# Description
<!--
> Please provide a summary of the changes made in this PR. Include
details such as:
> * What issue does this PR address or fix?
> * What new features or enhancements does this PR introduce?
> * Are there any breaking changes or dependencies that need to be
considered?
-->
Adds a really basic API to push notifications to the plater.
Plugin used in demo:
[plater_notification.py](https://github.com/user-attachments/files/31293074/plater_notification.py)
# Screenshots/Recordings/Graphs
<!--
> Please attach relevant screenshots to showcase the UI changes.
> Please attach images that can help explain the changes.
-->
<img width="2172" height="1241" alt="image"
src="https://github.com/user-attachments/assets/540319ca-a11a-4b48-9b80-82fb6b0849d9"
/>
## Tests
<!--
> Please describe the tests that you have conducted to verify the
changes made in this PR.
-->
<!--
> A guide for users on how to download the artifacts from this PR.
-->
[How to Download Pull Requests Artifacts for
Testing](https://www.orcaslicer.com/wiki/how_to_download_pr_artifacts)
* CLI: record command-line overrides in different_settings_to_system
Settings passed on the command line (--sparse-infill-density 25% ...) override
the loaded presets when m_extra_config is applied to m_print_config, but nothing
recorded them in different_settings_to_system. The exported project therefore
carried the new value with no mark that it was modified, and re-opening it in the
GUI reverted it to the system preset's value -- the same failure the preset-leaf
diff fixes for user presets, via a different source of override.
The key set comes from m_config, not m_extra_config. read_cli() puts only what the
user typed into m_config and setup() adds nothing but CLI-own defaults (none of
the keys run() materialises there is a preset option), whereas the CLI writes its
own values into m_extra_config (has_filament_switcher, filament_colour,
filament_map ...), which must not be reported as user overrides.
Values are snapshotted just before the apply and only keys the override actually
changed are recorded: a typed value equal to the loaded one modifies nothing, and
listing it would read as a spurious difference against what the GUI writes. Each
key lands in the column(s) whose preset type owns it -- process, every filament,
printer -- and a key already present is not duplicated. Keys no preset owns
(curr_bed_type, a project setting) land nowhere, as in the GUI.
Follow-up to #15595, split out at review.
* CLI: judge command-line overrides the way the value is read
Review follow-ups on the override recording:
- Lists were compared as whole serialized strings. read_cli() builds a fresh
one-entry list, so --nozzle-temperature 245 against 245,245,245 on a
three-filament project was recorded in every filament column although nothing
changed. Lists are now compared entry by entry with a missing entry read as the
first, as get_at() reads it (and as resize() pads).
- The log line fired for every changed key, including ones no preset owns
(curr_bed_type) and which therefore land in no column. It now fires only when a
column took the key.
- m_print_config.has(key) straight after apply(m_extra_config, true) was always
true, both configs sharing print_config_def; removed. columns.size() >= 2 also
always holds after the resize to filament_count + 2 -- different_settings_to_system
is not a CLI option, so nothing in between can shrink it -- but that rests on code
far away, so it stays a plain check rather than an assert: release builds compile
asserts out, and a _GLIBCXX_ASSERTIONS build would abort on columns[0].
Deliberately NOT done: comparing a key the loaded config lacks against its
built-in default. On reopen the GUI restores an unlisted key from the SYSTEM
preset, not the default. A 3MF written before an option existed leaves it absent
here, so --sparse-infill-density 20% (the default) against a Prusa system 15% would
go unrecorded and be reverted to 15%. Absent keys stay always-recorded:
over-recording is cosmetic, under-recording loses the value. Verified that such a
key really is absent at this point, rather than filled from the system preset.
Reported by HanifKoh and raistlin7447 in review of #15642.
* CLI: evaluate compatible_printers_condition in the compat checks
Slicing from the CLI with --load-settings exits with
CLI_PROCESS_NOT_COMPATIBLE (-17), "The selected printer is not compatible
with the process preset in the 3mf.", for process/printer pairs the GUI
accepts. Reproducible with stock, unmodified Prusa system profiles:
orca-slicer --datadir <datadir> \
--load-settings "<datadir>/system/Prusa/process/0.20mm SPEED @CORE One HF 0.4.json;<datadir>/system/Prusa/machine/Prusa CORE One HF 0.4 nozzle.json" \
--load-filaments "<datadir>/system/Prusa/filament/Prusament PETG @CORE One HF 0.4.json" \
--slice 0 --outputdir /tmp/out model.stl
The four compat checks in CLI::run did a literal name match against the
`compatible_printers` list only:
for (index ...) if (new_print_compatible_printers[index] == new_printer_system_name)
process_compatible = true;
Process profiles that declare compatibility through
`compatible_printers_condition` and leave `compatible_printers` empty are
therefore always reported incompatible -- the condition is never consulted.
