Commit Graph
275 Commits
Author SHA1 Message Date
Adithya Shankaran 8a42e8a1ac Refuse a network path before the automation roots open it
A UNC, device or non-disk verbatim path was canonicalized before the roots
refused it, so on Windows the machine contacted a host the agent chose.
Those prefixes are refused first, unless a root is already on the same
server and share. A link that leads to one is refused from the link's text,
including when the link is only an ancestor of the name. The Windows tests
drop the borrows clippy rejects, and cover a junction, an alternate stream
and a Kelvin-sign name. The command walk also notices a new file when the
folder's modification time has not moved.
2026-11-06 16:33:34 +00:00
Adithya Shankaran 8314fb87ef Search fonts beside a drawing when they stay inside the roots, and keep the walk honest
Under the roots the folder was skipped outright, so a drawing lost its own SHX fonts. It is now
skipped only if a font there leads out. The fixture deleted the running server's working directory
on reset, so a relative path met a missing folder and not the roots. Test folders now clean up when
a test fails. Three NO_FILES entries I added on the last merge were duplicates: those commands
already had a row.
2026-11-06 14:40:07 +05:30
Adithya Shankaran 0479ddc6ca Merge remote-tracking branch 'origin/main' into carry/1472-automation-roots 2026-11-06 13:09:34 +05:30
Tanish Manku befb9b6f61 Merge branch 'pr-2016' into land/2016 2026-11-05 14:36:35 -07:00
Tanish Manku ec0decfd5c Merge branch 'pr-1971' into land/2016 2026-11-05 14:36:35 -07:00
Adithya Shankaran 09e4f90740 Hold automation roots to exact names, the folder a name stands in, and every device name
A case-sensitive Windows folder can hold Root beside root, so names are now compared exactly. The
superscript COM and LPT ports and the NUL: form are device names too. A write must stand under the
write root as well as lead there: a link outside pointing in had its .bak written beside it, and a
plotter delete removed it. Paths reach the check as paths rather than as text, -WBLOCK no longer says
whether a file outside the roots is there, and SHX fonts beside a drawing are not searched under the
roots, since one can be a link leading out.
2026-11-06 01:50:01 +05:30
Adithya Shankaran c03fb712eb Merge remote-tracking branch 'origin/main' into carry/1472-automation-roots
# Conflicts:
#	crates/mcp/src/backend.rs
2026-11-05 23:52:39 +05:30
Zinadin ZidanandClaude Opus 5.5 04a279cff0 fix(mcp): a render with a plot style plots on white; "none" turns plot styles off (#803)
A plot style alone drew black lines on the dark canvas: the tables that
change colours print colour 7 (and, in monochrome.ctb, every colour)
black. With `plot_style` and no `background`, the render now draws on
white, as a plot prints on paper; a given background still wins.

`plot_style: "none"` (or "") draws every object with its own
properties, also in a layout that shows its own plot styles. An unknown
table's error lists the built-in tables and at most ten of the
drawing's own. The app's ui.render test now draws a red line, so it
fails if the plot style isn't passed on.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 12:09:31 +06:00
Zinadin ZidanandClaude Opus 5.5 017cc9b1fc fix(mcp): render takes a background and a plot style (#803)
The MCP render tool and ui.render, in the app and headless, always drew
on the dark canvas with no plot style, so a PNG placed on a page needed
the SVG export instead.

They now take `background` ("dark", the default; "white", "paper" or
"#rrggbb") and `plot_style` (a built-in plot style table such as
monochrome.ctb or grayscale.ctb, or one the drawing holds). Colour 7 is
drawn black on a light background and white on a dark one, as on a plot.
An unknown value is refused with what is accepted. Both copies of
ui.render read the options through one function in the engine
(render_look). The render tool's description also says it returns an
image block, not text.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 11:20:52 +06:00
Davit MatchakhelidzeandClaude Sonnet 5.5 fe67c048ec Leave ROADMAP.md and docs/localization-parity.md to maintainers
CLAUDE.md: PRs don't edit these docs. The affected rows are named in the PR description.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YMTAbGNRxnMEzRvYnQGphF
2026-11-05 09:01:04 +04:00
Davit Matchakhelidze 2c0fe29a02 Merge remote-tracking branch 'origin/main' into ka-language 2026-11-05 09:00:31 +04:00
Davit Matchakhelidze 7af5718b20 Merge remote-tracking branch 'origin/main' into fonts-georgian 2026-11-05 09:00:31 +04:00
Adithya Shankaran f09c50d6b8 mcp: refuse unclassified commands under the automation roots and walk the registry
Bring the rest of the file access under the roots:

- Image references had no check at all, and the loader threads lost the roots,
  which are per thread. Both are held to them now.
- The xref and image search looked outside the roots before any check, listing
  folders and answering whether a file exists. It now skips those places.
- setvar SAVEFILEPATH could point autosave and recovery anywhere, so it is
  refused under roots, and so is drawingrecovery.
- A command runs under roots only if file_access/commands.rs classifies it.
  Anything else is refused, whichever route starts it.

every_command_is_classified walks the registry for the table. mcp_roots_walk.rs
walks it through the real server with paths outside the roots, seeing reads by
access time and writes by metadata. The same walk with no roots is the control:
it has to find the same commands escaping.

