What the first Linux CI run of the 1.6 tree found.
- fastmcp 3's enable()/disable() append a Visibility transform per call,
and a tool lookup walks the transform stack as nested calls. The
lifespan re-applies the tool profile on every client connection, so an
in-process server grew two layers per session until every tool call,
system_status included, raised RecursionError (CPython 3.11 counts C
frames against the limit). The profile's transforms now replace the
previous ones, as _apply_discovery_mode already did; a test switches
the profile 600 times and then calls a tool. The local suite runs in
half the time it took with the stack growing.
- 21 fake-ActiveX tests build a real pywin32 VARIANT; they now skip
without pywin32, like the suite's other VARIANT-dependent COM tests,
and the Windows lane keeps running them.
- A test spelled a path through a directory it never created, which only
Windows normalises away; it creates the directory.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd
Measured 2026-09-22 by scripts/smoke_settings_com.py --build-dwt against the
running AutoCAD 2026 (25.1s), each on a scratch Documents.Add() document:
1. RPC_E_CALL_REJECTED (-2147418111) on Documents.Count straight after
Documents.Open, after Close(False) inside document_close, and in
_ensure_document_state: AutoCAD's own message filter refuses incoming
calls while it is still switching documents. pywin32 exposes no
CoRegisterMessageFilter, so _wait_out_rejected_call waits the window out
for read-only callables (0.25 s -> 1 s pauses, 20 s budget, inside
COM_CALL_TIMEOUT) at those sites and after Documents.Add/Open; any other
com_error is not retried.
2. layer_list raised "'IAcadLayer' object has no attribute 'Color'. Did you
mean: 'color'?": acax25enu.tlb spells IAcadEntity/IAcadLayer `color` and
`Lineweight`. Late-bound dispatch resolved .Color/.LineWeight
case-insensitively; with a makepy cache present (this machine's gen_py was
regenerated on 2026-09-17, and every object a method returns is then a
generated wrapper) the lookup is exact. All 17 sites now use the typelib
spelling, which works on both wrappers; the fakes model the measured names.
3. layer_state_restore raised "'IAcadObject' object has no attribute
'Count'": Dictionaries.Item and IAcadDictionary.Item are declared
IAcadObject* (the narrowing _com_unnarrow already handles for MLEADER
styles). The save path went through the typed Dictionaries.Add, so only
the reopen paths failed. Both now un-narrow; the fakes narrow like pywin32.
4. The Freeze put on layer 0 raises 'Invalid layer' whatever the value, even
False while thawed (color/Linetype/Lineweight/Plottable/LayerOn/Lock/
ActiveLayer all fine on it), so a state saved with 0 current restored with
two warnings and 0 unapplied. Freeze is written only when it changes the
layer.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd
`_file_identity` keyed on normcase(abspath), which folds case and separators
only, and only on Windows. Two names for one file — a hard link (which
Path.resolve() cannot fold either), an NTFS junction reached in-process, or
a case variant on macOS's case-insensitive APFS where normcase is the
identity — registered two entries: both rows said `saved: True`, a save from
either overwrote the other's bytes, and closing the 'clean' one dropped its
work as `discarded_changes: False`.
Identity now asks the disk: `os.stat` (st_dev, st_ino) when the path exists
(the samefile identity), `normcase(abspath)` only for a path not on disk yet
(a save target, or an owner whose file was removed) or a filesystem that
reports no inode. `_register_document`, `_path_holder`,
`_resolve_document_key`, the `_doc_path` setter and the quarantine's
same-path warning all go through the one helper. An untitled entry holds no
file and never matches a path.