For 0.20mm SPEED @CORE One HF 0.4 that condition is:
printer_notes=~/.*PRINTER_MODEL_COREONE[^_a-zA-Z0-9].*/ and
nozzle_diameter[0]==0.4 and printer_notes=~/.*HF_NOZZLE.*/
The GUI does not have this bug: is_compatible_with_printer() in Preset.cpp
treats an empty list as "no explicit constraint" and evaluates the
condition in that case.
Fix: replace the four loops with a check_compat lambda that calls
is_compatible_with_printer() -- the same helper the GUI uses -- wrapping
the already-loaded DynamicPrintConfigs in lightweight Preset /
PresetWithVendorProfile shells. The 3MF-embedded process/printer full
configs are kept in current_process_full_config /
current_printer_full_config so the condition can be evaluated for the
reprocess paths too; those fall back to the previous literal match when
the full config was not preserved.
Behaviour is unchanged where an explicit compatible_printers list exists:
is_compatible_with_printer() performs the same name match, and returns
true when both list and condition are empty, matching the existing
"old 3mf, no compatible printers, set to compatible" path.
Split out of #13731 (section 1) as a standalone, single-purpose change.
Orthogonal to the inherits-chain resolution work in #14718 / #15302 /
#15438; those decide which values a preset resolves to, this decides
whether the resulting pair is considered compatible.
* CLI: translate the 3MF's renamed compatibility keys before the compat check
The 3MF fallback fed the project config to is_compatible_with_printer() as-is,
but a project config does not carry compatible_printers or
compatible_printers_condition. PresetBundle::construct_full_config() erases both
and re-emits them as print_compatible_printers and
compatible_machine_expression_group; they are renamed back only on the
PresetBundle load path, which the CLI does not take. The check therefore saw no
list and no condition, read that as 'no constraint' and accepted every printer.
That is not just a wrong accept. An early true skips the !process_compatible
block that sets machine_switch, so the new printer is never appended to
print_compatible_printers and the exported 3MF stays marked compatible only with
the printer it came from -- which is exactly what that block exists to prevent.
Translate the two keys back before the check. Index 0 of the expression group is
the print preset; the group is filled print, filaments, printer.
Also note in the comment that profiles/BBL/{process,machine}_full/ are gitignored
and generated by nothing in-tree, so current_*_full_config is always empty and
this fallback is the only live path -- not the rare non-BBL case the original
comment implied.
Reported with measurements by HanifKoh in review of #15449.
Preset: add a config-level is_compatible_with_printer() overload
The CLI holds resolved DynamicPrintConfigs, not Presets, so it wrapped them in
throwaway Preset shells at the call site. Moving that into Preset.cpp puts the
compatibility policy -- including the documented fail-open on a malformed
compatible_printers_condition -- in one place for the GUI and the CLI, rather
than leaving a second copy of the plumbing in OrcaSlicer.cpp to drift.
Purely additive: neither existing overload changes, so no GUI behaviour moves.
Requested by HanifKoh in review of #15449.
(cherry picked from commit 14ca1972ef4d3c7d90935d159423013a40a6bd70)
* CLI: never overwrite a real compat key with an empty renamed one
7e7f0e3 translated compatible_machine_expression_group[0] into
compatible_printers_condition whenever the group vector was non-empty. A project
the CLI exported itself carries the real compatible_printers_condition AND an
all-empty group, ["", "", ""], so the valid condition was overwritten with
"", the check saw no constraint, and every printer was accepted.
That fixed GUI-shaped projects and broke CLI-shaped ones. Bisected across six
builds re-slicing one CLI-exported CORE One project with an MK4S: every build
before 7e7f0e3 gives 'compatible 0' and takes the machine-switch path; with it,
'compatible 1' and no switch.
The raw keys now win whenever they carry something; the renamed ones are only a
fallback, and an empty value is never written over a real one. Same for the list:
print_compatible_printers is used only when compatible_printers is absent or
empty and it itself is not.