Also drops two duplicated checks the rebase left in template.rs and compare.rs.
2026-11-04 23:56:13 +05:30
Nicholas Gower 0f33f430a7 security: confine plotter settings I/O and refuse the write root itself
- PLOTTERMANAGER's settings store now checks read access to the settings
  folder and each file, write access to the target before the folder is
  created, and write access before deletion (#1472 review). list surfaces a
  refused folder as an ordinary error.
- Reads are allowed under the read or the write root; writes stay under
  the write root. A write to the write root itself is refused, since its
  temporary file would land beside the root, outside it.
- Tests cover reads under either root, refusing the root as a write
  target, no temporary file outside the root, and every plotter action
  refused when the shared settings folder is outside the roots.
- Docs and CLI help describe the same rules.

This is 72961da from the #1904 branch, which the 2 Nov rebase dropped, ported
to main's atomic_write. Main's writers now use a fresh temporary file beside
the target, so the separate temporary-path checks are gone.
2026-11-04 22:26:20 +05:30
Nicholas Gower 46ee1a817d mcp: port the automation roots onto the clap CLI
The CLI's argument parsing moved to clap on main, so the hand-rolled parser
and its McpTarget change are rebuilt as two options, a usage error naming
to the modules main split out (wblock, import, export, plot, saveas,
compare, template, the LISP file builtins).
2026-11-04 22:14:27 +05:30
Tanish Manku 0d72152142 Merge main into land/2354 2026-11-03 18:21:00 -07:00
Tanish Manku 100d812155 Merge main into land/2354 2026-11-03 18:07:17 -07:00
Tanish Manku de753897c4 Merge main into land/1944 2026-11-03 18:00:03 -07:00
Tanish Manku 76d09aaddf Merge remote-tracking branch 'origin/main' into land/2354 2026-11-03 17:22:36 -07:00
Tanish Manku c2d62a2842 Merge main into land/1944 2026-11-03 17:11:42 -07:00
Tanish Manku 68e01dd487 Merge main into land/1230 2026-11-03 17:11:25 -07:00
Tanish Manku 8cdcd6f85c Merge #2354 (XCLIP and XCLIPFRAME) onto main 2026-11-03 15:00:20 -07:00
Tanish Manku 28421dd1d7 Merge #1944 (SHX shape fonts, big fonts and unifonts; linetype shapes) onto main 2026-11-03 14:59:22 -07:00
Tanish Manku 4c00b6732f Merge #2505 (raster images drawn on screen and in PNG output) onto main 2026-11-03 14:52:15 -07:00
Tanish Manku 58bb1e065e docs: date the view twist revision rows to the day they land 2026-11-03 14:50:59 -07:00
Davit MatchakhelidzeandClaude Opus 5.5 f9cb70500d docs: Georgian font fallback in the Georgian section
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 11:17:59 +04:00
Davit MatchakhelidzeandClaude Opus 5.5 e8408851cd docs: the Georgian face paragraph describes glyph support only
Choosing ka, its English fallback and the picker name come with the
language PR (#2016), which registers ka; this PR alone doesn't.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 11:17:59 +04:00
Davit MatchakhelidzeandClaude Opus 5.5 4b532f6c24 i18n(ka): translate the strings added upstream; prompts without the trailing "or"
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 08:01:13 +04:00
Davit MatchakhelidzeandClaude Opus 5.5 891e87a7db Merge upstream main into ka-language
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 08:01:13 +04:00
Davit MatchakhelidzeandClaude Opus 5.5 e931787a64 Merge upstream main into fonts-georgian
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 08:00:05 +04:00
mattlafy cbc9db75b1 Merge remote-tracking branch 'upstream/main' into pr/dview
# Conflicts:
#	crates/engine/src/prompt.rs
#	crates/engine/src/session/execute.rs
#	crates/io/src/dxf_read/entities.rs
#	crates/ui-egui/src/canvas.rs
#	crates/ui-egui/src/cmdline.rs
2026-11-02 21:53:26 -05:00
Books 2ac500925b DWG: a table cell's text height is an override only when the file flags it (part of #2501) (#2588)
* DWG: a table cell's text height is an override only when the file flags it (#2501)

acadrust's DXF writer writes group 140 for any cell with a style record, even a
default 0.18 with no override flag, and CADCraft read it as a per-cell height,
so cells ignored the table style's height. The bridge now clears the height of a
cell style that has neither the TEXT_HEIGHT property flag nor the content's
height override bit (0 reads as no override).

