summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Install offers the built pardes 0.23 binaries first, each with its checksum, ↵Gabriel Schneider5 days
| | | | | | | | | | | | then building from source `project.download` in style.typ holds the version, the SHA256SUMS link and a file per program (the static terminal build and the window build, with what each needs); install.typ shows, for each, the curl, the sha256sum -c of its own checksum and the move into ~/.local/bin, then the clone and the build as the alternative. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The landing shows each thing once, lists what is in pardes, and the guide ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | | | | | | | | | | | opens on an annotated pane The window shelf drops terminals-are-panes (`landing = false` in clips.toml; the casts show it) and keeps the PDF, image, make, diff and 9P clips. Captions say what happens: the live-text cast renames a variable, runs make and lands on the error; the 9P clip is a shell under 9ns where the session is a directory. Casts stack one per row at the page's width, each showing a frame of itself until it plays. Under the lead, a row of linked chips names what is in pardes: terminals, PDFs, images, language servers, diff review, scripting, themes. Install says how long the first build takes (62 s from a clean cache on 16 cores, measured with the dependencies already fetched), and `project` in style.typ has a `download` slot (url, sha256, arch) install.typ shows once a binary exists. #annotated draws a screenshot with numbered pins and their labels as text beside it; the guide opens with one of the orchard theme's shot: the workspace tag, a column's tag, a pane's tag, the box, the body and the scrollbar, before its glossary. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* An image's or a terminal's tag says petscii:on when it is drawn as glyph ↵Gabriel Schneider5 days
| | | | | | | | art, as it always is with no graphics; Petscii on|off still chooses for when there are The images and tagbottomimage goldens are re-recorded for the word. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A window whose focus moves to a PDF or picture while the cursor glides goes ↵Gabriel Schneider5 days
| | | | | | | | | | | | | idle: the glide stops instead of holding frames for ever glideCursor returned early when no cursor was drawn, so a glide under way (right-click a PDF from a tag: the cursor sets off for the new pane) was never stepped again and cursor_glide.moving() kept continuous() true: the GUI redrew at full rate, every frame, until focus came back to a cursor. With no cursor drawn the glide is now dropped; the next cursor lands. A pixel shell's only: the terminal has no cursor glide. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The hero clip middle-clicks make in the README's text, not a word added to ↵Gabriel Schneider5 days
| | | | | | its tag: text in a document runs as a tag word does Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The cast player keeps its colours in dark mode: the site's code-block rules ↵Gabriel Schneider5 days
| | | | | | no longer reach the player's own <pre> Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Every recording on the landing plays by itself, and only while it is on screenGabriel Schneider5 days
| | | | | | | | | | The GUI clips are autoplay muted loop playsinline with their controls, and every terminal cast autoplays and loops. A few lines of script watch them with an IntersectionObserver, playing each while a quarter of it is in view and pausing it when it leaves; without the script the videos still autoplay. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The terminal casts pin the orchard theme and show all four syntax colours, ↵Gabriel Schneider5 days
| | | | | | | | | | | | and the petscii cast opens its pictures on camera from a README pardes colours four syntax classes (keyword, string, number, comment), so main.c gains a comment to show the fourth; types and function names stay plain in any theme. The petscii cast starts on a README naming the pictures and the manual, right-clicks each open, no longer runs Petscii, and searches 'bit 21', which first appears on the CPUID page. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Show, don't tell: the guide, the tutor, scripting, setup and the landing ↵Gabriel Schneider5 days
| | | | | | lead show an example first and cut the prose it makes redundant; the reference's 9P tree and REPLs name Look and Exec Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The guide, the cheatsheet and the tutor point to helix's keymap for every ↵Gabriel Schneider5 days
| | | | | | key, and say where pardes differs Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The hero clip opens its output pane when make is clicked; the image clip ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | shows the Mandelbrot set and a seahorse-valley zoom, no Petscii; the landing lists the PDF and image clips first record_demos.py: text-is-the-interface starts on README and main.c alone, and the middle-click on make brings up the output pane before the right-click puts main.c on line 12. image-pane draws two PNGs, the classic view and a zoom into the seahorse valley; ls lists them, a right-click opens the set, Del closes it and a right-click opens the zoom. clips.toml and CLIPS run pdf-navigation, image-pane, pdf-tint-fit, then the hero, terminals, scripting and diff-review. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The remaining builtins in the reference, setup and the cheatsheet's 9P tree ↵Gabriel Schneider5 days