Found by a peer session re-testing the installed build.
* CLI: record user overrides in different_settings_to_system for 3MF export
Three sites in CLI::run wrote an empty `different_settings_to_system` column
and left a //todo:
//todo: support user machine preset's different settings
different_settings[filament_count+1] = "";
//todo: support system process preset
different_settings[0] = "";
//todo: update different settings of filaments
different_settings[filament_index] = "";
So a 3MF exported by the CLI does not record which keys the user actually
overrode relative to the system parent. Re-opening such a project in the GUI
then shows spurious "unsaved changes", and accepting that dialog can revert
inherited process/filament/machine values to system defaults.
The column could not be filled before because the CLI had no resolved view of
the parent preset. It does now: #15438 builds a PresetBundle for inherits
resolution, so the parent can be looked up by name and diffed against the
resolved leaf. This adds no extra loading -- the bundle is the one already
built, and the helper returns "" whenever it is unavailable or the parent
cannot be found, which is the previous behaviour.
Preset metadata is filtered out of the diff: `inherits`, the three
`*_settings_id` keys, and `compatible_printers` / `compatible_prints` and
their `_condition` variants, which have their own tracking columns
(`inherits_group`, per-slot lists) and would otherwise double-count.
A value already carried by the loaded JSON still wins for the process slot, so
presets saved with a `different_settings_to_system` field behave as before;
the computed value only fills the gap where that field is absent, which is the
case for every user preset in my datadir (0 of 47 carry it).
System presets keep an empty column: there are no user overrides to record.
* CLI: diff the filament slot before load_default_gcodes_to_config
The process and machine slots compute their different_settings_to_system column
before load_default_gcodes_to_config(); the filament slot did it after. That
call materialises absent gcode keys via option(..., true), and
DynamicConfig::diff only compares keys present in both configs -- so a gcode key
the resolved leaf did not carry would go from 'not compared' to 'compared as
empty against the parent' and land in the column as an override the user never
made.
Hoisted into a local above the call, guarded by load_filament_count > 0 so the
work is skipped exactly where it was before, and assigned at the original site.
The diff now also runs before config.erase("filament_settings_id"), which is
immaterial: cli_different_settings already filters filament_settings_id along
with the other *_settings_id keys.
This is a consistency fix rather than a demonstrated defect -- resolve_preset
merges the parent config, so in practice the gcode keys are already present on
both sides and the diff is unaffected. It removes the dependence on that
invariant, which the other two slots never had.
Reported by HanifKoh in review of #15595.
# Description
<!--
> Please provide a summary of the changes made in this PR. Include
details such as:
> * What issue does this PR address or fix?
> * What new features or enhancements does this PR introduce?
> * Are there any breaking changes or dependencies that need to be
considered?
-->
Every CI build leg compiles the whole tree from scratch: 42 to 57
minutes of each build job, on every push and every pull request, roughly
200 runs a week. This PR caches the compiled objects with ccache so that
a run only compiles what changed since the last push to main. With a
warm cache the compile steps take 1 to 4 minutes on all six legs and a
pull-request run finishes in about 30 minutes instead of 75.
Three prerequisites landed last week and made this measurable: #15537
took `GIT_COMMIT_HASH` off the compile line, #15552 made a build without
the precompiled header work on Windows, and #15501 stopped the Flatpak
job from rebuilding its dependencies.
## Changes
### Compiler cache in `build_orca.yml`
Each build leg (Linux x86_64/aarch64, Windows x64/arm64, macOS
arm64/x86_64) restores a cache entry keyed by that leg, compiles through
`ccache` via `CMAKE_<LANG>_COMPILER_LAUNCHER`, and prints its hit
statistics at the end of the job. The macOS universal combine does not
compile and is left out.
Who writes the cache is the important part. Cache entries are immutable
and a restore always takes the newest matching one, so every save is a
new entry that is never read again once a newer one exists. Therefore:
- **Pushes save.** After a successful save, the older entries for the
same leg on the same ref are deleted, so a branch holds exactly one
entry per leg. The save comes first, so a failed save leaves the
previous entry in place.
- **Pull requests restore only.** They read main's entries (GitHub lets
a PR read the base branch's caches) and keep nothing. Saving from PRs
would add about 6 GB per run that no other run can read.
