pardes.git/src/terminal.zig, branch main

Insert on a terminal's text says it edits a copy, run with Tab or B2, that the program's next drawing replaces, and that Ctrl-b types to the program

2026-10-02T04:33:55Z

Co-Authored-By: Claude Opus 5.5

An image's or a terminal's tag says petscii:on when it is drawn as glyph art, as it always is with no graphics; Petscii on|off still chooses for when there are

2026-10-01T16:44:34Z

The images and tagbottomimage goldens are re-recorded for the word. Co-Authored-By: Claude Opus 5.5

A terminal's petscii:on survives Dump and Restore: the dump says it, and the restored terminal draws its graphics as glyph art again

2026-10-01T10:16:43Z

An image pane's Petscii choice was dumped; a terminal's was not, so a Restore brought it back drawing pixels. dump.Terminal carries petscii, and a restored terminal is turned to glyph art when it was. Co-Authored-By: Claude Opus 5.5

A shell's prompt drawn on the alternate screen is a prompt: a terminal whose program was killed there is no longer busy for ever

2026-10-01T07:57:43Z

htop killed with `sig KILL` leaves the emulator on its alternate screen; bash comes back and draws its prompt there, marks and all, but promptInputReady refused any prompt on the alternate screen, so pty/status said busy 1 and pty/run answered a bare `busy` for good. A prompt mark there now counts. A run sent at such a prompt leaves the alternate screen in the emulator first, as a reused command pane does, so its output lands on the primary screen and is read whole, not `exit 0 cut` with nothing. Busy there with no prompt drawn, and no program name to give, a run is told `busy alternate screen`. Co-Authored-By: Claude Opus 5.5

pty/run after a clear answers what the command printed from the clear on, whole; cut only when its start really scrolled away

2026-10-01T05:47:36Z

A clear (ED 2, ED 3, a reset) under a running command marked its output lost, so `clear; ls` answered only "exit 0 cut". The output's start is now pinned again at the screen's top after the erase: what the command prints from there is its answer, whole, and a cut is said only when that start leaves the scrollback. Co-Authored-By: Claude Opus 5.5

A terminal's kitty graphics draw as glyph art where pixels cannot, and by choice: Petscii on the pane, TermImages for new ones

2026-10-01T03:53:56Z

Under an outer terminal without kitty graphics a tty pardes cannot pass a terminal pane's images on, so it draws them itself as an image pane's Petscii does: each row of an image is matched to C64-palette glyph art over the cells it covers (GlyphArt.renderRect, the exact-grid half of GlyphArt.render), cached while it is on screen. yazi is still told the kitty protocol works, so it keeps sending its previews. The same art is a choice anywhere: the Petscii builtin now takes a terminal pane too (PaneKind.picture), flipping its images between pixels and glyphs, and a terminal holding images, or drawing them as glyphs by choice, says petscii:on|off in its tag as an image pane does. TermImages real|petscii (default real) is what a new terminal starts with; it is in DumpConfig, /ctl and the reference. Co-Authored-By: Claude Opus 5.5

A terminal pane draws the kitty graphics its program places, so yazi's image and PDF previews show

2026-10-01T03:53:17Z

yazi asks the pane XTVERSION, and ghostty-vt answers "libghostty": yazi takes it for Ghostty and sends its previews by the kitty graphics protocol with Unicode placeholders (a=T,U=1, then U+10EEEE cells). ghostty-vt stored the images, but nothing drew them: the placeholders showed as tofu in the GUI and as raw codepoints in the tty, in every build since terminal panes answered queries (vkznsvzs, 2026-07-01). Before that no query was answered and yazi fell back to chafa's text art. Each frame, TerminalImages.zig reads the active screen's images and placements (virtual ones through their placeholder runs, and ones placed on a cell), cuts each into the cell rows it covers, maps those rows through the body's own row walk, clips them to the pane, and hands them on as ImagePlaces with exact geometry (NativePlacement.exact). The GUI draws them as textures; the tty passes them to an outer terminal with kitty graphics as one placement per row, its source cropped to whole cells. Placeholder cells draw no text. Scrolling, a delete, a resize, the alternate screen and a closed pane all fall out of reading the emulator's state every frame. A pane now knows its size in pixels: CSI 14/16/18 t are answered, the emulator's width_px/height_px follow the shell's cells, and the pty's winsize carries pixels (host_io.ptyWinsize), so yazi and a nested pardes size their images right. ghostty-vt's image store is 64 MiB per screen (its library default, 10 MB, refused a large preview), PNG transmissions are decoded with stb_image, and RGB images are converted only while shown. test/yazi_preview.py runs yazi in a hidden window on a PNG and a PDF and checks the captures, also through a tty pardes nested in the pane. Co-Authored-By: Claude Opus 5.5

Release Pardes 0.9

2026-10-01T03:12:17Z

Co-Authored-By: Claude Opus 5.5

A paste into a terminal is xterm's paste: bracketed when the program set 2004, and the control bytes a paste could smuggle a command in by, ESC first, become spaces, so no ESC[201~ in the text ends the brackets early

2026-10-01T03:12:17Z

Every paste path goes through edit.typeToTty (the GUI's and the tty's paste, p from a register, the acme chord's paste, a REPL's send), which now takes ghostty's paste encoder. A 9P write to a terminal's body or pty/data stays typed input, not a paste. The GUI test shell's stdin now takes a bracketed paste as one paste, as the tty does, not as keys. Co-Authored-By: Claude Opus 5.5

A program that asked for focus reports (1004) hears CSI I when its pane takes the keyboard and CSI O when it gives it up, to another pane, a tag or another window

2026-10-01T03:12:17Z

The core has a window-focus event now, from SDL's window focus and from the tty's own focus reports (vaxis turns 1004 on outside); the detached wire carries it: version 10. The test shell's ESC]777;mouse;focus;<1|0> fakes the window's. Co-Authored-By: Claude Opus 5.5