Moving an image whose crop is its own (`q re W n cm Do Q`) as part of a
group put a fresh `cm` before the clip path on every move, so twenty
moves left twenty `cm` operators in front of one `re`. It never nested
deeper, but the content stream grew with every drag.
When a `cm` already sits right before that clip (one an earlier move
added, or the file's own), `translate_images` now replaces it with the
product of the new and old matrices. The CTM at the clip is
`c · prev · outer` either way (ISO 32000-1 8.3.4, 8.4.4). A previous
`cm` that is not finite, or a product larger than 1e12, keeps the old
behaviour and inserts a separate `cm`.
The new test moves an image cropped smaller than itself twenty times and
checks that the image and its visible crop both moved by the full offset,
that the operator count stays at its value after the first move, and that
the caption after it is untouched. It fails without this change
(32 operators against 13).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The engine caches a page's text blocks for opening a paragraph and
applying its edit, but only while the document is unmodified: it was
keyed by the source identity, which edits keep. After the first edit the
cache never hit again, and the page was grouped again on every call (the
inline editor and the panel each call it after an apply).
pdfcraft_cos::Document::read_state() returns what the document reads as:
its source identity, its trailer and every changed object, compared as
allocations that the state keeps alive. A clone or a snapshot restored by
undo has the same state; any set, add, free or trailer change makes a
different one. The cache is keyed by it, so it hits after edits and never
returns another state's blocks.
On a page of 120 rows, text_blocks after an edit took 0.215 ms per call
and now 0.006 ms (release build, best of three; a throwaway timing test).
Tests: a read state matches a clone and not a document changed by set,
free or a trailer change, nor another document; after an edit the engine
groups the page once and undo and redo get their own blocks. The engine
test fails without this change.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
* M8: redaction removes the replies of the annotations it removes (#2659)
A reply (/IRT) placed away from the mark survived, and its /IRT kept the
removed annotation - its comment text and the QuadPoints of the redacted
words - in the saved file. Replies to a removed annotation, and replies to
those, now go with it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RUNQSPVXx3dDBabRZHAnjz
* M8: test replies listed before their parent, with a pop-up on a reply (#2659)
Review of #2697: the replies precede the annotation they answer in
/Annots, deepest first, and a reply has a pop-up. A single pass in
/Annots order would keep them; the test fails with the loop cut to one
pass, which the first test does not notice.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RUNQSPVXx3dDBabRZHAnjz
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Field detection read a path as a rectangle only when it had exactly four
corners, and gave up on a subpath once it reached five points. The French
income tax form 2042 draws its letter cells with a midpoint on each
vertical side (six points), so most of its cells were not found.
A subpath now keeps up to 16 points, and a point lying on a straight side
between its neighbours is dropped before the four-corner test. A point
that turns back along a side is not between its neighbours and stays, so
a path that doubles back is still not a box.
Claude-Session: https://claude.ai/code/session_01RUNQSPVXx3dDBabRZHAnjz
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
On wasm, detect_system_lang() always returned English, so Auto showed
English whatever the browser's language. It now reads
navigator.languages in order, then navigator.language, and takes the
first tag with a catalog, as the desktop builds do with the system
languages and as VectorCraft and FilmCraft do in their browser builds.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The Hand tool scrolled by the page widget's drag delta from inside the
page view's ScrollArea. egui applies that delta when the area ends, after
the frame's pages are laid out and drawn, so the page showed each move a
frame late and trailed behind the pointer. (Until the page view stopped
animating its scrolls, the delta was also eased in over several frames,
which is the stutter #924 and #1992 describe.) On top of that, the points
the pointer moved before egui tells a drag from a click were dropped, so
the page never caught up with the grab point.
Pan before the area lays the pages out instead: when the page widget is
being dragged with the Hand tool, set the scroll offset from the last
offset minus this frame's pointer delta, or minus everything moved since
the press on the frame the drag is recognised. The page now moves point
for point with the pointer in the frame it moves, and still stops where
it is let go (no fling). Other scrolls (go to page, links, zoom anchors,
compare sync) are untouched and still win when asked for in the same
frame.
