summaryrefslogtreecommitdiff
path: root/src/gui
Commit message (Collapse)AuthorAge
...
* upgrade: an Upgrade's dump and exec are upgrade.dumpInto and upgrade.execAs, ↵Gabriel Schneider12 days
| | | | | | | which the terminal's and the window's upgradeNow each spelled out (the dump's path, the dump, its write, the old binary held, the daemon told, 9P flushed, the exec and what a failed one undoes); the dump is freed with the core's allocator, which made it, where each frontend freed it with its own after a failed exec (a panic there, never reached); writeDump and openSelf are upgrade.zig's own now Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* replica: a client's join, the terminal's and the window's alike, is ↵Gabriel Schneider12 days
| | | | | | | replica.join (the session dialled, its build checked, its dump loaded and checked, or joined again for its view once its link ends, the core it leaves held meanwhile), where tui.zig and gui.zig each spelled out the dial, the check, the load, the redial and their timing; Link.reset takes the join, and Link.load, Link.joinedAs and Link.redial go Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* the four gui unit tests that failed on the base pass: the behaviour tests ↵Gabriel Schneider12 days
| | | | | | | allow the window's own rows (/status ends in `pointer in|out` on a window; a bare setting word moves a setting, a toggle to its other side and Lift, Crt or ShaderAnimation to their next state, and Shader and Pet, whose off means no file and no pet, take a name, not a flip); the tagline pointer-mapping test puts the workspace tag's text TAG_TEXT_INSET cells in, as it is drawn; the caret test that grew a tag past a row's end goes, as a tag wraps onto rows (up to MAX_TAG_ROWS) and no caret is past one Co-Authored-By: Claude Fable 5.1 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* after the merge: the window's frame probe follows a Restore's new core (it ↵Gabriel Schneider12 days
| | | | | | | pointed into the freed old one, so every frame after a Restore wrote freed memory); ov_pinch_scale and ov_touch_scroll_delta go (nothing wrote them once the pinch and touch_scroll events went; the Debug box loses its two dead rows, topbarexec and windowops regenerated alone and re-run); the perf baselines are named tui, as the frontend is (the perf gate had looked for perf-baseline-tui-*.json and found the tty-named files) Co-Authored-By: Claude Fable 5.1 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* pardes: a new panel's frame zero and the tagline band offsets are tested at ↵Gabriel Schneider12 days
| | | | | | | the end of pardes.zig, where both are written, so the tui runs them too (they ran in gui.zig alone) Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* host_io.Inbox's own tests move from gui.zig to the end of host_io.zig, where ↵Gabriel Schneider12 days
| | | | | | | the tui's unit tests run them too (they ran on the window's alone): a completion through a full output queue, the two PTY backpressure cases, the EOF reserve, the completions' order and reset, and the LSP query; PtyTests is host_io's, its waitBlocked pub for the gui test that still waits on a blocked reader Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* draw, gui: a tag caret's span is one Layer.caretSpan (past a full row's end ↵Gabriel Schneider12 days
| | | | | | | it sits at the tag's right edge, where the grid's is clamped), which the core's focused cursor box and the window's caret both take, in place of the rule written in each with a comment pointing at the other Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: one clipInstance cuts a quad to a pixel rectangle and its atlas window ↵Gabriel Schneider12 days
| | | | | | | in proportion, which for a glyph's 1:1 quad is the per-texel cut it made before and for a stretched band's the right one, in place of clipInstanceTop, clipInstanceRight and clipBodyInstance; a body layer's quad wholly past its bottom is empty where clipBodyInstance turned it over Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: one emitScrollRows draws a scrolling body's shifted rows, a glide's and ↵Gabriel Schneider12 days
| | | | | | | a touchpad's fraction alike (the rows the scissor shows, the caret a touchpad's block carries, then the gap: the glide's kept rows or the touchpad's one edge row), in place of emitGlideRows and its twin Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: a closed shell goes to host_io.retireShell as the tui's and macOS's do, ↵Gabriel Schneider12 days
| | | | | | | the window's own retired list gone (its slots, the reap on every spawn and reconcile, the loop's 16 ms poll while one waited); exit now waits out a shell's 100 ms grace (host_io.waitRetired) where it killed it at once, so a shell that saves on its hangup gets to, and one that ignores it is still killed and reaped before the window exits Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* fonts.zig's comments keep their whys and lose their history (max_fonts at ↵Gabriel Schneider12 days
