerating and diffing against the hand-built worklist: headings identical, only the intro and one sharper description differ.
|
| |
|
|
| |
I added a one-line description under each package and subsystem heading, pulled from its Emacs customization-group docstring or elpa package summary, with a parent-group fallback for sub-buckets. 109 of 120 buckets resolved; the rest are single-face buckets the face's own docstring already covers.
|
| |
|
|
| |
I added the missing buckets so every built-in subsystem face and package face leaves emacs-core. abbrev and which-func became emacs-general subsystems; git-gutter, git-commit, twentyfortyeight (2048-game), yasnippet, and edit-indirect became their own package headings. The orphans that were genuinely core (mode-line-active, the tty-menu faces, the line-number tick faces, fixed-pitch-serif) joined the core set. emacs-core is now 74 faces with no orphans left.
|
| |
|
|
| |
I regrouped the worklist into three top-level tiers: emacs-core (the standalone built-in faces), emacs-general (built-in subsystems like org, gnus, erc, diff, and vc, each a nested child), and one heading per third-party elpa package. The tier comes from where each face's defface lives, queried from the running Emacs: /usr/share/emacs is built-in, elpa is a package.
|
| |
|
|
| |
I added each face's Emacs docstring (first line) as the line beneath it, so the worklist reads as what every face is actually for. 1129 of 1293 faces carry one.
|
| |
|
|
|
|
| |
I added a hierarchical worklist of every known face: the live Emacs face-list unioned with everything the studio manages, grouped by package. Each face is marked DONE where the studio themes it, TODO where it doesn't, and each group is DONE, DOING, or TODO by how much of it is covered.
690 of 1293 faces are covered today. The gaps are mostly org, gnus, erc, custom, diff, and the built-in subsystems the package inventory doesn't reach yet.
|
| |
|
|
|
|
|
|
| |
The mode-line hover box (the raised bevel on clickable mode-line segments) came from mode-line-highlight, a face the studio never managed, so it fell through to Emacs's stock released-button default with no way to change it.
I added it to the generated UI face list, between mode-line and mode-line-inactive. The row and box control are already generic over that list, so they appear automatically. build-theme.el's UI emission is generic too, so the elisp side needs nothing.
The face isn't in the captured Emacs snapshot, so apply_hover_box_default seeds its box to the raised default in both build_uimap branches. That matches current behavior and leaves the user free to flatten or recolor it. The mock frame previews the hover by wrapping a mode-line segment in the face.
|
| |
|
|
|
|
|
|
| |
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.
|
| | |
|
| | |
|
| |
|
|
| |
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.
|
| |
|
|
| |
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.
|
| |
|
|
|
|
| |
I added a "Global Prefix Keybinding Alphabet" section to the messenger-unification spec. The per-app C-; prefix is a third keybinding surface, separate from the in-buffer chords (C-c C-c / C-c C-k / C-c C-a) and decision 6's cross-app verbs. Today the action leaf under each app is ad hoc: the same key means different things in Slack, Signal, Telega, and ERC. The section spells out the canonical actions, shows the inconsistency as a matrix, and proposes one leaf alphabet across all four, with the core seven verbs as the unifiable floor and the richer verbs as optional per-backend extensions.
I also added a smoke-first parity note to Phase 1 (build the controllable signel replacement to the capability floor, not its ceiling) and promoted the todo task to [#A] "Unify Signel and All Messengers into one UX" with a direct link to the spec.
|
| |
|
|
| |
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.
|
| |
|
|
| |
The studio opened with every column's span tints expanded, so the first view was crowded with colors the user hadn't asked to see. I set paletteShowFull to false in initApp, so the palette opens showing one tile per column and the existing arrow control expands the spans. The flip lives in the boot path rather than the module default, so the palette gates keep their full-palette baseline. The two that assert span tiles (#counttest, #paltoggletest) now opt into full mode explicitly. A new #paldefaulttest gate asserts the opening collapsed state.
|
| |
|
|
| |
The picker panel's background (#161412) sat a few shades off the page background (#0d0b0a), so it was hard to tell apart from the page. I gave it the studio's gold accent border (#e8bd30) and lifted the background to #1f1c19, both already used in the toolbar chrome, so the panel reads as a distinct surface. The #pickertest gate asserts the accent border and a per-channel background lift of at least 12 over the page, so the distinction can't quietly regress.
|
| |
|
|
| |
mu4e renders the open message with gnus, so the article-view headers, quote levels, signature, and inline emphasis are all gnus faces, not mu4e ones. gnus ships them as bright greens on a dark background, and the theme had no way to reach them. I added gnus as a bespoke view package: the article-view face set in face_data.py with palette seeds that mirror the mu4e header treatment, a realistic preview (header block, emphasized body, an 11-level quoted reply chain, signature), and a #gnustest gate that asserts every emitted face is a real gnus face. Theming and exporting gnus in the studio retires the green in the live view.