* Parity checklist: regenerated after merging main

* Tests: inherited table cell heights must stay unset (review on #2588)

A cell that inherits its height must read back as None, not an explicit 0.0625 override.

* Parity checklist: regenerated after merging main

* Test: a cell style with the TEXT_HEIGHT flag, content override bit 128 and a cell with no style record keep the right height (review on #2588)

* DWG: a table cell's content height override (group 144) is kept (review on #2588)

* Parity checklist: regenerated after merging main

* DWG: table cell height takes the first content's override and the style's override flags (review on #2588)

Only the first content's bit-128 height stands for the cell (one height for its text); an
invalid content height no longer keeps the writer's default 0.18; a height marked only in
the style's override flags (newer files) is kept.
2026-11-02 15:01:57 -07:00
STE1200andClaude Opus 5.5 5ec0d215d8 CLI: info and convert report what a damaged DWG left out (read notes, #859) (#2337)
A damaged DWG opens partially since #859 and the drawing's read notes
say how many object records were left out. OPEN echoes them and MCP
`open` returns them as `warnings`, but `cadcraft-cli info` and
`convert` stayed silent: a corpus round trip wrote thousands of skipped
records away without a word.

- convert: each read note goes to stderr as `IN: <note>`, next to the
  skipped-records warning (#872). The exit code stays 0.
- drawing.inspect (and so info, MCP inspect_drawing and the control
  channel): new field `readNotes`, next to `skipped`. Existing fields
  are unchanged.
- run already prints OPEN's notes with the session log.
- README and docs/control-protocol.md say where the notes appear.

Part of #1039.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-02 15:01:51 -07:00
RicSchonfelder 6660d88785 i18n(pt-BR): translate the rows the sibling catalogs cover (#2581)
* i18n(pt-BR): translate the rows the sibling catalogs cover

The Brazilian Portuguese catalog covered the reference catalog and every marked UI literal, but 191
strings that the German, French, Swedish and Ukrainian catalogs already translate still fell back to
English: the Hatch and Gradient dialog, the Block Editor and dynamic-block grips, the plot and drawing
units dialogs, multileader editing, selection preview settings and the unit names.

Rows follow the catalog's vocabulary (dimension = cota, hatch = hachura, grip = pinça, layout =
apresentação, multileader = multilinha, xref = referência externa), keep the prompt keywords, SPDX
tokens and technology names in English, and are appended after the existing entries.

* docs: leave localization parity updates to maintainers
2026-11-02 14:36:48 -07:00
aiongg 9b32882065 docs: tidy wording 2026-10-31 04:25:44 +00:00
aiongg ee715f4636 docs: tidy agent, contributor and asset docs 2026-10-31 04:17:25 +00:00
Tanish Manku c808939bd1 Merge remote-tracking branch 'origin/main' into pr-1693 2026-10-30 19:25:26 -07:00
Tanish Manku 8822462be2 Merge main into #1693 2026-10-30 19:19:59 -07:00
Tanish Manku e348a30ebd Merge main into #2460 2026-10-30 19:19:54 -07:00
1ecc2a130b DXF: no CLASS instance counts in R2000 files (AutoCAD LT 2026 hung opening them) (#1060)
* DXF: no CLASS instance counts in R2000 files (AutoCAD LT 2026 hung on them)

Since #635 every CLASS record carries group 91, the instance count, at every
version. The count came with the 2004 format: R2000 CLASS records are 1, 2, 3,
90, 280, 281. AutoCAD LT 2026 stops responding when it opens an R2000 DXF
whose CLASS records have a 91, so every R2000 file CADCraft saved locked it up
(a drawing opened from an R2000 file is saved as R2000 again).

Isolated black-box in AutoCAD LT 2026 with synthetic files: the same R2000
file without the 91 groups opens, as does the same content declared AC1032.

The counts are now written only when the file (or the DWG the bridge writes)
is 2004 or later; R2000 output drops the group from every CLASS record,
classes kept from a newer file included.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Test: binary R2000 drops group 91 from a kept foreign CLASS too

Review follow-up: the kept-class check covered R2000 only through ASCII
saves; it now also saves the fixture as binary R2000 and checks that the
ACME_OBJECT CLASS record is 1, 2, 3, 90, 280, 281.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Test: binary 2018 keeps a kept foreign CLASS's instance count

Review follow-up: the kept-class check now saves the fixture as binary at
both R2000 (no group 91) and 2018 (group 91 after the proxy flags, with the
class's count), through one helper.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Slopapalooza <307952167+Slopapalooza@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 19:16:30 -07:00
Tanish Manku 47ed3af26c Parity docs are updated by maintainers in batches, not in each PR (#2520)
docs/gaps.md conflicts in 153 open PRs and docs/ui-parity.md in 75, because every PR had to edit them. A PR now names its rows in the description.
2026-10-30 19:11:01 -07:00
Tanish MankuandAldo Zarate 3d623fc4d4 Saving over a DWG asks before it loses content, and saves keep a .bak (#2518)
Includes the .bak work from #1003.

Co-authored-by: Aldo Zarate <azarate@azf.com.mx>
2026-10-30 19:10:57 -07:00
Davit MatchakhelidzeandClaude Opus 5.5 5f6dbede5c docs: Georgian row count after the block-editor prompt
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YMTAbGNRxnMEzRvYnQGphF
2026-10-31 06:00:43 +04:00
Davit MatchakhelidzeandClaude Opus 5.5 651615e5c2 docs: Greek and Bosnian in the language lists; Georgian on the web comes with #1971
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YMTAbGNRxnMEzRvYnQGphF
2026-10-31 05:00:18 +04:00
Davit MatchakhelidzeandClaude Opus 5.5 66f65b763e docs: list bs beside ka in the supported interface languages
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YMTAbGNRxnMEzRvYnQGphF
2026-10-31 04:38:42 +04:00
Davit MatchakhelidzeandClaude Opus 5.5 126a734b5c Merge origin/main into ka-language
Keeps Bosnian and Georgian side by side; adds the new block-editor save prompt to ka.tsv.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YMTAbGNRxnMEzRvYnQGphF
2026-10-31 02:52:48 +04:00
Davit Matchakhelidze 6135745a1a Merge upstream main into fonts-georgian 2026-10-31 02:44:27 +04:00
43c41b8e68 bs: Add Bosnian (bs) to the interface languages (#2502)
* Add Bosnian (bs) to the interface languages

bs.tsv: Bosnian (Latin script, Ijekavian) catalog covering every key of
CADCraft's other catalogs, its wording shared with storytold/craft-i18n
locales/bs/ and the strings craft-i18n lacks translated new. Every bs-*
locale selects it; counts use one/few/other. Locale and plural tests,
localization docs and an attribution entry (AI-assisted, not yet
reviewed by a native speaker).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013Tayw8nmxoLLmKv13jK93z

* docs(localization-parity): Bosnian in the Languages table (CodeRabbit)

---------

Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
2026-10-30 15:46:30 -04:00
Books 8252d0e9bd Extents: an MTEXT's wrap width only narrows its box (#2465) (#2491)
* Extents: an MTEXT's wrap width only narrows its box (#2465)

A garbage group 41 width (3e14) became the box width, so the drawing extents were 3e14
wide. The width is now capped at the longest line's estimate. The width 0 branch is
unchanged. Tests: a_huge_mtext_wrap_width_does_not_inflate_the_extents and
a_narrow_mtext_wrap_width_still_sets_the_box_width.

* Parity checklist: regenerated after merging main

* Parity checklist: regenerated after merging main
2026-10-30 15:46:20 -04:00
2ez4gcxandClaude Opus 5.5 6ec6d97ea0 LISP: getvar and setvar take angle variables in radians (#2482)
* LISP: getvar and setvar take angle variables in radians

CADCraft keeps ANGBASE, SNAPANG, POLARANG, HPANG, CHAMFERD and
DIMJOGANG in degrees as far as sysvars::get/set go, which is right for
the SETVAR command, but AutoLISP's getvar returned those degrees and
setvar took them: (getvar "DIMJOGANG") gave 45.0, and
(setvar "ANGBASE" (/ pi 2)) set 1.57 degrees.

AutoCAD 2024 gives and takes them in radians from AutoLISP: each set to
30 by SETVAR, or to (/ pi 6) by setvar, reads 0.5235987755982987. The
LISP bridge now converts these six; the SETVAR command is unchanged.
getangle_is_relative_to_angbase sets ANGBASE in radians accordingly.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* SETVAR: keep real dimension variables real

A dimension variable whose style field is a real (DIMJOGANG, DIMSCALE
and the other real DIM* fields) is now stored as a real header value,
even when the first value set is a whole number. SETVAR DIMJOGANG 45
followed by (setvar "DIMJOGANG" 0.5) no longer rejects the fractional
degrees the radian value gives.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* SETVAR: store the value the dimension style took

A real dimension variable now stores the value its style field took after validation, so any input the style accepts (true for DIMSCALE gives 1) is stored as it was before.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 15:45:53 -04:00