* Control channel: don't lose the refusal reply to a TCP reset
When the control channel refused an unauthenticated connection (deadline
passed, overlong line, bad first request) it wrote the error line and
closed the socket at once. If the client had sent input the server had
not read, Linux answers that close with a reset instead of a FIN, and the
reset can discard the reply before the client reads it. That is the
flake in an_unauthenticated_connection_is_closed_at_the_deadline: the
dripping client's bytes were still unread when the server closed.
Now the server shuts down its write side after the reply and reads and
discards input until the client closes or a 1 s linger passes, and only
then closes. The new test a_client_refused_with_input_unread_gets_the_reason
pins it deterministically.
* Control channel: refused connections linger outside the auth places
A connection refused before auth used to keep its place among
max_unauthenticated while it lingered (up to 1 s), so refusals could
keep real clients out. Now `refuse` gives that place back and takes one
of a separate, smaller set of lingering places (Limits::max_lingering,
8) before the reply is written. When none is free the connection is
closed right after the FIN instead of lingering, so threads stay
bounded as #1806 intended.
Tests: a_lingering_connection_holds_no_place_to_authenticate and
only_so_many_refused_connections_linger.
* Control channel tests: a 5 s linger for the two linger tests
Their timing checks (the auth round trips finishing within the linger,
and a non-lingering connection failing a write within half of it) had
only 1 s and 500 ms of margin with TEST_LIMITS. Each now serves with its
own Limits and a 5 s linger; the assertions are unchanged and neither
test waits for the linger on its passing path.
text_layer wrote '/OCR BMC BT 3 Tr /PCHelv 1 Tf ... ET EMC'. Tr and the
font are graphics state, and ET doesn't reset them, so every content
stream after the layer on the page inherited 3 Tr: a header or footer
added after OCR was in the file but drew invisibly, in PdfCraft and in
MuPDF. The layer is now wrapped in q ... Q inside its marked content.
Claude-Session: https://claude.ai/code/session_01RUNQSPVXx3dDBabRZHAnjz
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The print-ready PDF copied the document's /OCProperties as they are, so it
printed the layers in their default state: a layer hidden in the Layers
panel disappeared on screen and came back on paper. Print as image already
used the view's layer choices (it renders with the document's config).
Once the layers have been changed, the copy imposed for printing now has
its default configuration (/D) set to them: /BaseState /ON, /ON the
visible layers, /OFF the hidden ones, its other entries kept. With the
layers untouched, the document's own configuration prints as before.
Claude-Session: https://claude.ai/code/session_01RUNQSPVXx3dDBabRZHAnjz
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Section titles are drawn upper case through to_uppercase(), and tool names
inside sentences lowercased through to_lowercase(). Rust's case mapping knows
no language, so in Turkish "Gelişmiş" became "GELIŞMIŞ" and "İmza" became
"i̇mza" (i plus a combining dot). Turkish maps i ↔ İ and ı ↔ I.
upper_label() translates and upper-cases a title with the Turkish rule when
the label has a Turkish row (an untranslated label is still English and keeps
the English rule); in_sentence() lowercases with it. Other languages are
unchanged. The test failed before the fix ("GELIŞMIŞ").
Seven tests failed at random on slow or busy machines, main included
(FreeBSD run 38834256439 at 5fdd820b). They now count the work or wait
for the state they check; no assertion is loosened and nothing retries.
- spot_colour_images_run_their_tint_transform_once_per_sample_value
counts tint-transform runs (exactly 256, not one per pixel) through
vendored hayro-interpret patch (34), color::tint_transform_runs, a
thread-local counter, instead of a 10 s deadline.
- shadings_with_large_sampled_functions_paint_quickly: no counter tells
the flat table from the hash map (both read every sample per build),
so the build each paint does is timed against reading its samples on
the same thread, fastest of 8 rounds each, and must stay under 10x
(flat table 1.1x, hash map 109x), instead of a 10 s deadline.
- page_contents_and_nested_forms_share_one_budget: the budget is
checked by its warnings; the deadline is only a hang guard, now 600 s.
- The pen-stroke tests hold finished renders back from the view
(DocView::hold_rasters, for tests) while they look at the settling
stroke, then wait by state for its raster.
- files_and_quit_from_the_operating_system waits for the tab before
deleting the file the open worker reads.
- image_pixels_follow_the_frame_while_a_corner_is_dragged waits for the
page raster on screen before each drag.
