- objects are found through a handle index when reading and writing clips, and removed in one pass (no scan per reference)
- the writer's handle seed starts past the clip objects, so a drawing with many clipped references gets no duplicate handles
- a clipped copy that runs out of the block budget marks the list truncated, like an unclipped one
- clipped block contents keep their own layers in plots
Review fixes from CodeRabbit on #2748:
- DVIEW TWist and view.set refuse to turn a locked active viewport ("The viewport is locked.")
- the untiled *Active VPORT record fits the extents rotated by the twist it writes, with the twist normalised
* MINSERT: a block reference repeated in rows and columns
MINSERT asks INSERT's questions, then the grid's as AutoCAD 2024 does
(measured with accoreconsole): the numbers of rows and columns (1 to
32767), the distance between rows or a unit cell when there is more than
one row, the distance between columns when there is more than one column,
then the attribute values once. A distance along an axis with one copy is
0, and one row by one column is a plain reference.
EXPLODE leaves a grid whole ("The object is minserted.") instead of
exploding its first copy and dropping the rest. MIRROR mirrors a block
reference as AutoCAD does, negating the X scale factor and a grid's
column distance, so a mirrored grid's rows stay on the right side. An
attribute with no default is asked without "<>".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* MINSERT JSON: refuse a distance that is not a number
`rowSpacing` and `columnSpacing` went through `f64_or`, which takes any
value that is not a number as the default, so `"rowSpacing": "15"` stored 0
and put every copy on top of the others without an error. They are now
checked as `rows` and `columns` are: absent or null is 0, a finite number is
kept, anything else is refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* EXPLODE: a block reference with different X and Y scales gives ellipses, not circles
EXPLODE of a block reference whose X and Y scales differ kept its circles
as circles with the mean scale, its arcs as circular arcs, its ellipses
with their old ratio and its polyline arcs as circular bulges: not the
shape the reference draws. BURST and INSERT with explode go the same way.
When the reference's transform isn't a similarity, a circle now becomes a
full ELLIPSE, an arc an elliptical arc, an ellipse gets its true axes, and
a polyline with arc segments (and no widths) its lines and elliptical
arcs. Everything else, and every evenly scaled or mirrored reference,
transforms as before.
The exact ellipse transform and the similarity test move unchanged from
snap.rs to cadcraft-geom (Ellipse::transformed, Mat3::similarity_scale)
and are shared by object snaps and EXPLODE.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* contributors: add Tyler-Chiriboga
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
EXTEND on an open polyline ran a ray along the chord of an arc end segment
and kept the old bulge, so the arc changed radius and centre, or the
boundary was reported missed when only the circle reached it. The arc end
segment now runs on round its own circle to the nearest crossing ahead,
turning the same way, and its bulge becomes that of the total turn, as
EXTEND does with an ARC. The polyline also keeps its constant width,
elevation and linetype generation, which the rebuilt polyline dropped.
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Dynamic blocks: index kept objects once per listing and copy each value record once
Listing blocks rebuilt a handle index of every kept foreign object for each
block with an extension dictionary, multiplying the block count by the object
count. The listing now builds one ObjectIndex and asks is_dynamic_block_in per
block; dynamic_block answers None before indexing when the block has no
extension dictionary.
templates() copied a value record's tags once per dictionary key, so a record
listed under many keys was copied many times, and it looked up each
dictionary entry by scanning every object. It now resolves entries through
its handle index and copies each record at most once (CWE-400, CWE-407).
* Dynamic blocks: list a block as dynamic from its graph object without walking the graph's members
From review of #2670. Changing SAVEFILEPATH released the old folder's lock while
files remained there, so another app could offer a running app's drawings. A
write failing while the app was idle waited for the next input to be shown. A
second drawing's failure stayed hidden behind a dismissed notice.
The frame now runs under a guard. A panic autosaves unsaved drawings at once,
even with SAVETIME 0, and shows a notice that doesn't block saving. A panic
hook writes a local report for every panic. Panics that keep coming stop the
interface and leave a window that closes, with the recovery copies kept.