| | | | | | are written as #word, the window-only ones included, so each links to its glossary entry Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The landing's terminal casts are recorded: live text, terminals as panes, ↵Gabriel Schneider5 days
| | | | | | | | | | | | | and images and PDFs in a plain terminal, each a tty pardes that test/record_casts.py drives in a pty of its own as a plain xterm The script answers only what an xterm answers, so images fall back to PETSCII glyph art and a PDF to its text layer. It stamps each scripted step on a virtual clock and writes the cells the step changed, as pyte saw them, so the same input makes the same file. The petscii cast generates its Mandelbrot views and reads the Intel SDM (--pdf) through a symlink. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Builtins are chips that link to a glossary generated from the registry, and ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | the virtual filesystem is drawn as a tree #word draws a builtin as a chip on the tag's ground over a rule (Look and Exec in acme's look green and exec red), a <code class="word"> in HTML, and links it to its entry in a new Builtins chapter wherever that chapter is in the same output (the book, the site). The chapter is drawn from docs/typ/builtins-tty.json and builtins-gui.json, which `zig build builtins-json -Dplatform=tty|gui` writes from the registry and the Help rows (src/builtins_json.zig, tools/gen_builtins_json.zig; summaryOf in ninep/ctl.zig is now pub); style.typ merges them, marking the window-only words. The doc steps refresh their build's file first, and each build's unit-test fails when its file is stale. Everyday words first (`everyday` in style.typ), then groups by leader prefix as Help has them, then settings, the words with no path, debugging and internals, and last an A-Z index of every word. `zig build book` joins the doc steps. #fstree draws a directory tree with its connectors, a description column and links to the sections that document each file; reference.typ draws the served tree with it, each pane, pty and column file on its own line, and a test in config.zig checks it against ninep/tree.zig's files both ways. The cheatsheet has a two-level version. On the site the tree reads as `tree` text in w3m and wraps descriptions under the names on a phone. The site's tutor draws its B1, B2, B3 and 1-2 chords as mouse glyphs; the landing credits Plan 9 and acme, as does the book's title page; install.typ no longer lists the mirrors (the footer does). Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Clicks in the site's clips stop, then click in place with a ring; the hero ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | | | opens on all three panes; new clips show the PDF and image panes; the Themes gallery has a shot of every theme record_demos.py: a click glides to its target, stops, and presses and releases with the pointer still (a ring and the button's name on the press, a pulse after), pausing before anything moves. The hero's README, main.c and output pane are all up from its first frame. pdf-navigation and pdf-tint-fit page the Intel SDM vol. 2 (--pdf, read where it is): PdfSections to CPUID, a search stepped with n, PdfTint's three looks, PdfFit both ways and a page address. image-pane opens a Mandelbrot PNG the script draws, then Petscii. `record_demos.py themes` shoots one scene in each theme of src/themes, in the Themes order, into docs/site/media/themes with themes.toml (ready). Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A PDF with a JPEG image no longer crashes ReleaseSafe and Debug builds: ↵Gabriel Schneider5 days
| | | | | | | | | | | | libjpeg and jbig2dec are built without UBSan, as the rest of MuPDF is libjpeg calls filter-dct.c's source manager through a function pointer; with -fsanitize=function on the caller and no type hash on the uninstrumented callee, the call trapped (Illegal instruction in jpeg_fill_bit_buffer). The Intel SDM vol. 2 crashed pardes on its first page. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Alt/Option-click is B2 and Super/Cmd-click is B3, sweeps included, as in ↵Gabriel Schneider5 days
| | | | | | | | | | | | | plan9port: a one-button trackpad can execute and look The core decides it at the press (mouse.emulateButton) and keeps it to the release, whatever keys are let go between. Ctrl-click stays a definition, and a program tracking the mouse still gets the plain click (Alt as its meta, Super as nothing). The SDL, macOS and web shells now pass Super; the macOS and web shells pass Alt on a click too. A terminal's mouse report has no Super bit, so in a terminal only Alt-click is B2. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Builtins are written by name: each chapter and the tutor define Look and ↵Gabriel Schneider5 days
| | | | | | Exec once and then say Look and Exec; the cheatsheet draws every mouse button as its icon; one-button clicks (Alt is B2, Super is B3) are in the guide and cheatsheet; building says how to send patches to the list Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Prose that names a mouse button in words drops the glyph beside it, so a ↵Gabriel Schneider5 days
| | | | | | screen reader or text browser reads each click once Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The docs call pardes's scripting interface a virtual filesystem, served over ↵Gabriel Schneider5 days