Product fix (#1858): a released drawing stayed painted only until any
raster of its page arrived. In full screen its edit is applied a frame
later, so a raster already under way could end the preview first. A
settling drawing is now marked committed when its AddAnnotation lands,
only rasters after that end its preview, and a failed edit drops it.
reading_lines grouped the joined operators by the stream their first token
is in and walked each group on its own. A marked-content sequence opened in
one stream (its /ActualText dictionary, or the whole BDC) and closed in the
next lost its replacement text: the export read "Glyphs" where every reader
of the joined streams reads "Replacement text".
The page's operators are now walked once. Each shown text still counts as
the stream its operator starts in, and text from another stream still starts
a new line, as before.
place_tagged wrapped the page's content in q/Q only when no Wrap stream
existed yet. Content appended after an earlier wrapper (by another tool, or
an older OCR layer) that left a cm, a clip or text state set then moved,
hid or distorted every later header, Bates number, watermark, added text,
flattened annotation and redaction cover.
Placing a mark on top now reads the page's actual operators instead of the
/PCMark tags. When the content parses cleanly with balanced q/Q and closed
text objects and paths, the stretch from the first stream that leaves state
set is wrapped, once, at a stream boundary no comment runs into. Content
that is already self-contained gets no new wrapper, so repeated marks don't
pile up.
Content that can't be read that precisely (stray bytes, unbalanced q/Q, an
unfinished text object or path, too large) is wrapped whole as before,
rather than refusing: redaction stamps its covers after removing the text
and can't back out. A q that content leaves open is closed after it, where
closing it changes nothing already drawn.
The wrapper's close now starts with a newline, so a comment at the end of
the content before it can't swallow the Q. Removing marks strips only
wrapper pairs that end the list, keeping one that later content still
relies on, and recognises the older close bytes too.
Pieces::join separated a page's content streams with a newline, which ended
a comment left open at the end of one stream. hayro, MuPDF and Poppler join
them with a space, so the comment runs on into the next stream and hides its
first line. Edit text, object_list, the redaction audit, redaction and the
Office/HTML export therefore saw text and boxes that nothing draws, and
redacting unrelated text on the page made the hidden content visible: the
rewritten stream dropped the comment.
- Pieces::join uses a space, so every reader of the joined streams agrees
with rendering.
- The parser records, per piece, whether a top-level comment runs into it
and out of it (boundaries only, not every comment). A redaction rewrite
of a piece ends an incoming comment before its first operator and leaves
an outgoing one open, so hidden content stays hidden.
- Reading for export parses the streams joined, instead of one by one.
- The redaction tag check compares the joined streams the same way.
- Pieces::piece_of uses a binary search.
RUSTSEC-2026-0370/0371/0372 were published today for zvariant, zbus_names and
zvariant_derive. All three are unmaintained notices rather than vulnerabilities,
and none has a fixed release to move to. They reach the tree as the D-Bus stack
behind accesskit_unix -> atspi (Linux accessibility) and as a dev-dependency of
pdfcraft-platform, so nothing ships in the other targets' binaries.
This is why main went red today; no dependency version changed.
craft-libs#169 is merged, so the fork pin is no longer needed. Pinned to
storytold/craft-libs 82a24b006eefa036b43c09a67fb9a3fbd17ffe63 (main after the merge) rather than the older 12281d07,
which predates resolve_absolute and the Windows colon refusal.
GoToR file specifications go through craft-docpath: network, device and
object-manager paths (\??\, /??/) are refused without touching the
filesystem, relative targets resolve against the source document's
folder, and a result that is not absolute is refused rather than read
against the working directory. Files named on the command line open
under their absolute path, so their relative links keep resolving.
An attachment opened as a tab is named with the same plain file name as
its Save suggestion, and Export certificate names the file the same way
from the certificate subject. Writing files into a chosen folder refuses
any name with a separator, a drive, or that is . or .., so no caller can
write outside the folder.
* fix(ui): validated signature display, snapshot tab reuse
Signature widgets with layered appearances baked the unverified state
into the page, so a valid signature kept reading 'Signature Not Verified'
while the banner and panel said valid. Validated verdicts are now drawn
into the icon and caption layers in the display copy only: the signed
bytes are untouched, saving keeps the original appearance, and files
without those layers or without a valid verdict render exactly as before.