* Autosave: a failed write is shown until one works (E6)
A failure only reached the log, so someone could work for hours with no
recovery copy. The autosaver keeps why each drawing's write failed; the app
says it once at the command line and shows a notice that doesn't block saving.
It ends when a write works or the drawing no longer needs a copy.
* Crash recovery: offered at the first relaunch, not 5 minutes later (E5)
A crashed app's lock file looked alive until it went stale, so a relaunch
within 5 minutes skipped its autosaves, and a suspended app's looked dead.
A running app now holds a system lock on the file, which the OS drops however
the process ends. File age still decides where locks don't work and for lock
files of older versions, which don't record holding one.
The folders are made by a new writer only user-named targets use, so autosave
and a dangling symlink's target keep the strict write_atomic. Save, Save As,
Export, Plot, WBLOCK, layout templates, render and the CLI report the topmost
new folder at the command line and in createdFolder.
A DIMENSION moved by an older AutoCAD keeps its *D block contents where
they were and the move in DXF group 12, and is drawn translated by the
insertion point. Group 12 was never read into Dimension, so the block
rendered at its pre-move position and the extents reached there too.
Read group 12 into a new Dimension::insertion_point, draw the block
translated by it, move its bounds with it in the extents, test the
block's text where it draws, and write group 12 back whenever the block
written is the one read from the file. Regenerated geometry (edits drop
the block link) sits at its definition points, so no offset is kept or
written for it.
Co-authored-by: rahul05ranjan <11793455+rahul05ranjan@users.noreply.github.com>
* OVERKILL: objects within the tolerance are duplicates wherever their numbers fall
OVERKILL compared objects by rounding every number to a multiple of the
tolerance. Two numbers closer than the tolerance on either side of a
rounding step round apart, so the objects were kept: at 0.01, circles at
x = 80.001 and 80.003 were merged but circles at 50.004 and 50.006 were
not. Arcs on one circle were bucketed the same way.
Duplicates are now found by shape (the JSON without its floating-point
numbers; counts and flags must match exactly) and a weighted sum of the
numbers on a grid of the tolerance, looking in the neighbouring cells too,
then compared pairwise to within the tolerance. Arcs and circles look in
the neighbouring cells of centre and radius and are grouped with a
union-find. Lines were already grouped by sorting and are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* OVERKILL: within the tolerance doesn't chain
Arcs and circles were grouped with a union-find over pairwise matches,
so A ~ B and B ~ C put A and C in one group even when A and C are
farther apart than the tolerance: a circle C was deleted for matching
only the arc between it and the kept circle A, and arcs were combined
across the chain. Each object now joins the first group whose every
member is within the tolerance of it, or starts its own.
In the duplicates pass, a followed object displaced an earlier keeper
that later objects had already matched, leaving them deleted up to twice
the tolerance from what stays. Followed objects are now looked at first,
so every deletion is within the tolerance of an object that stays.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* feat(inquiry): AREA, PERIMETER and DISTANCE system variables, as AutoCAD sets them
AREA, DIST and MEASUREGEOM now leave AREA, PERIMETER and DISTANCE for
getvar, read-only and not saved: the area just measured (the running
total in Add and Subtract mode) with the last area's perimeter, and the
distance (the total of Multiple points). After DIST's two points
LASTPOINT stays at the first point, as measured. AREA's Object option
measures splines, closed by their chord like open polylines.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Address review: spline area over narrow spans, finite distances, list order
- AREA of a spline samples it span by span (Spline::tessellate_spans), so
a narrow knot span no longer drops out of the area. The spread over the
spans is one helper shared with Spline::bounds, whose boxes are
unchanged.
- DIST refuses a distance that isn't finite on all three paths and leaves
DISTANCE as it was.
- sysvars::list puts the session's values after the drawing's header, so
it shows what get returns (AREA, ORTHOMODE, LWDISPLAY...).