| | | | | | 9P, and the guide opens on look and execute as the core idea Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The landing leads with live text and plays terminal casts first; the Themes ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | | | | | | | | | | page gets a gallery; install starts at the clone The landing's own words sit in one `landing` record in site.typ: a lead on the core idea (any text is live: right-click looks, middle-click runs), where it runs, and a status line whose version is read from build.zig.zon. Recordings come in two shelves: "In a terminal", asciicast players from docs/site/casts/casts.toml (the first autoplays; all loop and fit the width; data inline, so they play from file://), then "In its own window", the GUI clips. asciinema-player 3.17.0 is vendored in docs/site/player with its Apache-2.0 licence, from `npm pack`. The Themes page shows #theme-gallery() from media/themes/themes.toml, a thumbnail per theme opening the full picture; on paper it is nothing. Both shelves and the gallery show placeholders until their files exist. `project` in style.typ lists the mirrors (sourcehut, git.0x4200.cafe); install.typ clones from the first and names them all, and the footer links each. The README clones before it builds. The scripting interface is a virtual filesystem served over 9P, never just a filesystem, in the tagline, captions and README. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The landing shows the four recordings, pardes is ISC-licensed, and the ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | | | install lines fit a phone clips.toml names the recordings (text-is-the-interface, terminals-are-panes, scripting-over-9p, diff-review), each ready, its caption checked against what the clip shows. The first ready clip plays at the top, full width and on its own (autoplay, muted, looping), and the grid below shows the others, three across. Videos and posters keep the recordings' 960x700 shape. The footer links the source at git.0x4200.cafe/pardes and names the ISC license, from `project` in style.typ. LICENSE holds the ISC text, and the README names it and says the libraries keep their own (MuPDF, on by default, is AGPL). The install lines put their comments on lines of their own, and on a phone a shell line wraps rather than scroll its end away. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Scripts find the session as m=\$PARDES_MOUNT, else 9p; SPC q quits; upper ↵Gabriel Schneider5 days
| | | | | | keeps a selection's newline and its tag words once; the reference takes line files, init errors, dumps, dials and Save on +Tutor Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The documentation site's landing page has four short silent clips in ↵Gabriel Schneider5 days
| | | | | | | | | | | | | docs/site/media, recorded from a hidden test window by test/record_demos.py text-is-the-interface, terminals-are-panes, scripting-over-9p and diff-review: each a webm (VP9), an mp4 (H.264, faststart) and a poster PNG at 960 px, with captions.txt. The script drives a PARDES_TEST session on the virtual clock through stdin mouse events, keys and 9P, keeps the settled frame after each step, draws the pointer and loops each clip with a dissolve back to its first frame. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The docs become a website: a page per chapter, a landing page with ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | recordings, one Typst bundle run docs/site/site.typ writes every page in one bundle compile: the landing page (the tagline, the recordings listed in media/clips.toml, links to start from), a page per entry of `chapters` in style.typ, and the tutor. Every page shares a tag-line header with the chapter words; #doc links resolve to the right page and anchor; a chapter with four or more sections gets an "On this page" list. Links are relative and pages flat, so it reads from file:// or any static server. `zig build site` installs it to <prefix>/share/doc/pardes/site and fails plainly without typst. The look is 0x4200.cafe's lapis layout (framed panels, hard shadows, a tag-line trail, Departure Mono headings over Crimson Pro) in orchard's greens and amber, forestbones_light when the reader prefers light; both fonts are OFL and shipped. In HTML a key is one pill with its + and ›, and a mouse button is an inline SVG in currentColor. The tagline lives once in style.typ for the sheet, the book and the site, and the single page html.typ gives way to the site. The landing leads with one plain sentence (columns of panes, click to run or open, helix keys, scripts through files), then the tagline and a line on where it runs; a hero image is the first ready clip's poster; "Start: the guide" leads the links, "Hands on: the tutor" next, the rest a plain row; the captions use no undefined words. docs/typ/install.typ holds the install lines once: setup.typ has an Install section that includes it, and the landing includes it too. Building says up front it is for contributors and points at setup to install. In HTML the guide's glossary is a closed <details> (#glossary, the same heading and table on paper), each drawn mouse carries hidden text (left-click, hold left, click middle) for screen readers, w3m and copying, and the footer takes a source link and a licence from `project` in style.typ once decided. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The docs give back the day-one lines the trim cut, the tutor becomes ↵Gabriel Schneider5 days