| | | | | | | 512, the Zig 0.16 move, the flake a shared scratch path once caused), and the module doc names its readers: Output.zig's picker, exec.zig's Font, the windows' fallback faces Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: gripDecor draws a grip's mode bars from its first row without a loop ↵Gabriel Schneider12 days
| | | | | | | over the rows it skipped, Post's vignette ceiling is shadowCeiling's (the same sum), ShaderBuild drops expand and a Wake struct whose ctx was never set, and ShaderBuild's and Post's headers say an attached window runs the bundled passes alone (no SPIR-V travels the wire) Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: the test feed reads its ESC]777 events through one prefix and keeps one ↵Gabriel Schneider12 days
| | | | | | | cut short for the next read whatever its kind (a split view; one was handed to vaxis's parser before) Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: PARDES_SDL_PRESENT, PARDES_SOFT_PRESENT and PARDES_TEST_FONT_SCALE go, ↵Gabriel Schneider12 days
| | | | | | | knobs no test and no doc names: the present mode is still the first of immediate, mailbox and vsync the window takes, and a soft present still comes of a transparent window or a refused claim Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: the input's core and the frame's are never null (every Input is made ↵Gabriel Schneider12 days
| | | | | | | with one, both renderFrame callers pass one), so Input.core, renderFrame's, buildOverlay's and the pointer mapping's lose their optionals, and wheelAt its no-core branch Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: a pty slot never holds a closed fd (closePty empties the slot in every ↵Gabriel Schneider12 days
| | | | | | | path), so the fd guards on it, its kill_at, reap's pass over the ptys and shutdownPtys' kill of ptys stopPtys has already emptied go Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: the page is its colour, a [3]u8, not a Ground whose clear was never ↵Gabriel Schneider12 days
| | | | | | | true (WindowOpacity is a uniform now), so the transparent clear colour and the clear_bg tracking in resolveCell and emitInstance go; the concealed-text test keeps its one real case Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: an instance's transition fields are written once: emitInstance no ↵Gabriel Schneider12 days
| | | | | | | longer spells out the defaults CellInstance already has, and setTransitionFields with no track leaves a new instance as it is instead of zeroing it again Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: dead declarations and one-line helpers go: setenv, Msg, MessageBatch, ↵Gabriel Schneider12 days
| | | | | | | the topbarPaneBorderPixels alias, blink_redraws (counted, never read), wakeSdl (sdlWake again), anchorInset, bodyLayerAt, addChecker, guiFontName, handleFinger's forward to handlePairFinger, and a test of std.AutoHashMap Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: a pointer event or the gamepad puts the pointer through one pointAt ↵Gabriel Schneider12 days
| | | | | | | (where it is, and the cell under it), and its moves are posted through pointerMouse rather than spelled out three times Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: mouseCell returns the cell, never null (a window point maps to the grid ↵Gabriel Schneider12 days
| | | | | | | since no post pass moves one), so its nine orelse branches go, and pointer_mapped, which then always equals pointer_present, with them Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: the capture and the soft present copy the frame down through one ↵Gabriel Schneider12 days
| | | | | | | downloadTarget, each still submitting and timing its own wait Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: the cells' and the images' instance buffers grow by one growInstances, ↵Gabriel Schneider12 days
| | | | | | | the images' made on first need as the cells' are rather than at start, in place of ensureVbuf and ensureImageBuffers Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: one texture2d makes every 2D texture, the atlas, an image's, the scene, ↵Gabriel Schneider12 days
| | | | | | | the capture and the post chain's ping-pong and bloom ones, in place of six fills of the same create info Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: one makePipeline builds every pipeline, the cells', the decor's, the ↵Gabriel Schneider12 days
| | | | | | | overlay's, the images' and the post chain's, from its two shaders and its vertex input (none for the post chain's triangle), in place of four copies Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* the pinch and touch_scroll events go: the window and the web page sent them ↵Gabriel Schneider12 days
| | | | | | | and the core only kept them for two numbers on the Debug line; their wire tags (0x1b, 0x1c) now answer BadTag, and the NaN check they had moves onto pdf_scroll's distance Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: the touch debug overlay goes (circles, trails, hex labels in their own ↵Gabriel Schneider12 days