- LASTPOINT after DIST's Multiple points is pinned to AutoCAD 2024's
measured behaviour: the last point entered, also after Length and Undo.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Address review: measure every knot span of a spline in AREA
A spline with more knot spans than half the sample cap fell back to an
even grid over the domain, which can step over a narrow span and measure
the wrong area: a 100 x 100 square as a degree-1 spline, its right and
top edges two spans a millionth wide after 4,500 spans along the bottom,
measured 4,998.75.
Spread the samples over the spans at any span count: at least one
segment per span, within the cap up to MAX_SAMPLES spans and one each
past it, as Spline::length integrates every span. One evaluation per
span costs 85-285 ms on a million spans in a release build, against
32-52 s for the length AREA already takes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Address review: share AREA's spline samples by how much each span bends
Past the sample cap, tessellate_spans gave each knot span one segment, so
a bent span among many straight ones was measured by its chord. Example:
a quadratic spline along the x-axis, one control point 1000 above it,
past 4,500 spans. It measured 750; the area is 1750/3.
Each span now keeps one segment. A span whose control points are
collinear is straight, so its chord is exact. tessellate's count is
shared among the spans that bend, by the cube root of a measure of their
control points' convex hull, which holds the span and its chord. The
shares are rounded on the running total, so they use the whole count
however many spans share it. The example now measures 583.33343, below
the cap and past it.
The spreading by parameter width that this replaces goes, and
span_params with it: bounds is main's code again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* DIST: a measurement refused as too large leaves LASTPOINT at the first point
Session::feed sets LASTPOINT to the second point before DIST checks the
distance, and the refusal returned before DIST put the first point back,
as it does after a measurement. It now does so before refusing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* AREA: a spline too long to measure is too large, not a perimeter of 0
Spline::length reads a length past f64's range as 0, and AREA took that
as the spline's perimeter. Spline::checked_length returns None there,
and AREA reports the perimeter too large, as it does a polyline's.
The overflow was hidden twice: in the sum, and before it in the
derivative, which read as zero speed once past f64's range. The length
integral now takes the derivative before that (Spline::velocity), so an
overflowing speed counts too. param_at_length keeps the speed it had,
and finite lengths are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Say "or" once in prompts that offer options
Twelve prompt messages ended in "or", and the command line adds "or"
before a prompt's options, so they read "or or": HELIX height, XLINE angle
and offset, BLEND, four LENGTHEN prompts, two NCOPY prompts, PEDIT and
table breaks. They now read as AutoCAD's do.
The interface catalogs are keyed by these messages: each key loses its
"or" and each translation its trailing word for "or"; where the shorter
key already had a translation, the longer row is removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Say "or" once in HELIX's radius prompts; rename the Vietnamese keys
HELIX's base and top radius prompts ended in "or" before their Diameter
option too, so they read "Specify base radius or or [Diameter]". They
now read "Specify base radius or [Diameter] <1.0000>:", and their keys
lose "or" in all 17 catalogs, as the other twelve did.
The Vietnamese catalog, merged from main, gets the same renames as the
other catalogs; where the shorter key already had a translation, the
longer row goes. The unused "Specify rotation angle or" row goes from
every catalog too: RECTANG has asked "Specify rotation angle" since
#2554.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Write option prompts without "or", as AutoCAD does
AutoCAD writes " or [" only when a prompt asks for something besides its keywords ("Enter block
name or [?]"); a prompt asking for one of them has none ("Enter an option [...]", "Enter layout
option [...]", "New color [Truecolor/COlorbook]"). The 27 messages read without "or" from
AutoCAD 2024's command line are listed in NO_OR and displayed that way.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Address review: the window's command line drops "or" as the prompt does
The command line in the window draws a prompt's keywords itself and
wrote " or [" after every message, so -LAYER still read "Enter an option
or [?/Make/...]" there, and questions such as AUDIT's had "or" too. Only
the prompt text (the control channel, the state JSON, the tests) had
dropped it.
Prompt::keyword_lead gives what goes before the keywords; display() and
the command line both use it, and the command line translates only
" or [". A render test reads what the command line draws, in English and
German, for -LAYER, LINE and AUDIT. Suggested in review.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The interactive save only knew about DWG, so overwriting a 2013+ DXF lost SAT-text
solids on a command-line note. loss_on_save now lists them, and EXPORT reports them
with the DXF wording.
Found while testing crates/engine/src/cmd/array.rs:380 (run_arraypolar, `fill: f64_or(p, "angle", 360.0)`)
Co-authored-by: Claude Haiku 5.5 <noreply@anthropic.com>
* DWG save keeps OLE objects (OLE2FRAME): stop reporting them as not saved
acadrust reads OLE2FRAME from our DXF into its own entity and writes it to
DWG of every version we save, frame and embedded data unchanged; only the
"Not saved in the DWG file" warning (and the QSAVE loss prompt) listed it.
not_saved_as now skips the types the DWG bridge writes (DWG_WRITES).
Tests: a synthetic OLE2FRAME through DWG (resaved, and as 2000 and 2010
DWG) keeps every byte; OLE objects are not reported at any version.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* DWG_WRITES matches the type as stored, not case-insensitively
The DXF given to the DWG bridge carries an unknown entity's type as stored, and
acadrust reads only "OLE2FRAME" into its OLE entity: any other spelling becomes
its Unknown and is dropped, so it must stay in the "Not saved" report. The test
adds a lower-case "ole2frame" that is still listed. (Review comment on #2582.)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(engine): WBLOCK writes only what the block or selection uses (#2423)
WBLOCK of a block, or of selected objects, copied the source drawing's
whole layer, linetype, style and block tables into the new file, so a
5-object symbol carried 40 unused layers and 20 blocks. The written
drawing is now purged of what nothing in it uses, as AutoCAD's WBLOCK
writes it: purge::purge_unused runs PURGE All's scan on a plain Drawing
until a pass removes nothing (a block can free its layer), keeping the
defaults and the current ones. Writing the entire drawing is unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(engine): PURGE's scan counts leaders' dimension styles and styles used by other styles
From the review of #2621: PURGE (and now WBLOCK, which runs its scan)
took as unused a dimension style only a LEADER used, a linetype only a
dimension style or a dimension's override named (DIMLTYPE, DIMLTEX1,
DIMLTEX2), and a text style only a complex linetype's text used, and
removed them while the objects still named them. The scan counts all
three now; text_styles_in_use, which STYLE's delete also asks, includes
complex linetypes' text styles.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(engine): PURGE's scan counts multiline styles' linetypes and table styles' text styles
From the review of #2621: a linetype only a multiline style's element
used, and a text style only a table style's cell style used, were taken
as unused and purged while the styles still named them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: austin62jake <340374338+austin62jake@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* feat(engine): INSERT of a drawing file keeps an attribute definition's field
The attribute definitions of a drawing file read by INSERT name=file lost their
extension dictionary, so the ACAD_FIELD dictionary and the FIELD objects of an attribute
whose value is a field (the file name in a title block) were gone from the block.
adopt() now copies that object tree to the drawing under new handles, with the FIELD
class and a field list if the drawing has none.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* foreign: index the kept objects once when walking and copying a dictionary tree
A file can keep many objects; a linear lookup per pointer grew with their product.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: andrewbrake <6925387+andrewbrake@users.noreply.github.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
* OOPS brings back what -WBLOCK erased
-WBLOCK erases the objects it writes, and OOPS brings back what the
last ERASE or BLOCK removed; the two landed separately, so OOPS never
saw what -WBLOCK erased. -WBLOCK now erases through erase_kept, as ERASE
and BLOCK do.