| | | | | | exercises on a +Tutor copy, the README builds the GUI and says what a terminal needs, and Shift-Esc says it needs the kitty keyboard protocol Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The docs trimmed to their path: a guide with Words you'll see and one Where ↵Gabriel Schneider5 days
| | | | | | commands run and panes go table, scripting's first tag word Fmt and eight traps, setup as the one home of the pager and the servers, a scannable cheatsheet, an honest README, the edge cases moved to the reference, and the pager's colours Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A name held for its newline that no newline could make good (a terminal's, a ↵Gabriel Schneider5 days
| | | | | | component over 255 bytes, a path too long) is refused by the write that held it, not at the close (monkey9p's close_runs_only_line_files; follows sswkwwvv) Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The web build compiles again: a dump asks whether a picture is on disk only ↵Gabriel Schneider5 days
| | | | | | where there is a disk (fixes zkmtpmmq) Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A bare Save on a +-named pane that is no output (+Tutor) asks for a path, as ↵Gabriel Schneider5 days
| | | | | | +New's does, never writing a file named +Tutor Co-Authored-By: Claude Opus 5.5 <[email protected]>
* SPC q runs Exit, refused once over unsaved panes as the tag word is, as ↵Gabriel Schneider5 days
| | | | | | helix's :q Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A setting only the other frontend has (Font, GripWidth in a terminal) is ↵Gabriel Schneider5 days
| | | | | | passed over in the init file, not failed: one init file serves the window and the terminal Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Every pane pardes spawns gets $PARDES_MOUNT, the session's own directory ↵Gabriel Schneider5 days
| | | | | | | | | | | | under 9ns --mntgen ($NINE_MOUNT/pardes/<name>), unset when it is not known 9ns says --mntgen with NINE_MNTGEN=1 (cloud9). Under `9ns --unix` the mount is the server it dialled, never this session (its socket does not exist until it starts), so it stays unset there; scripts fall back to `9p -a "unix!$PARDES_9P"`. Nothing is statted: the name comes from the session's socket and the environment. selfmount checks both modes. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* `> exec` through a mount no longer hangs the editor, and a close holding a ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | | | | last line is answered at once: the line runs at the editor's next step A shell's truncating open of exec (or look, ctl, log, pager) sends a setattr that changes nothing; it waited for the editor to be quiet, and through a 9ns mount that quiet could never come: 9ns answers nothing else on the mount while the setattr is out, and the editor's step may be out reading a pane's file through it. This was the "truncating open of exec hangs, only in selfmount" mystery. Only a truncate of a pane's file or a tag waits now. A close holding a last line with no newline hands it to the editor's step (tree.runClosedLines, beside fillClosedPagers), as a /pager close does, and is answered at once; a refusal is the log's err. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* `pardes --wait` whose $PARDES_9P no longer answers says the session is gone, ↵Gabriel Schneider5 days
| | | | | | exit 1, instead of drawing a screen of its own in the asker's terminal Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A dump no longer embeds a picture or a PDF that is on disk: Restore reads it ↵Gabriel Schneider5 days
| | | | | | | | | | | from there, as a clean file's text is Every Dump carried each image's and PDF's bytes in base64 (a 900 KB PDF made a 1.3 MB dump), and a Restore showed that copy however the file had changed. One on disk now goes by its path alone; one not on disk (no file behind it) still has its bytes kept. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A +Pager keeps its program's colours: the session parses what /pager is ↵Gabriel Schneider5 days
| | | | written after its directory with ghostty-vt, SGR becomes spans drawn over the plain text, every other escape is dropped, and PagerColor off pages it plain
* Saving the init file in a pane (Config) applies its settings at once; its ↵Gabriel Schneider5 days
| | | | | | | | | | | | commands are not run a second time A setting changed in the init file took effect only at the next start. The host's Save now hands a written init file's text back to the core, which applies its setting lines again (each idempotent), with the same notice and err for one that fails; a command line in it (`Newcol`) belongs to the start and is left alone. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* The init file reads past a UTF-8 BOM, and every line it fails is said and ↵Gabriel Schneider5 days
| | | | | | | | | | | | logged with its line number, never silent A BOM made the first word unknown, a word no builtin has was skipped without a word, and a setting that refused its value said nothing (the init file ran with announcing off). Each failing line now leaves a notice and an `err - init file line N: <why>` record; `#` still starts a comment only at a line's start, so `#ff0000` stays a value. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* `answer z` quotes `z`, a bare `name` says it names nothing, and `Repl ↵Gabriel Schneider5 days