Tests: the `_case_folding_fs` guard probes the temp filesystem instead of
`os.path.normcase`, so the case-variant tests run on macOS where they used
to skip while the bug was live. New: hard link (backend and over the wire),
junction (mklink /J; a directory symlink elsewhere), a case variant with
normcase given its POSIX definition, and the spelling fallback for a save
target that is not on disk. Four of the five fail on the previous backend.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd
document_close's promise (never drop unsaved work unless told to) and the
`saved` row of document_list are only as good as _DocState.dirty, and three
headless mutators never set it: a new layout tab, the $TILEMODE / active
layout write, and a paper-space VIEWPORT all changed the file yet closed
without refusal, reported discarded_changes: False, and were absent on disk.
COM refused the same sequence (Saved is AutoCAD's DBMOD), so the two engines
disagreed. Each now calls _mark_dirty() after its write, which also pushes
the undo snapshot every other mutator already has.
The regression test runs the save -> mutate -> list/close/undo/redo -> close
with save sequence for all three and checks the change reached the file.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd
`drawing_apply_iso_layers` recorded its standard in `_active_layer_set` on
the backend, and `_role_layer` reads it ahead of the PlanSpec to decide where
`dimension_auto` / `construction_xline` file their output. It was neither a
`_DocState` field nor cleared by `_reset_document_state`, so an iso13567
sheet next to a mech one pushed the mech drawing's dimensions onto
`M-DIMEN-T-N` (created on demand), silently — and `drawing_critique` could
not see it because the same stale set drove the critique. The attribute
predates the registry (v1.5 `drawing_new` did not reset it either), but the
per-document claim of Task 21 covered it and the measured attribute list
missed it.
- `_DocState.active_layer_set` + `_ActiveField("active_layer_set")` on
`EzdxfBackend`: the set follows the active entry, bidirectionally.
- `PremiumContract._reset_document_state` clears it (declared on the mixin
like `_plan_spec`), which fixes COM's `_ensure_document_state` switch and
`drawing_new` / `drawing_open` / `disconnect` on both engines.
- tests: `test_active_layer_set_is_per_document` (headless registry, the
reviewer's repro plus the reverse direction) and the COM activate test now
also asserts the set is cleared and `_role_layer('dim')` is back on DIM.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd
aeab950 claimed one registry entry per file but only the two save paths
compared with normcase. _register_document keyed by the plain
os.path.abspath string, so on Windows drawing_open of an already-open
file spelled with another case registered a second entry (no `reloaded`,
open_documents: 2); from there every document_activate/close of either
entry was refused as ambiguous, and drawing_close(save=True) — which
bypasses the resolver — wrote the second entry's bytes to the shared
file while the surviving row reported saved: True. The resolver's
by_path route had the same plain-string comparison, so over the wire a
lowercase drive letter was refused 'no open document named ...' while
ComBackend._com_find_document lowercases and accepts it.
One helper, _file_identity(path) = normcase(abspath(path)), now decides
every registry comparison: _register_document consults _path_holder so
another spelling of an open file reloads it; the _doc_path setter re-keys
only when the file changes, so a case-variant save keeps the key a caller
holds; _resolve_document_key's by_path compares identities;
_close_document_sync refuses at the write when another entry holds the
file; _clear_quarantine's same-path warning uses the same rule. Keys stay
spelled as the caller spelled them — the identity is for comparing, not
for showing.
Five tests pin it (four skip on a case-sensitive filesystem, where two
spellings of one file cannot exist), including the wire route through
fastmcp.Client that the review measured.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd
drawing_save / drawing_save_as onto a file that another open entry holds
used to succeed: the registry then carried two entries for one path, the
resolver's exact-key shortcut picked the stale entry on document_close, and
the file ended up holding the other document's bytes while both rows still
reported saved: True. Both save paths now refuse before doc.saveas (as
AutoCAD refuses SAVEAS onto an open drawing), the _doc_path setter raises
rather than let two entries share a path, and _resolve_document_key refuses
a shared file on every route — exact key, path and basename — naming the
breach instead of advising "pass the full path" into the unchecked shortcut.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd
EnvironmentContract carries every member of the track E interface; the two
drawing_properties_* members are @capability('dwgprops') in this worktree only
and become @abstractmethod when group C's implementations merge.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd
