Files
bambuddy/backend/app/services/slice_preview.py
T
maziggy d5153f1de3 feat(slicer): live progress + filament discovery polish + OrcaSlicer warning
End-to-end live progress, two correctness fixes, and a UX warning around
  the upstream OrcaSlicer bugs we discovered while testing.

  LIVE PROGRESS
  =============

  Wire OrcaSlicer / BambuStudio's --pipe progress channel through the
  sidecar -> Bambuddy -> persistent toast so a user-initiated slice shows
  "{name} -- Generating G-code (75%) -- 47s" instead of just elapsed time.
  The same wiring covers the SliceModal's filament-analysis preview slice
  (the real slice that fires before profile picking, used to discover
  which AMS slots an unsliced plate consumes) and the embedded-settings
  fallback path triggered by Orca's --load-settings segfault on complex
  H2D models.

  - Sidecar (orca-slicer-api/bambuddy/profile-resolver, separate commit):
    switch /slice from execFile to spawn, mkfifo per request, parse the
    CLI's structured JSON progress events into a per-process
    ProgressStore, expose GET /slice/progress/:requestId.
  - Bambuddy backend: slicer_api.slice_with_profiles + slice_without_profiles
    accept request_id + on_progress, spawn a 1Hz parallel poller that
    forwards each snapshot via SliceDispatchService.set_progress(job_id,
    ...) onto the matching SliceJob; GET /slice-jobs/:id includes the
    latest snapshot on every poll. The 404 from the early-race window
    (POST fired before sidecar's progressStore.start) is treated as a
    retry rather than terminal -- otherwise the poller bailed before any
    progress could ever arrive.
  - /api/v1/slicer/preview-progress/:requestId proxies the sidecar's
    progress endpoint for the modal's filament-discovery flow (the
    /filament-requirements call is server-originated; the browser can't
    reach the sidecar directly).
  - Frontend: SliceJobTrackerContext re-renders the persistent toast with
    the new format when a useful progress frame is present, falls back
    to elapsed-time-only when the sidecar hasn't emitted yet or doesn't
    support progress. SliceModal.FilamentAnalysisSpinner generates a
    per-(source, plate) UUID, polls the proxy at 1Hz, and mirrors the
    inline spinner contents into a separate persistent toast so the
    preview slice doesn't feel silent either.

  CORRECTNESS FIXES
  =================

  - MakerWorld imports were persisting URL-encoded filenames verbatim
    ("stormtrooper-helmet%20h2d.3mf"). Backend now urllib.parse.unquote
    s the manifest-supplied name and the URL path-tail fallback before
    passing to save_3mf_bytes_to_library; frontend defensively
    decodeURIComponent s in the slice toast / analysis spinner so
    already-imported rows display cleanly without a backfill migration.
  - The fallback path's slice_without_profiles call now forwards the
    same request_id + on_progress as the primary slice_with_profiles
    call so the toast keeps updating across the segfault -> embedded-
    settings retry boundary instead of going blank.

  ORCASLICER WARNING
  ==================

  Verified two upstream OrcaSlicer CLI bugs reproduce on the latest
  nightly (2.4.0-dev, 2026-04-28) with the help of an isolated AppImage
  extract and a minimal sentinel-value-injected cube fixture:
    - OrcaSlicer/OrcaSlicer#12426 -- SIGSEGV in
      update_values_to_printer_extruders_for_multiple_filaments on
      painted multi-extruder 3MFs (commented on the existing thread,
      not a new issue)
    - OrcaSlicer/OrcaSlicer#13386 -- CLI strict-validates parameter
      values BambuStudio writes by default (solid_infill_filament: 0,
      tree_support_wall_count: -1, prime_tower_brim_width: -1) and
      rejects with exit 238, even though Orca's own GUI tolerates
      them (filed by us alongside this change)

  Settings -> Workflow -> Slicer card renders an amber inline warning
  under the preferred-slicer dropdown when orcaslicer is selected,
  linking both upstream issues and recommending BambuStudio until the
  fixes land. Option stays pickable -- users who only slice STLs aren't
  affected by either bug.
2026-04-28 16:36:49 +02:00

