screen-clone was 2.6x slower than Zig. The main cost was in
PageList::clone_range: it created a throwaway page, recycled it into the
pool and Page::reinit then memset the entire page again, faulting in
every cell. Untouched pages now go to a clean_pool that create_node
reuses without re-zeroing. Page::init allocates cells with
bytemuck::allocation::try_zeroed_vec (Cell derives Zeroable), getting
lazily-zeroed memory from the allocator without unsafe code, and
clone_partial_row_from bulk-copies plain-cell runs in rows with managed
memory. screen-clone clone: 78 ms -> ~28 ms (Zig ~27.8 ms).
osc::Parser::next_slice appends payload bytes in bulk while behaving
exactly like per-byte next() (2048 limit, overflow), verified by a
property test; the osc-parser bench uses it (54.5 -> ~32 ms, Zig ~28).
Still slower: screen-clone render (~2.4x), dominated by per-row style
vectors in RenderState::update.
Test Plan:
- cargo test --workspace (no failures)
- ghostty-bench +screen-clone / +osc-parser with --profile production
- render.rs ports render.zig: RenderState, the per-frame snapshot the
renderer consumes (rows, cells, styles, graphemes, selection,
highlights, cursor, colors), updated incrementally with Zig's full
redraw triggers and per-page/per-row dirty tracking. Cells are stored
struct-of-arrays; use the accessors.
- highlight.rs ports highlight.zig (untracked, tracked, flattened).
- formatter.rs ports formatter.zig: plain, VT and HTML output for
terminal, screen, page list and page, with selection, rectangle,
unwrap, trim, palette, codepoint maps, hyperlinks and pin/point maps.
Screen's test dump helpers now delegate to it.
Cargo.toml also picks up dependencies for the kitty graphics and search
modules that are being ported concurrently.
Test Plan:
- cargo test -p ghostty-terminal --lib (1243 passed; formatter 100/100,
render 15/15 Zig tests)
Ports ZIG/src/terminal/Terminal.zig - printing with grapheme clustering
and Unicode widths, wraparound, insert mode, charsets, cursor movement,
scroll regions and left/right margins, erase/insert/delete, tabs, modes,
alternate screens, save/restore cursor, DECALN, full reset, palette,
mouse/focus/paste flags, kitty keyboard, titles/pwd, semantic prompts,
hyperlinks, protected mode, synchronized output and resize.
stream_readonly.rs implements ghostty_vt::stream::Handler, applying all
state-changing actions to a Terminal and ignoring queries; it is the
base for termio's side-effecting handler. StringMap uses the onig crate
for Oniguruma-compatible regexes (needed by Ghostty's URL regex).
Not yet wired: kitty graphics (hooks and dirty flags are in place where
Zig calls into it, storage is being ported separately) and the tmux
control-mode viewer.
Test Plan:
- cargo test -p ghostty-terminal (849 passed, 5 ignored; all 372
Terminal and 33 stream_readonly tests ported; ignored ones are blocked
on the PageList parity pass)
Ports ZIG/src/terminal/{Screen,ScreenSet,Tabstops,Selection}.zig on top
of the page/PageList layer. The cursor holds a TrackedPin; Zig's cached
page_row/page_cell pointers become accessor methods. Selection methods
take the PageList explicitly because tracked bounds are handles.
Screen tests use minimal plain-text dump/selection-string helpers that
follow the formatter's semantics; they will delegate to the real
formatter once formatter.zig is ported.
Gaps found in the PageList layer are recorded for the parity pass in
docs/rewrite/notes/ghostty-terminal-pagelist-requests.md; three resize
tests are #[ignore]d pending those fixes (PageList::resize keeps
trailing blank rows Zig drops). Kitty image storage and StringMap are
placeholders.
Test Plan:
- cargo test -p ghostty-terminal (443 passed, 3 ignored; Screen 183/184,
Selection 18/18, Tabstops 4/5 - skipped tests use allocator failure
injection)
First cut of the terminal core memory model from ZIG/src/terminal:
size, point, style, hyperlink, ref_counted_set, page and PageList.
Design (see docs/rewrite/ARCHITECTURE.md decisions 2-4):
- Cell/Row are 8-byte u64 wrappers with Zig's exact bit offsets.
- Page owns row/cell Vecs plus style set, grapheme and hyperlink maps;
Page::layout reproduces Ghostty's byte-size formula on 64-bit Linux so
scrollback-limit pruning matches.
- PageList keeps nodes in a generation-checked slab with a free list and
list-owned tracked pins addressed by handles.
This is explicitly partial (docs/rewrite/notes/ghostty-terminal-
pagelist.md): 151/202 PageList tests ported; reflow rebuilds a snapshot
instead of Zig's incremental ReflowCursor; point<->pin conversion walks
from the list head (no active/viewport caches); RefCountedSet doesn't
reproduce probe-dependent ID reuse. A follow-up pass finishes parity.
Test Plan:
- cargo test -p ghostty-terminal (239 tests)
Start of the full Rust rewrite of Ghostty 1.3.1 (reference source in
../ghostty). Sets up a workspace whose crates mirror Ghostty's Zig modules
(vt, unicode, datastruct, os, input, config, terminal, font, renderer,
termio, core, apprt-gtk) so porting work can proceed in parallel per crate.
Also adds:
- docs/rewrite/{ARCHITECTURE,PLAN,STATUS,AGENT_GUIDE}.md: crate map, core
design decisions (safe Rust, slab-based PageList, logical capacity
accounting), phased plan, progress tracker, and porting rules.
- Workspace clippy config relaxing pedantic lints that are pure noise for
low-level terminal code (casts, long functions).
- Shared vt types needed by several work packages up front: kitty
keyboard flags (kitty/key.rs) and the CSI separator bitset (SepList).
Test Plan:
- cargo test -p ghostty-vt (3 tests pass)
- cargo clippy -p ghostty-vt --all-targets -- -D warnings
ddidderr