Release builds install each vendor as its preset cache alone. The
read-only preset load the CLI uses to resolve an inheriting user preset
passed allow_cache = false to keep caches from being written, which
also stopped them from being read, so every vendor fell back to JSONs
that are not installed and the CLI failed.
The flag now only gates writing: a read-only load reads caches and
writes none. The filament library is also read from its cache whenever
that is all that is installed, so a vendor updated over the air still
resolves against it.
* Refresh a CLI Project's Filament Settings From Their System Presets
The CLI loads a project's printer and process settings as the GUI does,
taking every key the project does not list as changed from the current
system preset, but it kept the stored filament values. A project saved
before a profile update then sliced with old filament values on the
command line and with the current ones in the GUI.
Every project filament that no loaded filament replaces is now resolved
by its system preset name and fed to the filament merge the up-to-date
path already uses, which keeps the keys listed in
different_settings_to_system and maps per-variant values onto the
preset's variants. This covers a plain run, --uptodate without
--uptodate-filaments, and the slots --load-filaments leaves empty. The
merge tells refreshed entries from loaded ones per entry instead of by
the global loaded-filament count, and the entries are kept in slot
order. A project filament saved under a name the presets have since
split per nozzle is resolved through the name conversion the GUI uses,
which PresetBundle now exposes.
* Check the Project Refresh Test's Result Directly
Shellcheck SC2181: test the checker's exit status in the if instead of
reading $? afterwards.
* Save device url in all cases and load printer url after hot-reload finishes
* Recreate web view from scratch as only URL fix seems not robust enough
* Add the same robust browser recreation for WebViewDialog
It should eliminate possible issue with blank Home and other pages
in the same way as Printer page
* Remove redundant fallback leftover
* Fix webview reset state and replay Project info on page reload
The first-show webview reset now runs only on Windows, reloads the last
printer URL and resets the Project page's ready state. The Project tab
replays its 3MF info whenever the page reloads, so it no longer goes
blank after a theme switch or a slow first load. The Device tab no longer
loads an extra time on first open, and the Home tab no longer navigates
twice. NeedsRecreateOnShow() logs is_recreating_gui so one language
switch shows whether the reset ever fires.
* Build plugin pages on first show
A language switch rebuilt every plugin page's browser while the main
window was being recreated, which left plugin tabs blank on Windows.
Plugin pages are now lazy pages, never prebuilt, and are removed left
to right so removing pages never builds one only to destroy it.
A plugin page's script now starts when its tab is first opened;
messages posted before that are dropped.
---------
Co-authored-by: SoftFever <softfeverever@gmail.com>
Co-authored-by: Noisyfox <timemanager.rick@gmail.com>
Co-authored-by: SoftFever <103989404+SoftFever@users.noreply.github.com>
* Add Missing Includes Across src/libslic3r
Every libslic3r source and header now directly includes the headers declaring what it uses, rather than relying on the precompiled header or transitive includes. Generated with clang-tidy misc-include-cleaner, with libslic3r headers spelled libslic3r/... so they resolve outside the library's private include paths. MultiMaterialSegmentation.hpp, Support/SupportParameters.hpp and Format/STEP.hpp are made self-contained by hand.
* Make the libslic3r Headers Compile on Their Own
Each now includes, or forward-declares, what it uses instead of relying on what its includers happened to include first. Left out: I18N.hpp, which errors on purpose when included from GUI code, and VoxelizeCSGMesh.hpp and SLA/bicubic.h, which nothing includes and which no longer compile at all.
* Add the Includes Missing From the Hand-Fixed libslic3r Headers
clang-tidy would not edit these headers while they failed to compile on their own, so the first pass skipped them. With the headers now self-contained, a second pass adds the rest.
* Keep Windows Setup Ahead of the Added libslic3r Includes
Print.cpp and Thread.cpp open with a _WIN32 block that has to come first; without the precompiled header, Print.cpp otherwise reaches windows.h through OCCT with NONLS defined and boost/regex fails. OpenVDBUtils.cpp and SLA/SupportTreeBuilder.cpp had includes inside #ifndef NOMINMAX, which libslic3r defines on Windows, so those were skipped there. .clang-tidy also ignores the MSVC STL and UCRT internals, Boost.Multiprecision's fwd.hpp and CPython's Windows include directory.