Other valid signatures still get the verdict mark over their widget,
sharing one style with the Signatures panel, and hide it again once the
appearance itself shows the verdict.
View signed version and View revision opened a new tab on every click,
and every tab showed the document title, so repeats piled up identical
tabs. Repeats now reuse the snapshot tab, snapshot tabs read like their
source plus the suffix in the tab strip, window title and Combine picker,
and carry a notice bar naming the revision with a Close that asks about
saving first.
Signed-off-by: Chiranjeevi U <244287281+chiranjeevi-max@users.noreply.github.com>
* Address review comments on validated appearances and trust refresh
Count every field when checking shared appearance streams, so a
caption shared with an unvalidated field is never rewritten for the
valid one alone.
Pick the icon slot by drawing instead of length, and only when a
single layer qualifies. Bare rectangles never count as the icon, and
ambiguous or missing slots fall back to the overlay badge.
Revalidate and invalidate together through one helper, so trusting a
certificate always drops rendered pages. Trust alone still marks
nothing dirty.
Tests: mixed-verdict sharing, longer decorative borders, competing
and doubled layers, capture-before preservation, trust invalidation
without zooming, and signed Session round trips with rendering,
saving and passwords.
Signed-off-by: Chiranjeevi U <244287281+chiranjeevi-max@users.noreply.github.com>
* Prove the icon slot geometrically and cover real layered signatures
A lone line drawing is still not the icon: frames and borders span the
box while the status mark lives in a corner of it. The icon now has to
sit strictly inside the appearance frame and cover at most half of it;
anything else falls back to the overlay badge. Widgets without a
measurable frame veto the rewrite the same way.
Tests: parsed extents and inhabitant checks as units, both reported
border repros plus a frameless widget, and real layered signatures
built with cms::sign_detached through Session open, trust, rendering,
save and reopen, plain and password-protected.
Signed-off-by: Chiranjeevi U <244287281+chiranjeevi-max@users.noreply.github.com>
* Identify the icon by shape instead of geometry
A lone line drawing is still not the icon, but size and position were
the wrong signals: the real status mark is large, and geometry blocked
it. Frames, borders and panels are rectangles by nature while the mark
is not, so a drawing made only of axis-aligned rectangles is decoration
however it is stroked. Singularity, shared-stream and template checks
are unchanged; anything ambiguous still falls back to the overlay badge.
Tests: rectangularity as a unit, nested line-drawn frames, and a
triangular fixture mark in place of the rectangular one.
Signed-off-by: Chiranjeevi U <244287281+chiranjeevi-max@users.noreply.github.com>
* Budget the appearance walk and keep the marks with the bytes
The walk that finds a signature's icon and caption capped depth,
layers and fields, but not the work: a dictionary whose entries all
point at the next one is visited once per entry per level, so a small
hostile file never finishes opening. The walk now counts its visits
and gives up, and a field whose tree cannot be read to the end stops
the whole pass, because a layer we never read may be the one two
fields share and a rewritten shared layer puts one field's verdict on
another field's page. The sharing count is exact again with that: two
widgets of one field pointing at the same appearance claim it once,
or the field vetoes its own rewrite.
A mark used to outlive the bytes it described. A failed display write
left the page showing the file's own caption while the verdict still
claimed its state was drawn; a metadata edit dropped the mark while
the page kept drawing the state, so the badge landed on top of it. The
write fallback now drops the marks and logs the loss, and metadata
carries them over.
Tests: a self-referential appearance tree, which never finished before
the budget, two widgets sharing one appearance, a metadata edit that
keeps its mark, a save that goes through the writer instead of
short-circuiting on the opened file, and the snapshot bar's Close
button pressed for real.
Signed-off-by: Chiranjeevi U <244287281+chiranjeevi-max@users.noreply.github.com>
* Cite the new regressions in parity, restore main's installer note
The appearance entry now cites the bounded walk and the save through
the writer, and says what a tree that cannot be read to the end does.
The installer entry carried main's older text, which still claims
signed installers are planned; main's newer note is taken verbatim.