183 lines
6.9 KiB
Python

"""Preview-slice cache for the SliceModal.
The slice modal needs the per-plate filament list before the user picks
profiles. For sliced files this lives in ``Metadata/slice_info.config`` and
the ``/filament-requirements`` endpoint can read it directly. For unsliced
project files it doesn't exist yet — only the slicer can produce it, since
Bambu Studio applies its own pruning to painted-face data at slice time.
This module wraps the sidecar's ``slice_without_profiles`` call so the
endpoint can run a preview slice with the project's embedded settings,
parse the result's slice_info, and return the actual filament list. Results
are cached by ``(kind, source_id, plate_id, content_hash)`` so repeat
opens of the modal on the same plate are instant; LRU eviction keeps the
cache bounded. Hash invalidation handles in-place file replacement; no TTL
is used because preview-slice output is deterministic for a given file
content.
"""
from __future__ import annotations
import asyncio
import hashlib
import logging
import zipfile
from collections import OrderedDict
from io import BytesIO
import defusedxml.ElementTree as ET
from backend.app.services.slicer_api import (
SlicerApiError,
SlicerApiService,
)
logger = logging.getLogger(__name__)
_PREVIEW_CACHE_MAX = 256
# Cache values: list[dict] on success, [] on parsed-but-empty (slicer
# returned a 3MF without filament data for this plate — caching the negative
# avoids burning 30s+ per modal open on a known-bad input).
_preview_cache: OrderedDict[tuple[str, int, int, str], list[dict]] = OrderedDict()
# Per-key locks prevent N concurrent modal opens on the same (file, plate)
# from launching N redundant preview slices — only the first one runs, the
# rest wait and read from the cache. Locks are evicted alongside cache
# entries to keep the dict bounded; we do NOT cache transient sidecar
# failures (network errors etc.) so those retry naturally on next request.
_preview_locks: dict[tuple[str, int, int, str], asyncio.Lock] = {}
def _content_hash(file_bytes: bytes) -> str:
return hashlib.sha256(file_bytes).hexdigest()[:16]
async def get_preview_filaments(
*,
kind: str,
source_id: int,
plate_id: int,
file_bytes: bytes,
file_name: str,
api_url: str,
request_id: str | None = None,
) -> list[dict] | None:
"""Run a preview slice for ``plate_id`` using the file's embedded settings,
parse the resulting slice_info, and return the per-plate filament list.
Returns ``None`` when the preview slice fails — the caller should fall
back to whatever heuristic it has (typically the project_filaments +
painted-face approach in ``threemf_tools``).
"""
h = _content_hash(file_bytes)
key = (kind, source_id, plate_id, h)
cached = _preview_cache.get(key)
if cached is not None:
_preview_cache.move_to_end(key)
return cached
lock = _preview_locks.setdefault(key, asyncio.Lock())
async with lock:
# Re-check after acquiring the lock — another coroutine may have
# populated the cache while we were waiting on it.
cached = _preview_cache.get(key)
if cached is not None:
_preview_cache.move_to_end(key)
return cached
try:
async with SlicerApiService(base_url=api_url) as svc:
result = await svc.slice_without_profiles(
model_bytes=file_bytes,
model_filename=file_name,
plate=plate_id,
export_3mf=True,
request_id=request_id,
)
except SlicerApiError as e:
logger.warning(
"Preview slice failed for %s/%s plate %s: %s",
kind,
source_id,
plate_id,
e,
)
return None
except Exception as e: # noqa: BLE001 — never break the modal on sidecar issues
logger.warning("Preview slice unexpected error: %s", e)
return None
filaments = _parse_filaments_from_sliced_3mf(result.content, plate_id)
# Negative-cache the parse failure: a slice that succeeds but yields
# no parsable filament data for this plate is a deterministic
# property of the input. Re-running the slice produces the same
# result, just N seconds slower. Empty list signals "preview was
# tried, no usable data" so the caller can fall through.
cache_value: list[dict] = filaments if filaments is not None else []
_preview_cache[key] = cache_value
if len(_preview_cache) > _PREVIEW_CACHE_MAX:
evicted_key, _ = _preview_cache.popitem(last=False)
# Drop the matching lock so the dict doesn't grow forever.
# Safe to discard: the lock isn't held here, and any later
# request for the same key will mint a fresh lock.
_preview_locks.pop(evicted_key, None)
return filaments
def _parse_filaments_from_sliced_3mf(content: bytes, plate_id: int) -> list[dict] | None:
"""Extract ``<filament>`` entries for ``plate_id`` from a sliced 3MF's
Metadata/slice_info.config. Returns ``None`` on any parse error so the
caller knows to fall back."""
try:
with zipfile.ZipFile(BytesIO(content)) as zf:
if "Metadata/slice_info.config" not in zf.namelist():
return None
data = zf.read("Metadata/slice_info.config").decode()
except (zipfile.BadZipFile, OSError):
return None
try:
root = ET.fromstring(data)
except ET.ParseError:
return None
for plate_elem in root.findall(".//plate"):
idx = None
for meta in plate_elem.findall("metadata"):
if meta.get("key") == "index":
try:
idx = int(meta.get("value", ""))
except (ValueError, TypeError):
pass
break
if idx != plate_id:
continue
out: list[dict] = []
for f in plate_elem.findall("filament"):
fid = f.get("id")
if not fid:
continue
try:
slot_id = int(fid)
except (ValueError, TypeError):
continue
try:
used_grams = float(f.get("used_g", "0"))
except (ValueError, TypeError):
used_grams = 0
try:
used_meters = float(f.get("used_m", "0"))
except (ValueError, TypeError):
used_meters = 0
out.append(
{
"slot_id": slot_id,
"type": f.get("type", ""),
"color": f.get("color", ""),
"used_grams": round(used_grams, 1),
"used_meters": used_meters,
"tray_info_idx": f.get("tray_info_idx", ""),
},
)
return sorted(out, key=lambda x: x["slot_id"])
return None