* Re-Add libslic3r Includes After the Clipper2 2.0.1 Migration
Rebasing onto main took main's version of the files the Clipper2 migration rewrote, so their added includes are restored here, along with includes for main's new code. Clipper2's individual headers are now ignored by clang-tidy: they only build the Z variant through clipper2_z.hpp, which defines USINGZ first, so including clipper.core.h and the like directly broke ClipperZUtils.cpp.
* Ignore Clipper, libpng, mcut and Boost.Polygon Internals in clang-tidy
Each only works through a wrapper or umbrella header: libslic3r/clipper.hpp or clipper_z.hpp configure Clipper before including it, png.h pulls in libpng's config headers, and Boost.Polygon's headers only compile through polygon.hpp or voronoi.hpp.
* Ignore minilzo's Config Headers in clang-tidy
lzoconf.h and lzodefs.h are internal to minilzo.h, which is what the code includes.
* Add Missing Includes Across the Remaining Sources and Tests
Covers src/slic3r/Utils, src/slic3r/plugin, src/slic3r/Config, src/libvgcode, src/dev-utils, src/OrcaSlicer.cpp and tests/, the directories left after src/slic3r/GUI and src/libslic3r. Generated with clang-tidy misc-include-cleaner. libvgcode's own headers are included by relative path as in the rest of that library, and Catch2 and pybind11 with angle brackets as elsewhere in the repo.
* Make the GUI and Test Headers Compile on Their Own
Each now includes, or forward-declares, what it uses instead of relying on what its includers happened to include first. Headers that only compile on one platform, or that nothing built includes, are left alone.
* Keep Windows and nanosvg Setup Ahead of the Added Includes
OrcaSlicer.cpp and several tests set _WIN32_WINNT, WIN32_LEAN_AND_MEAN or NOMINMAX before including Windows.h, and the profile validator defines NANOSVG_IMPLEMENTATION before any libslic3r header. The added includes had landed above those blocks, which broke the Windows build.
* Add the GUI Includes the First Pass Missed
Covers headers that only became editable once they compiled on their own, and wx symbols whose suggested header changed as the clang-tidy ignore list grew after the src/slic3r/GUI pass.
* Keep the Added Test Includes Below the NOMINMAX Guard
test_marchingsquares.cpp and test_texture_displacement.cpp had includes inside #ifndef NOMINMAX, which the tests inherit as defined on Windows from libslic3r, so those were skipped there. .clang-tidy also ignores the MSVC STL and UCRT internals, Boost.Multiprecision's fwd.hpp and CPython's Windows include directory, as in #16068.
A tool change added from the layer slider switches heads mid-print like a
painted color, but the parallel-mode checks only looked at the filaments of the
plate's objects, support and prime tower. A one-filament plate with a slider
change to a copying head's filament passed them. They now include those tool
changes, as the plate's warning badge already did.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Only the primary and its Span tools print a Span plate's colors; the other
active tools replay them. The multicolor rule never checked where each color
went, so a filament on a copying or mirroring tool, an unused one, or past the
end of the physical extruder map was accepted. It is now refused, and the
message names the tools that print colors and where the map sends the
offending filament.
The plate's warning badge also read the raw physical_extruder_map rather than
the effective one slicing uses, so on printers that set no map it disagreed
with the slicer.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three commits: G-code export caches its filament config slot and repeated
option lookups, per-plate bed type overrides follow the printer's multi-bed
support, and the gizmo checkboxes and texture displacement panel get styling
and refresh fixes. No conflicts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The name field rebuilt its suggestions in wxEVT_COMBOBOX_DROPDOWN, which
ComboBox sends after it has sized and shown the popup from the items it
already held. With every suggested name taken, the popup opened around an
empty list: a small empty box on GTK, a black one on Windows. The list is now
rebuilt from the field's own mouse-down and double-click, which run ahead of
ComboBox's handlers. With nothing to offer the popup stays closed and the
click focuses the field instead.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The editor only replaced an empty name, so a row could be given a name another
row already had. A plate stores its mode by name and find_imex_mode() takes the
first row with it, so the second row was unreachable, and the plate's mode
list repeated the name: left-click stuck on it, or looped without getting back
to Primary.
An edited name that another row already carries, or the reserved Primary name
in any case, is now replaced when the edit is committed: "copy" becomes
"copy 2". The edited row yields, so plates keep resolving to the row they
meant, and tabbing through a field without changing it checks nothing.