Signed-off-by: Chiranjeevi U <244287281+chiranjeevi-max@users.noreply.github.com>
* M9: preserve rounded decoration beside raster status glyphs
* M9: place appearance progress beside signature history
* M9: retain appearances with ambiguous curved layers
* M9: keep signature verdicts in viewer overlays
Defer generic appearance rewriting: caption text and a unique drawing do not establish an icon role. Preserve original artwork for saves, image and text exports, and both print modes while retaining snapshot reuse and trust refresh. Cover chamfered borders and mixed signer/caption streams through signed Session round trips.
---------
Signed-off-by: Chiranjeevi U <244287281+chiranjeevi-max@users.noreply.github.com>
Co-authored-by: skylar taylor-barrick <hi@skylar.cc>
* fix(render): sampled functions index a flat sample array (vendored hayro-interpret patch 31)
A Type 0 sampled function kept its samples in a HashMap keyed by each point of
its /Size grid (two SmallVecs per key, SipHash), and the renderer builds a
shading's functions again every time it is painted, so every paint hashed one
insert per sample. A 2,097,152-sample ramp painted 30 times took 38 s; the
pdf.js corpus's bug1721218 (InDesign) spent most of a 1.5 s page on it.
The samples now stay in stream order in one Vec, and a grid point is found by
strides (first input fastest). The interpolation keeps the same recursion and
arithmetic, so results are bit for bit the same, and a point the stream ended
before still gives no value. The repeated-paint page draws in well under a
second. The maximum sample value no longer computes 2^32 in a u32 (a panic with
overflow checks on at 32 bits per sample).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MoEx9Nj8KC4WKDcf5YaokQ
* fix(render): one decode of an image serves every size it is drawn at (vendored hayro-interpret patch 32)
The decoded-image cache (patch 24) keyed each decode by the size the image is
drawn at, but only the JPX filter decodes a reduced resolution for a size:
every other filter decodes the whole image. So a JPEG or Flate scan missed the
cache at every zoom step, was decoded again, and was kept once per zoom level.
The key now carries the target size only when JPX decodes the image, its
/SMask or its /Mask (read from /Filter or /F, erring towards keeping the
target). `InterpreterCache::decoded_image_decodes` counts the cache's decodes;
the new test draws a Flate image at five sizes and sees one decode, and a
copy whose soft mask is JPX-encoded still decodes per size. Both assertions
fail with the old keying.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MoEx9Nj8KC4WKDcf5YaokQ
* fix(render): patch meshes copy their vertices' colours less (vendored hayro-interpret patch 33)
A Coons or tensor-product patch is drawn as 722 triangles over a 20 x 20 grid
of points, and every vertex's colours (a SmallVec) were produced or copied up
to five times: each grid cell interpolated its four corners itself (1,444
interpolations a patch for 400 points), cloned all six vertices of its two
triangles, and the mesh sampler cloned each vertex again to move the triangle
into device space. Profiling page 8 of the annual report in #1584 put a
quarter of its time in those clones.
Each grid point's colours are now interpolated once, a cell's second triangle
takes the vertices its first copied, and the sampler moves a patch's freshly
made triangles instead of copying them (`TriangleSampler::sample_owned`; a
free-form triangle mesh, which the shading keeps, is still copied). The same
inputs give the same values: rendering the first three pages of every pdf.js
corpus file is byte for byte the same. Release build, Linux: report page 8
773 -> 696 ms, personwithdog.pdf 149 -> 129 ms, issue13520.pdf 104 -> 87 ms.
The new test pins the colours near each corner, at the centre and between them
for a Coons and a tensor-product patch (taken before the change); swapping two
vertices' grid points makes it fail.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MoEx9Nj8KC4WKDcf5YaokQ
* docs: roadmap entries for faster sampled functions, image decodes and patch meshes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MoEx9Nj8KC4WKDcf5YaokQ
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A cover counted only when it hid 40% of the line's words or box, so a
black box over one number in a long line, and an unapplied redaction
mark over it, audited clean while the number was still extractable. A
cover that hides half of any one word now counts as well.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RUNQSPVXx3dDBabRZHAnjz
Resolve independent progress-document conflicts while retaining upstream updates and the existing regression-tested fix. Refresh the relevant milestone notes; feature scores remain unchanged.
Fresh checks: cargo fmt --all -- --check, git diff --check, and layering/assets/parity gates. Focused production/test files and lockfiles match the prior green CI head. Full CI will rerun on this merge.