The store is therefore a flat ~7 GB (one entry per leg: Linux ~1 GB,
Windows ~2 GB, macOS ~0.6 GB), not a growing one. The
`hendrikmuhs/ccache-action` only installs and configures ccache; restore
and save go through `actions/cache` with one path string, because the
cache service only matches entries saved under the identical path and
the action spells it differently on Windows. A failed ccache install
falls back to an uncached build rather than failing the job.
### Precompiled header off when the cache is on
With `SLIC3R_PCH` left on, a warm cache hit only 19 % of compiles: Clang
stamps the PCH with the build time, CMake does not pass
`-fno-pch-timestamp`, and everything that includes the PCH (libslic3r
and libslic3r_gui, ~750 files) missed every run. `build_linux.sh -p`
exists for exactly this reason. The workflow now exports
`ORCA_EXTRA_BUILD_ARGS=-DSLIC3R_PCH=OFF` whenever ccache is enabled,
which brings the warm hit rate to 98.4–98.9 %.
The cost is on cold compiles, which are 25–60 % slower than today's PCH
build (ccache preprocesses every miss before compiling it, and the miss
compiles without PCH). Main pays this once after an image update or a
wide header change; PRs pay it only for the files their change
invalidates. A change to a header included by half the tree
(`PrintConfig.hpp`, `Preset.hpp`, `Model.hpp`) lands a run at 1.2–1.9×
today's time. `ccache`'s depend mode would remove the preprocessor pass
and is the natural follow-up.
### Includes the precompiled header was supplying on macOS
A build without PCH had never been tried on macOS. Three files used what
`pchheader.hpp` happened to include: `LocalesUtils.cpp` needs
`<sstream>` and `<iomanip>`, and `AmsMappingPopup.cpp` /
`PhysicalPrinterDialog.cpp` need `<wx/tooltip.h>`. libstdc++ and the GTK
wx port pull these in transitively; libc++ and the Cocoa port do not.
This is the macOS counterpart of #15552 and is worth merging on its own.
### `ORCA_EXTRA_BUILD_ARGS` pass-through
`build_linux.sh` already forwarded this variable to the slicer
configure. `build_release_macos.sh` now reads it into an array
(shellcheck-clean), and `build_release_vs.bat` appends it on both
configure lines, so CI can add a CMake option without editing three
scripts.
## Behaviour reviewers should know about
- **Main-only cache writes need `actions: write`** on the workflow token
to delete the previous entry. The default token already has it (the
nightly deploy steps write with it), so no `permissions:` block was
added. A fork PR's read-only token never reaches the delete step.
- **A runner image update cold-starts the cache** as configured, because
ccache keys the compiler by its mtime and every image rebuild reinstalls
it. Images updated 20260819 → 20260828 during this work, about every one
to two weeks. Keying on the compiler version string (`compiler_check`)
would avoid that; left as a follow-up since it changes every hash.
- **What is now the critical path:** the two Flatpak jobs (46–66 min,
untouched here), the orca-test-repo regression suite run inline in the
Linux job (7 min), and NSIS/PDB/MSIX packaging on Windows (6 min). Those
are the next wins.
- **Open question:** CI still drives `build_release_vs.bat`. #15552 gave
`build_win.bat` a `--cache ccache --no-pch` option; moving the Windows
job onto it would replace the batch-file change here.
# Screenshots/Recordings/Graphs
<!--
> Please attach relevant screenshots to showcase the UI changes.
> Please attach images that can help explain the changes.
-->
Compile step of each build leg, minutes. Main's numbers are from run
34324625046.
| Leg | main | cold, PCH on | warm, PCH on | cold, PCH off | warm, PCH
off | ~50 % of headers changed | 5 source files changed |
|---|---|---|---|---|---|---|---|
| Linux x86_64 | 48 | ~75 | (19 % hits) | ~110 | **2.6** | 91.3 (452/928
misses) | 3.4 |
| Linux aarch64 | 41.7 | 53.4 | 52.2 (178/927 hits) | 58.8 | **2.5** |
55.8 (452/928) | 2.9 |
| Windows x64 | 57 | 85.5 | — | ~105 | **2.4** | 76.8 (449/974) | 2.5 |
| Windows arm64 | ~45 | 64.4 | — | ~78 | **4.3** | 59.6 (450/974) | 4.2
|
| macOS arm64 | 51 | ~71 | — | — | **0.8** | 79.7 (453/947) | 0.9 |
| macOS x86_64 | ~43 | 70.2 | — | — | **1.0** | 68.1 (411/742) | 1.0 |
Warm hit rates: 98.4–98.9 % on every leg; the 11–14 misses are what any
commit changes (version stamp and its includers). The "50 % of headers"
column is a real event: #15251 and #15416 merged into main between two
runs, changing 20 headers that reach 453 of 870 translation units.