Resetting a row to a saved name that another row has since taken does the
same. The plate's mode list comes from imex_plate_mode_choices(), which lists
each name once, so a profile that already has duplicates still cycles.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The editor took its background from the app's window default, which in dark
mode is a gray the dark-mode walk has no entry for. In dark mode it never
matched the page's palette color, and an editor built in dark mode kept that
gray after a switch to light while the controls inside it changed. It now
takes the page's color, which the walk remaps both ways.
Its labels and G-code text had no color of their own, so on Windows they took
the system's text color, which follows Windows' theme rather than Orca's. They
now start from the page's label and input text colors, which the walk maps
with the page.
The walk also ran the tool tiles' role colors through the palette map, which
turned their labels gray after a theme change. The tiles carry wxBU_AUTODRAW,
the flag it skips, so the role colors stay as set.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The remove button drew imex_remove.svg, a redrawn delete.svg, and the help
button drew the mascot question icon. They now use delete and icon_qusetion,
the tip icon the send-print dialog uses, and imex_remove.svg is gone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Fix gizmo checkbox styling and Texture Displacement resize
Use shared BBL checkboxes in Texture Displacement and restore white toolbar checkmarks so other gizmos keep proper checkbox contrast. Also fix Texture Displacement resizing only after mouse movement by requesting additional frames while its layout is still changing.
* Fix gizmo checkbox styling and Texture Displacement resize
Use shared BBL checkboxes in Texture Displacement and restore white toolbar checkmarks so other gizmos keep proper checkbox contrast. Also fix Texture Displacement resizing only after mouse movement by requesting additional frames while its layout is still changing.
* Update GLGizmoTextureDisplacement.cpp
---------
Co-authored-by: yw4z <ywsyildiz@gmail.com>
The update merged the same upstream commits this branch already carries. Its
GLCanvas3D resolution kept the assembly view painter block after the ghost
render, which section view removed along with the members it calls, so that
tree does not build. This merge keeps the removal; nothing else differs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Five commits: a section view for the 3D canvas, a faster G-code preview, user
preset values kept on extruder variants they don't list, debug build CMake
fixes, and a 30 minute timeout on macOS notarization.
Two conflicts, both in GLCanvas3D. The header's were neighboring
declarations, kept from both sides. In the transparent pass, the section view
removed the assemble view's painter block that followed our ghost render;
the ghost render stays where it was, at the end of that pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
notarytool submit --wait has no timeout. On 2026-10-02 it hung for
over 5 hours in a main build. Since #16044 a new push no longer
cancels a running main build, so nothing stopped it and six waiting
main runs were replaced without starting.
Over the last 30 days the step succeeded 206 times, with a median of
4.3 minutes and a maximum of 14.8.
The plate's mode button showed one icon whatever the mode. It now shows the
mode the plate slices as, the one its ghosts follow. imex_mode_kind() reads it
from the heads beside the mode's primary: none is Normal, any Span head is
Custom (the multicolor modes), otherwise any Mirror head is Mirror, and heads
that all copy are Copy. One Mirror head is enough because an IQEX mirror mode
copies within the primary's gantry.
The icons are Felix14-v2's: the four kinds, each in light and dark with a hover
state, replacing the single mode icon. Two fixes to them: the light Normal
border used the dark theme's gray, and an opacity="undefined" attribute hid one
of its strokes in nanosvg. The knight outlines are drawn at 1.0 rather than
0.8, matching the other plate icons and keeping Copy and Mirror apart when
zoomed out.
Co-authored-by: Felix14_v2 <75726196+Felix14-v2@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Fix debug build after qhull upgrade
We upgraded qhull from 8.0.1 to 8.0.2 in 504a5d3b70, which contains a commit qhull/qhull@16159c648c `use same CMake target name for Debug and non-Debug`, so this target name check is no longer required
* Fix issue like `IMPORTED_LOCATION not set for imported target "opencv_world" configuration "RelWithDebInfo".` when build Debug config
Seven commits: painted multi-material segmentation made deterministic, missing
includes added across src/slic3r/GUI, U1 high-flow nozzle variants, Windows
ARM64 build and HTTPS fixes, and a rule added to the wxWidgets agent skill.
Three conflicts, all include lists: upstream's include pass and ours each added
to the same lists in GCodeViewer.hpp, PartPlate.hpp and PartPlate.cpp. Resolved
as the union of both. Three includes both sides had added at different places,
which git kept twice, are kept once: Color.hpp and <set> in PartPlate.cpp,
<sstream> in Plater.cpp.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The IMEX fixtures set seven nozzles but kept multifilament_config's single
filaments x filaments flush block. get_flush_volumes_matrix splits that block
across the nozzles, leaving each with 7 values, and
ToolOrdering::reorder_extruders_for_minimum_flush_volume then reads them as a
7 x 7 matrix, past the end of the buffer. One of the affected tests segfaulted
on Windows x64; ASan reproduces the overflow in that test on Linux, where it
passed only by luck.