| | | | | | | | | | | | | klingon` is invalid (EINVAL), its message citing no doc path A refused answer quoted the word `answer`, not the choice refused. A `name` with nothing after it was an unknown control message. A Repl of no such language failed EIO with a pointer into the reference; it now says `invalid language` (EINVAL, here and through a mount) and how many more there are. A builtin's failure that says invalid anywhere is EINVAL, as 9ns reads it. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A rename preview's rows are one format whichever file they are in: ↵Gabriel Schneider5 days
| | | | | | | | | `file:L:C-E`, the range's end too, for the asked-about file as for the others The asked-about file's rows said `file:L:C` while other files' said `file:L:C-E`. lsp-rename-files is re-recorded for the one row it shows. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A look at a path not there under a peer's mount is a miss, the write ↵Gabriel Schneider5 days
| | | | | | | | | answered, as any path to no file is It failed the write with the peer's "file does not exist". A file not found, here or through a mount, is now said and logged as a miss. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Mount dials: a relative socket path is a bad dial address, a socket path not ↵Gabriel Schneider5 days
| | | | | | | | | | there says no such socket (ENOENT), and the hint says tcp! takes a numeric address `Mount peer rel/x.sock` dialled relative to wherever the editor ran; `Mount peer unix!/nonexistent` said `dial failed: no answer` (EIO); and the hint offered `tcp!host!port` where only a numeric address is taken. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* /pager takes one directory an open, as pty/run takes one line: a second ↵Gabriel Schneider5 days
| | | | | | | | | | line, in the same write or the next, is refused EINVAL Through a mount `printf 'a\nb\n' > pager` arrives a line a write, so the second asked for another +Pager on the same open, its answer replacing the first. An open that asked already refuses another. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A bound REPL whose program exited is skipped, as the docs say: out of the ↵Gabriel Schneider5 days
| | | | | | | | | | | | question, the running one taking the text, and a done command REPL's tag says no id Two bound REPLs always asked, an exited one offered and, chosen, sent nothing. Now an exited one is left out: one still running takes the text with no question, none running is said (`no program running`), and nothing runs as a command. A command pane bound as a REPL whose command is done no longer shows its id in the tag. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* An exec or Tty from a pane whose directory is a peer's path refuses before ↵Gabriel Schneider5 days
| | | | | | | | | | | making any pane: `exec: <dir>: no such directory`, no `exit 127` pane left A peer's directory (`/n/peer/x`) is no directory on this host, where the shell starts, so a command there ran in a pane that could only say `(pwd) exit 127`, and a Tty made a new/del pair. missingDir now counts such a path missing, as it does a local directory that is not there. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* An unsaved pane restored from a dump that says nothing of its disk (no ↵Gabriel Schneider5 days
| | | | | | | | | disk_hash, an older dump) asks at its first Save, as for a file changed on disk What its text was read against is unknown, so a Save could write over a newer file without a word; it now asks once first. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A command or terminal pane refused for room is refused before anything is ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | logged or answered of it, and a serial is never handed out twice A command pane with no room logged `new N` and `run N ls`, was answered to its exec, then went with `del N` and no `exit N`; and sync gave N back (`next_serial -= 1`) to the next pane made, so a script holding N found another pane under it. Tty did the same. Both now take a refused placement back at once, before the run is logged or the serial answered (the pane was never announced, so it leaves no record), and fail the write with no room. The decrement is gone: gaps are fine, and a serial once said names one pane for good. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* A command line runs when its newline comes, or at its open's close, never ↵Gabriel Schneider5 days
| | | | | | | | | | | | | | | | because a write was short: a long line cut by a mount is one line A write shorter than its message, and not a multiple of 4096, was taken as whole, its tail a line. 9ns cuts a FUSE write of up to 512 KiB into 65512-byte Twrites, the last one short, so a 600 KiB `Edit ,c/.../` arrived cut and its tail ran as an unknown control message; the 1 MiB refusal failed the same way. Each open now holds a partial line until its newline, or its release, which ends the last line; an Edit block is whole when it closes, as before. A name write with a NUL still fails its write at once. selfmount writes a 600 KiB Edit line and a 1.1 MiB line through a real 9ns mount; building.typ says how lines are cut. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Grep's miss says `Grep: text not found`: it is a literal search, and ↵Gabriel Schneider5 days
| | | | | | | | | "pattern" suggested a regular expression Still ENOENT through a mount ("not found" is in 9ns's table, and the unit tests' errno-words check holds it so). The guide's quote follows. Co-Authored-By: Claude Opus 5.5 <[email protected]>
generated by cgit v1.3 (git 2.53.0) at 2026-10-06 07:21:37 +0000