As AutoCAD 2024 does (accoreconsole, the drawing's entnext order read
back): OOPS after -WBLOCK brings its objects back, each where it was and
with its handle; only the last erase counts, -WBLOCK's or ERASE's; and
-WBLOCK * erases nothing, so OOPS still brings back the erase before it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* OOPS: keep the last erase through a refused REFEDIT command; Block Editor
During REFEDIT, a command that removes an object outside the working set
is undone (refedit::after_command), but erase_kept had already replaced
what OOPS kept: the objects came back, and the erase before them was
lost. erase_kept now leaves OOPS as it was when an object it removes is
outside the working set. That covers BLOCK, -WBLOCK, ERASE and the
Delete key.
In the Block Editor, its objects keep the handles they have in the block
definition, so in_blocks found them there and OOPS brought back nothing
erased in the editor. in_blocks now skips the block open there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Polyface meshes stay one object through DXF and DWG
A polyface mesh (POLYLINE flag 64) was read as loose 3D faces. It is now
read, drawn, exploded into its 3D faces and written back as a POLYLINE.
Hidden edges aren't drawn. A mesh too large for the 16-bit groups 71-74
is saved as 3D faces.
Split from #1005.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* A polyface face ends at its first zero index
A face record with a gap ([1, 2, 0, 3]) has two corners and is dropped,
not read as a triangle.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Polyface meshes: write only faces that name their vertices; ID gives their Z
The writer drops the faces the reader would drop, so no record names a
missing vertex, and the 16-bit check only needs the counts. ID and DIST
give a mesh the Z of its nearest visible edge, as they do for 3D faces.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
There was no OOPS. It now brings back what the last ERASE removed (the
Delete key too), or BLOCK deleting the objects it makes a block of, as
AutoCAD 2024 does (accoreconsole, the drawing's entnext order read
back):
- Each object keeps its handle and goes back directly above the object
that was under it, or under the one that was above it when that is
gone. An erase records both neighbours (EntityStore::neighbours).
- Only the last erase counts; entdel doesn't. OOPS says nothing and is
one undo step.
- An object stays out when it is already in the drawing, when an undo
took the drawing back to before it was made, when its layer, block or
layout is gone, and, during REFEDIT, when it isn't in the working set.
REFCLOSE forgets the working copies; the Block Editor has an OOPS of
its own.
- The command's label is in the 17 interface catalogs.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
Measured in AutoCAD 2024 (accoreconsole):
- N, E, S and W are taken on their own in any angle format (LINE to
@10<N goes north at AUNITS 0, 1 and 4).
- A bearing or a compass point is an angle like a number, measured from
ANGBASE in the ANGDIR sense: with ANGBASE 90 and ANGDIR 1, @10<N45dE
draws at 45 degrees and @10<N east; (angtof "N") is 0 there.
- <<< takes them from east counterclockwise; << (degrees) refuses them.
main took the letters in surveyor's units only, and read them and
bearings as absolute directions. One cardinal() now serves every angle
format, and the test that pinned the surveyor-only reading expects the
measured values. -UNITS uses the shared parser.
Also write the -UNITS tests' scripts with \n escapes again; they had
become multi-line strings.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Inside a routine, commands recorded no undo step (the routine becomes
one when it ends), so U inside a routine took back the step from before
the routine: after drawing a line D, ERASE B then U in one routine
brought B back and removed D.
As AutoCAD 2024 does (accoreconsole, the drawing read back after one
routine), U inside a routine now takes back the routine's last command,
with what the routine did after it (entdel, entmake); U again, UNDO n
and REDO go on through its commands. With none left, U takes back the
drawing's step before the routine, as before, and the routine's step
then starts from there. Once the routine ends it is still one undo step.
The routine's steps live in the session's LISP state (RoutineSteps), not
in the drawing's undo stack, so a long routine doesn't push the drawing's
earlier steps out. A command inside the routine ends what REDO could
bring back; UNDO Control One and None apply. DocState::swap_in is the
one restore step undo, redo and the routine share.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Tanish Manku