| | | | | | | 3x5 font, the click flash, and the ink runs the labels drew in), the two-finger scroll and tap stay; the overlay draws the pet and the cursor's glide in one call, and renderFrame no longer takes the Debug flag only that overlay read Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: the headless grid host goes (PARDES_TEST_GRID: runGrid, its host table, ↵Gabriel Schneider12 days
| | | | | | | its frame dump and its test), with --gui-grid in test/agent_session.py and its test, its only user, run by hand; the test feed now says only whether its input ended Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: what draws a frame takes it *const instead of @constCast-ing it, the ↵Gabriel Schneider12 days
| | | | | | | image placement helpers name SavedImagePlace instead of anytype, resumeWindows marks the live view's window once after its loop instead of on every pass, and gripButton and applyResize drop parameters they ignored Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: the tests' Guis come from one fixture, testGui(cell_w, cell_h, ↵Gabriel Schneider12 days
| | | | | | | tagline_width, tagline_height), every other field its default, in place of 33 setups written out field by field Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* host_io: the shells a dead session left are one Adoptions table (Adopt's ↵Gabriel Schneider12 days
| | | | | | | ask, adoptShell, a spawn's take, and the Adoption a terminal pane takes), where the tty and the window each kept the same three fields and the same adopt, adoptShell and spawn lookup Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* host_io: a Restore's new core is restoreCore's (the clients that replicate ↵Gabriel Schneider12 days
| | | | | | | the old one let go, the dump read and restored, its failure said once, the old core's pane and theme watches dropped), with both ponytail notes, where the tty and the window each wrote those steps out; each keeps its own ptys, pipes and views, and the tty's dump import goes with its one use Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* file_watch: every pane's watch is dropped by one call (unwatchAll), where ↵Gabriel Schneider12 days
| | | | | | | the tty's Restore, the window's Restore and a tty test each wrote the same eight-line loop over every pane Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* host_io: the image, PDF and syntax subsystems a session draws with start and ↵Gabriel Schneider12 days
| | | | | | | stop in one place (startWorkers, stopWorkers), where the tty's local session, the window's session and its test grid each started them and deferred their stop in eleven lines of their own Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* FrameStats: the timedemos leave the binary: PARDES_FRAMES_DEMO names a text ↵Gabriel Schneider12 days
| | | | | | | script (parseScript: `<ms> [<n>x<every>] <action>` a line, a wheel, touch, key, keys, phase, click, rail_wheel, paste, exec or end), the fifteen compiled ones are tools/demos/*.txt, each read back step for step identical to its builder before the builders went (11,401 steps the binary no longer carries: the Debug binary's .data.rel.ro 8,566,256 to 7,944,088 bytes, an idle detached session's RSS 38.3 to 37.3 MB), and the keystroke one is test/perf.zig's own data, parsed for its tty and written beside its files for its window; a window whose script does not read does not start; a test reads every script in tools/demos Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* vaxis_input: the core's cell styles as vaxis draws them (style, color, the ↵Gabriel Schneider12 days
| | | | | | | underline mapped by its name) and a pasted key's text (pasteBytes) are vaxis_input's, the tui's copies gone; the gui's test-mode stdin takes a paste's keys through the same pasteBytes, which also knows the keypad's Enter (its own copy did not) Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* image: contain and its Fit, a wrapper over containGeometry kept for the ↵Gabriel Schneider12 days
| | | | | | | window built without MuPDF, are gone: gui.zig's two places there (nativePlaceDrawable, the native image instance) ask containGeometry, null exactly when the old size was zero, and the contain test asserts on it Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* surface: a frame's images are a std.ArrayList(ImagePlace), where they were a ↵Gabriel Schneider12 days
| | | | | | | slice of optionals grown by hand (appendImagePlace's doubling realloc, nimages beside it) whose slots were never null: every reader's `orelse continue` and `.?` go (gui, tui, terminal, TerminalImages, ninep/screen, test/pdf.zig, kitty and the two PDF benches); macos.zig's loop the same, passing ast-check but not built here Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* merge the window's new-shell fix (a new terminal's first pty/run waited for ↵Gabriel Schneider12 days
| | | | | | | ever, its greeting typed at the same first prompt after the window's resize; the canary's Newcol refused, the runner's 80x24 terminal shrinking the window), onto the window tier: every process the runner starts is kept off the user's display (WAYLAND_DISPLAY and DISPLAY unset, SDL_VIDEO_DRIVER=offscreen, in Child.spawn), as a second pardes-gui a case started could open a window there; a window's grid is its case's (Detached.Options.grid: its terminal's size and PARDES_TEST_COLS and _ROWS), the goldens regenerated at the 120x40 they were made for (they had been captured at the 80x24 the runner bug gave) and twice reproduced, the monkey at 80x28; gui.zig's comments point at test/process/window.zig; tui 38 (2 skipped), gui 33 and 5 skipped Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* rename: the terminal frontend is the tui (-Dplatform=tui, src/tui/tui.zig, ↵Gabriel Schneider12 days
| | | | | | | --tui with --tty kept as its old spelling, the process tests' backend .tui and --backend=tui, docs/typ/builtins-tui.json, the Settings report's platform), and the builtin Tty is Term (its registry entry, SPC n t, every default tag, the docs, tests and snap scripts), Tty still run as Term through exec.acme_words, Tty+fish too on a tag and a ctl; tty as a pane's raw mode and as the OS's terminal unchanged Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* gui: Upgrade refused while the display has no screen (SDL lists none: the ↵Gabriel Schneider2026-11-01
| | | | | | | screen off and its outputs gone), in words, the session as it was: the new process starts SDL's video afresh and failed there (`SDL_Init: The video driver did not add any displays`), after the dump and the exec, so the session ended, its shell left to the daemon (a window session upgraded with the screen off, 2026-11-01; its dump restored it) Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNAtZenFv4weoMD3cge
* views: a client that joins again has its own view back, layout, focus and ↵Gabriel Schneider2026-10-30
| | | | | | the panes opened in it: an Upgrade's clients and one whose bye says it joins again (a divergence's, wire 19's `bye.again`) leave their view kept for them (`views_away`, digested and dumped, so an Upgrade's new build has it; taped `awayView`), the client names it in its next hello (`Hello.view`, handed on by the daemon with the connection) and its join takes it back (taped `rejoinView`) rather than a copy of the session's; the live view is never a kept one, `views.count` counts the others, and a join with every view taken lets one go (every Upgrade made each client's own panes leave its view: the hunt's 80 ms cell lost `+Search` and `+New`) Co-Authored-By: Claude Opus 5.5 <[email protected]>
* replica: what stopped each input's speculation is counted by name ↵Gabriel Schneider2026-10-30
| | | | | | (PARDES_REPLICA_STATS's `stops`: a host function's, a site's, or a poison's reason, which tape.poison now takes; PARDES_REPLICA_TRACE says it with the input); a window client's PARDES_TEST_LATENCY says each input (`I seq ns at-once`) and, after each unit, the first of its own not confirmed (`U ns seq`), so a harness finds the frame that shows an input that waited Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Adopt and a terminal reader's failure are inputs (Pardes.adoptShell, ↵Gabriel Schneider2026-10-30
| | | | | | readerFailed; tape.Input adopt_shell, reader_failed): tty.zig's and gui.zig's adoptShell ran the Tty word and cleared the pane's greeting straight on the core, and the window's reader failure set its pane's mode, between steps and untaped, so every client of a session that took in a dead session's shells joined again (the hunt's H15: `the unit of 0/48 diverged: the core called poll_frame where the tape has a host (10)`, gone after); replica.zig's test, which the old writes fail Co-Authored-By: Claude Opus 5.5 <[email protected]>
* Messages: a detached session's messages go on the clock in the pump, after ↵Gabriel Schneider2026-10-30
| | | | | | its wait, not in the host's wait: the host's call (tty waitInput, gui waitDetached) was no input, its clock read untaped, so a replay (and every replica of a session nobody attached) kept a message its session had let go; with MessageLinger 0 (test/daemon.py's sessions) a tape whose session said a failure never replayed, on main too (pane 0's msg, digest 6, 7 of 7 runs; 0 of 6 after); test/cross_replay.py's session says MessageLinger 0 again Co-Authored-By: Claude Opus 5.5 <[email protected]>
* remote clients: a pointer's hit carries the place it was made on ↵Gabriel Schneider2026-10-30
| | | | | | (Layer.Place, on the wire, version 17) and a step resolves it from that, never from this process's frame (`surface`, `.local`), which a session never draws for a client: a window's pointer over a tag diverged it at once (the user's `tape byte 119`, 15 rejoins in 25 s) and a click there aborted it; a replay that meets a site its tape lacks, or makes a syscall that is none, is a divergence that goes on harmless (poisonOf), not a panic, and a speculation that makes one waits (Tty's shell lookup, untaped, is a value site now, as Shell's was); a copy over a queue is refused, not asserted; a session's last client, while no one else watches it, leaves it its layout (leaveView, settled as closeView is), which its rejoin copies; a redial waits only after an Upgrade's `refused` (a divergence's redialled a gone session 20 s); a dying session shuts its clients' links before its trace, and an --attach client ends its ssh and socket whatever way it exits; a far join's hashing stops at a blob once its client is let go (its leave waits for it in the step); a confirmed speculation's state is checked against its unit's digest, and the trace gives a key's and a pointer's words; test/daemon.py: 200 far joins level, a window's pointer swept over tags and body, ssh stand-in clients refused and detached, a client of a killed session ending in 2 s; in-process tests for each Co-Authored-By: Claude Opus 5.5 <[email protected]>
* remote client: a write to the session that fails is its hangup (a client cut ↵Gabriel Schneider2026-10-30
| | | | | | off as it lags said nothing and exited 0 when its own write found the link shut before its reader did); a window client's frame is marked rendered as it presents, so the live harness reads its render as its core's time; test/daemon.py's window clients: one far (PARDES_REPLICA_FAR) put off the session's state rejoins from the core it left, and again across an Upgrade (its `refused`); a client's echo stats in PARDES_REPLICA_STATS; --ps beside them lists the session alone, a window's as (gui, attached), its clients none Co-Authored-By: Claude Opus 5.5 <[email protected]>
* gui: the window a replica client (pardes-gui --attach=<socket>:<name> and ↵Gabriel Schneider2026-10-30
| | | | | | <host>:<name>, as the tty's, far through ssh as amort's: its blobs by hash and its stream in zstd) and a window session serving replicas (a terminal's client or a window's): the client feeds its replica what its window does, read over its own view (views.within) and applied after, draws its view between steps, takes the stream on its own reader and rejoins as the tty's does (replica.Link, the client's half shared with it, the far path's Have with it); the session takes clients at its step boundary, one person's input a wait, a client's view no window of its own, its font and opacity answers taped; a client says its frontend first (Event.frontend, in wire 16 beside the far word) and the tty and window builds step alike (one desktop semantics: each has the other's words, which a view takes by its own frontend, builtins.registry.here); a check asked 300 ms after the last input and check, not per wake; test/daemon.py --window-clients Co-Authored-By: Claude Opus 5.5 <[email protected]>
* replica clients over a socket: the session streams each its dump and every ↵Gabriel Schneider2026-10-30
| | | | | | | | | | | | unit after it (replica.zig `Peers`, a writer and a reader thread a client, units at each step boundary, a client's input stepped there in its own view), and `pardes --attach=<socket>:<name>` and `--attach=<host>:<name>` (through `ssh -tt -L`) are a client in the tty host, its keys applied at once by its own replica and drawn by its own process wire.zig version 15: welcome (the client's id and view, the dump's length, the session's UTC offset and digest), dump pieces, units (with the authority's clocks), terminal messages both ways, a check. A unit carries one digest, at its end (`unit_digests`). The client: tty.zig `Shell.remote`, `feed` (a terminal's input to the replica and the session); the session's clock guessed from the last unit's (tape.zig `Guess`); a pump woken only by a client's post reads no new time (`mail_only`), and a unit of another's ending where the last ended leaves the speculation as it is (`neutral`); the rollback copy taken while idle (`Replica.ready`); views a frontend here does not draw left undrawn. A lagging client never holds the session: its queue is cut off past 64 MiB (PARDES_REPLICA_LAG_KB). Detach in a client ends that client; a Restore lets every client go. daemon.zig: a replica the daemon hands over is the session's loop's (a note), not a stub thread's. A window session refuses one in words. test/daemon.py `--clients`: two clients (one through the delay relay) type into a detached session and agree with it, one killed outright, one detached, a third typing into a shell. Co-Authored-By: Claude Opus 5.5 <[email protected]>
* upgrade: a newer build at the path a session started from is said at a ↵Gabriel Schneider2026-10-29
| | | | | | focus-in, once a build, the path read just after the first frame (off a startup's way); an upgraded process lets the binary it replaced go (held, its deleted file kept its blocks) and sweeps its old image's prompt files (kept for an exec that failed); test/daemon.py --upgrade asks through /ctl, checks the fds and prompt files level, and runs on FreeBSD (procstat, the daemon's registry, no kept old binary there) Co-Authored-By: Claude Opus 5.5 <[email protected]>