pardes.git/src/pdf_bridge.h, branch release-0.24

A PDF page taller than one raster is drawn at the size it is shown, in bands around the screen, not squeezed into 4096 rows and blown up

2026-10-01T03:14:45Z

The render geometry (pdf_bridge.c) clamped a page's LONGEST side to max_dimension: 4096 in the SDL policy, 1200 in Kitty's. A page 841.89 x 4818.9 pt, fit to a 1570 px wide pane, came out 715 x 4096 and was drawn 2.2 times too large at 1x, 4.4 times at 2x; Kitty's raster was 210 px wide. Every request now caps a raster ROW at max_dimension, and caps the height too only while that keeps the page at least as large as it is shown (RenderRequest.display_width/height, set from the fit). A tall page fit to width is rasterized at the pane's own width, in SDL, and at Kitty's 96 dpi as any other page is there. Such a page is not one buffer. A raster over 8192 rows (a texture's and Kitty's 10000-pixel limit) or over 64 MiB is banded (pdf_view.zig): its slot holds the rows on screen and a screenful either side, in 512-row chunks, and is rendered again, keeping the chunks it has, once the screen comes within half a screenful of its edge. Each chunk comes from its own render reaching 32 rows past both edges, so a row of the page is the same however it was scrolled to, and no chunk shows its clip edge: band rows are within 3 levels of a whole-page render (108 pixels of 45 million on the reported PDF). Highlights paint over the band's clean rows as over a page's, and the band uses the page's CTM, so search, selection and pointer geometry are unchanged. A display-list render now culls to the band (a scissor in page space). No pixel changes, but a frame that renders new rows of the reported page at 2x fell from 33-43 ms to 13-16 ms (pdf-scroll-bench, --cell 16x32). design.pdf's scroll-bench pixels are identical. Co-Authored-By: Claude Opus 5.5

Repaint PDF highlights by row, send rasters by shared memory, and animate messages

2026-10-01T03:12:14Z

PDF highlights (hover preview, search, selection) are baked into page rasters, and any change re-rendered the whole page with MuPDF; the TTY then re-sent it as base64 (4.7 MB a page), the GUI as a new texture. Worse, a pointer motion over a PDF invalidated the page even when no preview was shown, so every motion paid that. Now: - A raster whose baked highlight set equals the wanted one is left alone. - A highlighted page keeps its clean rows (before highlights and tint); a change repaints only the rows of quads that differ, running MuPDF's highlight pass (pardes_pdf_paint_highlights) over those clean rows and tinting them: the operations a full render performs, so the pixels are identical. MuPDF band renders are NOT bit-identical to a whole page (edge rows, resampled images), so they are never used to patch; the comment claiming otherwise is corrected. - ImagePlace.patch hands shells the changed rows; the GUI uploads just those rows into the texture it holds. - The TTY probes kitty shared memory (t=s) with an id vaxis never reaches and sends rasters as a /dev/shm name when the terminal reads it; direct base64 otherwise (ssh). - Shells that take row patches (GUI, TTY with shm) repaint a selection while it is dragged instead of only on release. Latency elsewhere: - TTY: an animating frame no longer sleeps 16 ms blind; a tick thread posts into the input queue, so input inside the frame is handled at once. - TTY and GUI: queued pointer motions coalesce to the last. - GUI: a skipped swapchain image re-arms the frame (3 retries); animations still tick while nothing presents. - Editing: the line index is carried across an edit instead of rebuilt from a scan of the whole file per keystroke. Messages fall into their row (ease-in; the GUI slides the band out from under the tagline, a terminal fades it), stay until the next input as before, linger MessageLinger ms (default 800), and dissolve (ease-out). MessageAnimation toggles it; both are settings, in Config and startup files. The snapshot harness pins the old behaviour. The detached server now ticks animations. A restored terminal comes back live: the old screen and scrollback (dumped as clean VT by ghostty's formatter, replayed at the new size; older dumps fall back to their rendered text), a dim "restored history" marker, then a new shell in the directory it was in. Right-click on a line number in a file pane looks at that line (a sticky context header's number included). Measured with an external pty driver (TTY), an in-process fence trace (GUI, PARDES_TEST_LATENCY), and test/pdf_pointer_bench.zig (pixel identity against the baseline and a whole-page oracle); balanced A/A/B rounds, paired per-round statistics. Messages stack: each event gets its own row and its own fall, linger and dissolve; a line keeps its row until it leaves and a new one fills the first free row. Announcements and statuses are replaced in place, not stacked. MessageFall, MessageDissolve and DumpDir are settings Config reports. Co-Authored-By: Claude Opus 5.5 (1M context)

Add macOS backdrop blur and preserve PDF ink opacity

2026-10-01T03:12:14Z

Follow embedded PDF links through Look

2026-09-15T20:24:42Z

Refactor panes and filesystem; replace FUSE with 9P

2026-09-07T16:59:12Z

Consolidate pane, layout, memory and host code. Serve 9P by default over Unix sockets, with runtime mounts and optional TCP/QUIC transports. Remove FUSE and obsolete proof-of-concept examples. Fix highlighting and terminal-history performance, expand differential and stress-test infrastructure, sort navigation results while preserving the next occurrence, add syntax-colored Braille minimaps, remove SPC-k, and document 9P interaction as a repository skill.

pdf: continuous scroll bench harness and per-frame render path

2026-08-15T14:58:29Z

replace ArrayLists with bounded storage

2026-08-10T12:17:07Z

support large PDF outlines

2026-08-10T12:17:07Z

expose MuPDF PDF outline metadata

2026-08-10T12:17:07Z

render MuPDF pages directly into owned RGBA

2026-08-10T12:17:07Z