|
| |
|
|
| |
The old mu4e preview was cramped and referenced faces that aren't in the mu4e inventory (mu4e-moved-face, mu4e-attach-number-face, mu4e-cited-1..7, mu4e-compose-header-face), so those rendered unthemed. I rebuilt it as a realistic mu4e screen: a status bar, a column header, and one row per message state (unread, replied, flagged, forwarded, draft, trashed, related) with the current line on the highlight background, then a compact message view and the compose separator. It now exercises all 27 mu4e inventory faces and only those. The new #mupreviewtest gate asserts every data-face is a real mu4e face.
|
| |
|
|
| |
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.
|
| |
|
|
| |
markdown-mode fell back to the generic preview (bare face names). I added renderMarkdownPreview, a realistic README that exercises the markdown faces in context: front matter, headers, bold/italic, inline and fenced code, links, lists and checkboxes, a blockquote with a footnote, a table, strikethrough, highlight, math, and inline HTML. A PREVIEW_KEYS map in app_inventory routes markdown-mode to the renderer, and the #mdtest gate checks every data-face it emits is a real markdown face.
|
| |
|
|
| |
The generic .pkgbar button rule outweighed my .viewnav rule on specificity, so the arrows rendered at default button size. Scoping the rule under .pkgbar wins the cascade and restores the compact arrow size.
|
| |
|
|
| |
I added left and right arrow buttons flanking the view dropdown. They step the selection to the previous or next item and re-render the faces table and preview, so you can walk the list without reopening the dropdown. A pure stepViewIndex helper clamps the index to the option range, no wrap. stepView sets the selection and calls onViewChange.
|
| |
|
|
| |
The contrast column showed "5.4 PASS". The number's color already encodes the tier (green AAA, grey AA, red fail), so the PASS/FAIL word was redundant. I dropped it and put the WCAG meaning in the cell's hover via a pure contrastTitle helper. crHtml now renders just the colored number. verdictFor stays for the covered-overlay worst-case readout, which is unchanged.
|
| |
|
|
| |
A non-default height looks identical to the default in the size input, so a stray 1.1 hides in plain sight. I added a small gold corner flag on any per-face setting cell (fg, bg, style, inherit, size, box) whose value differs from the face's seed default. A pure faceBoxNonDefaults helper computes the per-box flags. buildPkgTable resolves fg/bg to hex before comparing, so a palette-name-vs-hex difference doesn't read as a change.
|
| |
|
|
| |
The assignment-view dropdown listed package faces in APPS build order (bespoke apps first, then inventory). generate.py builds them that way, so the list wasn't alphabetical. I added a pure appViewKeysSorted helper that orders the app keys by display label, and buildViewSel uses it. The @code and @ui editor entries above the divider are unchanged.
|
| |
|
|
| |
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.
|
| |
|
|
|
|
| |
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.
|
| |
|
|
|
|
| |
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.
|
| |
|
|
| |
make task-sorted wraps todo-cleanup.el --archive-done, moving DONE/CANCELLED level-2 subtrees from Open Work into Resolved. Ran it: archived this session's closed tasks.
|
| |
|
|
| |
The warning box under the palette wasn't useful there; the same ΔE info is reachable per-chip via the nearest-ΔE tooltip and inline contrast. Remove renderPaletteWarnings, the #palwarn element, its CSS, and the #deltatest gate. paletteWarnings still runs for the per-chip nearest distance.
|
| |
|
|
| |
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.
|
| |
|
|
| |
<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.
|
| |
|
|
| |
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.
|
| |
|
|
| |
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.
|
| |
|
|
| |
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.
|
| |
|
|
| |
Same fragility as the sibling tests: it stubbed env-laptop-p (now unused by the direction logic) and passed only because the batch frame is landscape. Stub the direction directly.
|
| |
|
|
| |
These tests stubbed env-laptop-p to pin the dock edge, but the aspect-ratio rework dropped env-laptop-p from the direction logic, so the laptop case resolved to the batch frame's real (landscape) aspect and split a third window. Stub cj/--ai-term-default-direction directly.
|
| |
|
|
| |
The literal {bg:MAP['bg'],fg:MAP['p']} repeated 32 times across app.js, palette-actions.js, and the browser gates. Replace it with a groundPair() helper. Named groundPair, not ground, to avoid colliding with the local ground bindings destructured from columnsFromPalette. No behavior change; node tests and browser gates are the safety net.
|
| |
|
|
| |
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.
|
| |
|
|
| |
The aspect-ratio rework dropped env-laptop-p from the direction logic, so these tests stubbed a dead function and depended on the batch frame's real dimensions. Stub cj/--ai-term-default-direction directly.
|
| |
|
|
| |
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.
|
| |
|
|
| |
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.
|
| |
|
|
|
| |
- 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.
|
| |
|
|
| |
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.
|