The fixtures now repeat the block once per nozzle, as the GUI does, and size
flush_multiplier to match, since append_full_config takes the nozzle count from
it. The helper is shared in test_helpers.
Co-authored-by: HanifKoh <76276251+HanifKoh@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Keep User Preset Values on Extruder Variants They Don't List
A user preset stores the variant list its parent had when it was saved.
When the parent later gains variants, update_diff_values_to_child_config
matched variants by name only and left the new ones at the parent's
value, so the user's settings were silently replaced there, and a
re-save wrote the system values into the user's file.
An unmatched parent variant now takes the child's first variant of the
same extruder, the rule slicing already uses in get_config_index_base.
A child without a variant list covers the parent's first extruder. The
name match also no longer indexes the child's extruder ids when it has
none.
* Share One Variant Column Rule Between Slicing, User Presets and Projects
Three places chose which variant column a value comes from, each with
its own copy of "the same variant and owner, else the owner's first
column": get_config_index_base when slicing, the user preset merge in
update_diff_values_to_child_config, and normalize_filament_values_to_variants
for projects and the CLI.
find_variant_column now holds that rule and map_variant_columns applies
it to a variant list, so a change to how missing variants are filled
reaches all three. Each caller keeps its own copy step. There is no
behaviour change: G-code is identical before and after. The one
relaxation is that get_config_index_base no longer reads past a short
id list when its two lists differ in length, which its assert already
rules out.
* Rename variant column helpers to variant index
---------
Co-authored-by: SoftFever <softfeverever@gmail.com>
* fix: include bundled UV binary for arm64
* fix: update unit test CI
* Install unit-test numpy only with the bundled uv
* Simplify the unit-test script's uv lookup
---------
Co-authored-by: SoftFever <103989404+SoftFever@users.noreply.github.com>
Co-authored-by: SoftFever <softfeverever@gmail.com>
The ARM64 jobs pinned CMake 3.31 because CMake 4 dropped pre-3.5 policy
compatibility and its ARMASM support broke Boost.Context. Both are
handled now. deps/CMakeLists.txt sets CMAKE_POLICY_VERSION_MINIMUM on
CMake 4, and Boost.Context uses the winfib implementation on ARM64, so
nothing assembles with armasm.
CMake 3.31 also predates VS 2026. Its InstallRequiredSystemLibraries
treats the v145 toolset as v143, searches only the VS 2017-2022 install
directories and finds no runtime, so the ARM64 installer ships without
msvcp140.dll and vcruntime140.dll. CMake 4.2 and newer find the VC145
redistributable.
get-cmake also installs Ninja, so the ARM64 jobs now use its latest
release instead of the one already on the runner, as x64 does.
The install now fails when InstallRequiredSystemLibraries returns no
msvcp140.dll or vcruntime140.dll, after a configure warning naming the
CMake and MSVC versions. A CMake that predates the Visual Studio in use,
on a developer machine or after the next runner image update, then
stops the installer build instead of shipping one that cannot start.
The build_win.bat prerequisite installer drops its matching 3.31.8 pin.
This is the bulk of the profile changes for U1 that led to the cooling
catiant connection.
The ugly end gcode is used by the printer's UI to complain if normal
nozzle is installed but the file was sliced for highflow.
[How to Download Pull Requests Artifacts for
Testing](https://www.orcaslicer.com/wiki/how_to_download_pr_artifacts)
* Make Painted Multi-Material Slicing Deterministic
Painted (multi-material) models sliced to slightly different G-code on
every run: ±1 µm wall coordinates and reordered islands. Hashing each stage
of the segmentation across runs showed the projected painted lines and the
per-layer Voronoi segmentation were stable; the raw top/bottom projections
from slice_mesh_slabs() were not. Three causes, all thread-order dependent:
- slice_slabs_make_lines() appends each slab's intersection lines from a
parallel facet loop and never restored a canonical order, so the loop
start vertices and polygon order from make_slab_loops() depended on
scheduling. Sort every slab's lines with the same key slice_make_lines()
already uses.