Whole run, before and after (a pull-request run; wall clock to the last
non-Flatpak job):
| Job | main (run 34324625046) | warm cache (run 34444305385) | what
remains |
|---|---|---|---|
| Windows arm64 | 50.0 | 15.7 | compile 4.3, NSIS 3.5, cache save 1.6,
deps restore 1.1, cache restore 1.0 |
| Windows x64 | 67.4 | 13.2 | NSIS 3.2, PDB 2.6, compile 2.4, MSIX 0.5 |
| Linux x86_64 | 57.5 | 12.6 | orca-test-repo regression 7.6, compile
2.6 |
| macOS x86_64 | 46.9 | 6.1 | free disk space 2.3, compile 1.0 |
| Linux aarch64 | 43.9 | 4.6 | compile 2.5, apt 0.9 |
| macOS arm64 | 54.9 | 4.4 | free disk space 1.7, compile 0.8 |
| macOS universal | 7.7 | 2.2 | signing and notarisation only on main |
| Flatpak x86_64 / aarch64 | 66.6 / 46.5 | unchanged | full compile
inside flatpak-builder |
| **Wall clock** | **75 min** | **31 min** (Flatpak excluded; 66 with
it) | macOS runner queueing now exceeds job time |
Cache storage: one generation per leg is 400–680 MB compressed at PCH
on, 0.6–2 GB at PCH off; six legs ≈ 7 GB. Without the delete step, 21
main pushes a week would hold ~80 GB of entries that are never read.
## Tests
<!--
> Please describe the tests that you have conducted to verify the
changes made in this PR.
-->
- Thirteen CI runs on this PR, one change per run, with the ccache
statistics printed by every leg: cold (34336272234), warm with PCH
(34346737197), cold and warm without PCH (34352209577, 34364791732), the
macOS include fixes (34435044411, 34435968806 with `ninja -k 0` to list
every remaining file, 34439532375), all legs warm (34444305385), the
keep-only-newest cleanup (34450381312, then 34452640274 after the
Windows CRLF fix), the half-tree invalidation (34452640274), the
five-file change (34463517683, 34464720539), and this final shape
(34466377763, restore-only).
- Unit tests on all five platforms, the profile slice check, the Windows
build-script suite, Shellcheck and the universal DMG build all pass on
the cached binaries.
- The cleanup was verified against the PR's own cache scope: 44 entries
from the earlier runs reduced to exactly one per leg, on all three
platforms, after fixing the CRLF that made `gh cache delete` fail on
Windows.
- A libc++ syntax-only pass over all 1986 C++ translation units on Linux
found the `LocalesUtils.cpp` include; the two wx includes only surface
in a real macOS build and were found with a keep-going build in one
round.
<!--
> A guide for users on how to download the artifacts from this PR.
-->
[How to Download Pull Requests Artifacts for
Testing](https://www.orcaslicer.com/wiki/how_to_download_pr_artifacts)
A miss used to cost a preprocessor pass for the hash and then the real
compile. With the depend mode ccache hashes the include list the
compiler reports, so a miss costs only the compile. Ninja already asks
every compiler here for that list.
Clang records the modification time of every input in the precompiled
header, so a fresh checkout produces a different header and every file
that includes it misses the compiler cache. -fno-pch-timestamp makes the
header reproducible, and pch_defines lets ccache cache the header itself.
The precompiled header no longer has to be turned off when the cache is
on.
Every CI leg compiled the whole tree from scratch, 42 to 57 minutes of
each build job. Objects are now cached with ccache, one entry per leg
kept on the branch that built it: a push saves the cache and drops the
previous entry, a pull request restores main's and keeps nothing.
The precompiled header is turned off whenever the cache is on: Clang
stamps it with the build time, so every file including it missed. With
it off, a warm run hits 98.5 to 98.9 % of compiles and the compile steps
take 1 to 4 minutes; a cold run costs 25 to 60 % more than before, and a
change to a widely included header lands in between.
Ian Chua