0.9.17 Error selecting filement in slots #39

New Issue

harrygeier · 2026-05-31T09:08:23+02:00

harrygeier commented

2026-05-31 09:08:23 +02:00

I installed .9.17 and as predicted i can select filaments

But it does not work correct yet:

i save it:

Slot just shows PLA below and if i click again it brings up a totally different material

I installed .9.17 and as predicted i can select filaments But it does not work correct yet: <img width="324" alt="image.png" src="attachments/ebc42683-c255-4a15-ad23-ec16e4a8bb65"> i save it: Slot just shows PLA below and if i click again it brings up a totally different material <img width="398" alt="image.png" src="attachments/c5b3058e-8b51-48f6-b87a-65da13539b75">

image.png

27 KiB

image.png

23 KiB

viewit added the

bug

label 2026-05-31 10:03:31 +02:00

gangoke commented

2026-05-31 10:15:47 +02:00

I took a peek at orca_filaments.json
Theres many duplicate ids. In your example Both Anycubic PLA High Speed and Tiertime Generic PLA Silk use id: GFL96

I took a peek at orca_filaments.json Theres many duplicate ids. In your example Both Anycubic PLA High Speed and Tiertime Generic PLA Silk use id: GFL96

😕 1

viewit commented

2026-05-31 10:55:11 +02:00

I think both the JSON and a mapping in the code are causing the errors. The fix will be ready tonight or tomorrow at the latest.

harrygeier commented

2026-05-31 11:32:19 +02:00

Also, i do not see my own profiles from Orca, which would be great to have instead of 950 never used ones to scroll along..

viewit commented

2026-05-31 18:25:54 +02:00

orca can not send profile data to the bridge . i think we can make an import function ..

viewit commented

2026-05-31 18:51:51 +02:00

Update — fix is ready on the dev branch, lands in 0.9.18. Three things turned out to be wrong in 0.9.17:

multiColorBox/setInfo was sent over the wrong MQTT topic (slicer/printer/...). The printer silently ignored the command and the next multiColorBox/report then overwrote the cached material in the bridge — so on re-open the dialog showed whatever the printer reported, not what you just saved. Now sent over web/printer/... like the Anycubic Slicer Next does (verified by live MQTT sniff + Workbench-Vue source).
Save flow was racing. Profile-override POST and material POST fired in parallel; depending on which finished first the bridge state could be inconsistent. Now sequential + reloads the local profile map before closing the dialog.
OrcaSlicer filament IDs are not unique — @gangoke nailed it. orca_filaments.json has 68 duplicate IDs; OGFL99 alone is shared by 136 different vendor profiles because Orca lets vendor profiles inherit a generic filament_id from the base. The bridge now uses (vendor, name) as the primary selector, which is 100% unique across all 1002 profiles. The ID is still sent as a hint, but the actual match relies on the vendor+name pair.

Also the slot card now shows the vendor under the material (e.g. PLA / Polymaker) so you can see at a glance which override is active-

Also, i do not see my own profiles from Orca, which would be great to have instead of 950 never used ones to scroll along..

Good point @harrygeier. Two ways forward:

Filter the dropdown to vendors you actually have configured at the printer — easy default, no extra work for the user.
Import your own Orca profiles via a config endpoint — you point the bridge at your local OrcaSlicer user profile directory once, the bridge picks up your custom presets next to the system library.

Ill open a separate enhancement issue for that so it does not get tangled with the present bug fix.

If you want to test the dev build earlier let me know — otherwise it ships with 0.9.18.

Update — fix is ready on the dev branch, lands in 0.9.18. Three things turned out to be wrong in 0.9.17: 1. **`multiColorBox/setInfo` was sent over the wrong MQTT topic** (`slicer/printer/...`). The printer silently ignored the command and the next `multiColorBox/report` then overwrote the cached material in the bridge — so on re-open the dialog showed whatever the printer reported, not what you just saved. Now sent over `web/printer/...` like the Anycubic Slicer Next does (verified by live MQTT sniff + Workbench-Vue source). 2. **Save flow was racing.** Profile-override POST and material POST fired in parallel; depending on which finished first the bridge state could be inconsistent. Now sequential + reloads the local profile map before closing the dialog. 3. **OrcaSlicer filament IDs are not unique** — @gangoke nailed it. `orca_filaments.json` has 68 duplicate IDs; `OGFL99` alone is shared by 136 different vendor profiles because Orca lets vendor profiles inherit a generic `filament_id` from the base. The bridge now uses `(vendor, name)` as the primary selector, which is 100% unique across all 1002 profiles. The ID is still sent as a hint, but the actual match relies on the vendor+name pair. Also the slot card now shows the vendor under the material (e.g. `PLA` / `Polymaker`) so you can see at a glance which override is active- > Also, i do not see my own profiles from Orca, which would be great to have instead of 950 never used ones to scroll along.. Good point @harrygeier. Two ways forward: - **Filter the dropdown to vendors you actually have configured at the printer** — easy default, no extra work for the user. - **Import your own Orca profiles via a config endpoint** — you point the bridge at your local OrcaSlicer user profile directory once, the bridge picks up your custom presets next to the system library. Ill open a separate enhancement issue for that so it does not get tangled with the present bug fix. If you want to test the dev build earlier let me know — otherwise it ships with 0.9.18.

viewit referenced this issue

2026-05-31 18:52:57 +02:00

Slot profile dropdown: filter to / import from users own OrcaSlicer profiles #41

viewit commented

2026-05-31 18:54:14 +02:00

Tracking the vendor-filter / own-profile-import as a separate enhancement: #41

viewit commented

2026-05-31 20:28:06 +02:00

Shipped in v0.9.18 — three fixes:

multiColorBox/setInfo now goes over the right MQTT topic (web/printer/...), so material/colour changes actually stick on the printer.
Save flow is sequential and reloads the local profile map before closing the dialog.
Profile selector is (vendor, name) instead of filament_id (the IDs are not unique in OrcaSlicer — OGFL99 alone is shared by 136 vendor profiles, thanks @gangoke).

The slot card now also shows the vendor under the material so you see at a glance which override is active.

Note: setting a slot only works when the printer is idle, and empty slots cannot be labelled. The follow-up to show only your own vendors / import your local Orca profiles is tracked in #41.

Closing — please reopen if the round-trip still misbehaves in 0.9.18.

Shipped in [v0.9.18](https://gitea.it-drui.de/viewit/KX-Bridge-Release/releases/tag/v0.9.18) — three fixes: 1. `multiColorBox/setInfo` now goes over the right MQTT topic (`web/printer/...`), so material/colour changes actually stick on the printer. 2. Save flow is sequential and reloads the local profile map before closing the dialog. 3. Profile selector is `(vendor, name)` instead of `filament_id` (the IDs are not unique in OrcaSlicer — `OGFL99` alone is shared by 136 vendor profiles, thanks @gangoke). The slot card now also shows the vendor under the material so you see at a glance which override is active. Note: setting a slot only works when the printer is idle, and empty slots cannot be labelled. The follow-up to show only your own vendors / import your local Orca profiles is tracked in #41. Closing — please reopen if the round-trip still misbehaves in 0.9.18.

viewit closed this issue

2026-05-31 20:28:06 +02:00

Sign in to join this conversation.

3 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: viewit/KX-Bridge-Release#39