- segmentation_top_and_bottom_layers() wrote a layer's shell projections
into neighbouring layers' vectors from the parallel loop, relying on a
parity double-buffer that assumes TBB ranges are exactly one group wide
and aligned, which blocked_range does not guarantee; two threads could
append to the same vector. Each source layer now records its projections
in its own slot and they are gathered per target layer in source order.
- The painted-line sort in post_process_painted_lines() was not a total
order: projections of one span from facets of different colours tied on
every key and the first one won the span. Colour and end points now break
the tie.
Three multi-threaded runs of each painted fixture now give one G-code;
unpainted output is unchanged.
* Test That Slab Slicing Does Not Depend on the Thread Schedule
Projects a dense, tilted sphere with slice_mesh_slabs() on one thread and
then three times multi-threaded, and requires the polygons to match exactly,
vertex order included. Fails without the canonical line sort, passes with it.
* Add Missing Includes Across src/slic3r/GUI
Every GUI source and header now directly includes the headers declaring what it uses, rather than relying on the precompiled header or transitive includes. Generated with clang-tidy misc-include-cleaner, plus one hand edit making CalibrationPanel.hpp self-contained.
* Drop the OS-Specific Includes Added Outside Their Platform Guards
GLib, GTK, D-Bus and POSIX headers are only used inside platform #if blocks, which already include them. Added unconditionally at the top of the file they broke the Windows build.
* Add the clang-tidy Configuration That Generated These Includes
Only misc-include-cleaner's missing-include check, with the headers it must never suggest: per-platform, internal and OS-specific ones that would break other platforms or are not meant to be included directly.
* Match Windows Paths in the clang-tidy Ignore List
Header paths use backslashes on Windows, so every / in a pattern is now [/\\]. The Windows SDK headers are ignored alongside the other OS-specific ones, and the list is one pattern per line. Suggested by @raistlin7447 from a Windows clang-cl run.
filament_minimal_purge_on_wipe_tower became variant-keyed upstream, which
invalidated the comment claiming none of full_config's keys were.
WipeTower2::extract_wipe_volumes indexes it by raw filament slot. Not reachable
with shipped presets; noted rather than worked around.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
ComboBox::SetLabel is overridden and writes the text control when the text
control is shown, which it is on an editable combo. Clearing the label after a
pick therefore erased the name. Removed; the label is never written on this
control.
Suggestions are also rebuilt when the list opens. Built in add_row they filtered
against only the rows that already existed, so a row was offered names the rows
below it had taken.
Drops the 1px inset on the Primary label, which matched a frame the name field
no longer has.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
138 commits. The count is large because texture displacement merged with its
whole branch history behind it, going back to July, alongside config and preset
file locking across instances, a foundation for configurable printer agent
connections, and a day of smaller fixes and CI work.
Two conflicts, both the same shape: each side had appended to a sorted list and
git could not choose an order. libslic3r's CMakeLists gained InstanceLock
alongside our IMEXHelpers and IMEXZones, and the preset bundle loading test
gained an include for ParallelResolve alongside ours for IMEXHelpers. Both sides
kept, alphabetical. No logic conflicted.
Verified: 740 targets build clean under -Werror, and the Release suite passes
1740 of 1740, up from 1665 -- the 75 new cases arrived with the merge and all
pass. That mattered more than usual here, since preset loading and config
locking are both areas the IMEX preset code touches.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The name field becomes an editable ComboBox, the same pattern the sidebar uses
for parameters like sparse infill anchor length: predefined entries in a
drop-down, with the text still typeable. The mode table is authored for whatever
hardware the user has, so a closed list would be wrong -- nothing in the slicer
reads a mode's name except as the key a plate stores -- but the conventional
names are worth offering rather than leaving everyone to retype them.
What is offered follows the tool grid rather than a fixed list. Four carriages
get iq-copy and iq-mirror; multicolor needs a Span partner beside the primary
and a second gantry to copy the pair onto, so mc-copy and mc-mirror appear only
on a grid that can hold one. imex_resolve_routing() already refuses a multicolor
mode with no Span on the primary's gantry, and suggesting a name it would then
reject is worse than not suggesting it. Names a row already uses are dropped, so
the list only ever offers what is still free.
Two things about ComboBox matter when it is editable, which nothing else in the
tree does -- the other 78 call sites all pass wxCB_READONLY:
GetValue() returns the drop-down selection whenever there is one, so a name
typed after picking a suggestion would be silently discarded. Every read goes
through GetTextCtrl() instead. Field.cpp reconciles the same way for its own
open enums.
The constructor hands its value to TextInput as the LABEL -- the small
right-aligned slot a unit like "mm" occupies -- because a read-only combo hides
the text control and shows the label in its place. Left there, the name rendered
as a greyed echo beside the hint while the field itself sat empty. The combo is
built empty and the value written to the text control, and the selection handler
clears the label again afterwards, since SetSelection() writes there too.
The mode column widens to 176 to leave room for the drop-down arrow, with the
header spacer deriving from the same constant.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* Lock Config and Preset Files Across Instances and Write Them Atomically
Every running instance shares one OrcaSlicer.conf and one user preset
tree, and nothing kept their writers apart. Two instances saving at the
same moment, or the cloud preset sync thread writing while the GUI thread
saved, could interleave, and a reader in another instance could open a
preset JSON or .info file between truncate and close and get a partial
file, dropping that preset for the session with a parse error.
Add InstanceLock, a scoped guard that serialises the threads of one
process through a recursive mutex and other processes through an advisory
OS file lock: flock on POSIX, held on the guard's own descriptor so no
other close in the process can drop it, and LockFileEx on Windows. The
outermost guard opens the lock file and closes it on release, so nothing
stays open between saves and a data dir can be removed once nothing is
saving into it; the file itself is kept, since deleting it would let a
third instance lock a fresh file while the second still holds the old
one. It is best effort: when the lock file cannot be opened or locked, or
another instance still holds it after a second, the guard logs once and
lets the write proceed, then leaves the file alone for ten seconds, so a
hung instance never blocks every other one and a holder stuck in a
debugger does not cost a stall per save. The guard sits at the leaf
readers and writers: set_sync_info_and_save() calls save_info() under the
preset collection mutex, so a batch lock around save_user_presets() would
invert the order against the sync thread. The user preset scan reads its
files on worker threads without the guard, since the mutex would
serialise them, and takes it per file in the serial commit step, so a
save never waits for the whole scan. Each read keeps the bytes of the
preset and its .info as they were before parsing; commit compares them
with the disk under the guard and reads a file that changed again, so it
never deletes or writes back over another instance's newer save, nor
installs a .json and .info from two different saves; a preset another
instance removed in the meantime is not installed. Without the guard, in
a cool-down, the scan still loads the presets but leaves their files
alone: an unreadable file stays for the next scan, and a derived
compatible printer is not written back. Read-only scans, which is what
the CLI does, take no lock and create no lock file.
AppConfig holds OrcaSlicer.conf.lock in load() and save(); load is
included because the Windows path restores from the .bak copy. Every
user preset writer and reader holds user.lock: Preset::save(), which
writes no .info when the preset itself could not be written, since an
.info without its preset reads as a cloud deletion request, save_info(),
reload() and remove_files(), each preset the scan commits, the
bundle metadata reads and write, the .info removal after a cloud-confirmed
delete, the orphaned-.info scan on the sync thread, the bundle folder
removal on unsubscribe and the physical printer writers and delete
paths. A bundle import extracts under cache/ into a folder per process
and per import, where no scan reads.
Preset JSON, .info, bundle metadata, physical printer and config files,
and the caches and state files that already used a temporary by hand,
now go through write_file_atomically(), which writes <file>.<pid>.<n>.tmp
beside the target and renames it over, so a reader that never waits sees
a complete old or new file. A symlink is followed; a target that is not
a regular file is written in place; and when no temporary can be created
beside an existing target, or the rename itself is refused, by a Windows
reader holding the file open or a mount that cannot replace in one step,
the helper writes in place as before, since losing the save is worse
than a torn read. On POSIX the rename replaces the
target atomically where the old code removed it first and left a window
with no file at all; only a mount that refuses a one-step replace gets
the old remove-then-rename. A crash between temporary and rename leaves
the temporary behind, which no scan reads. Preset::save() returns
whether it wrote the preset, so the scan counts a compatible printer it
could not write back as an error. A failed config write keeps
the config dirty, and the idle handler waits ten seconds before retrying
while an explicit save always tries.
* Run the Cross-Process Lock Test on Every Platform
The test that checks the guard yields to a lock held elsewhere forked a
child to hold it, so it was left out on Windows. The OS lock belongs to
the handle on Windows and to the open file description elsewhere, so a
second handle in the same process is refused like another instance
would be. The test now holds the lock that way and runs everywhere.
Rodrigo Faselli