The service runs as a separate Docker container. It does not modify Bambuddy or InvenTree source code. Bambuddy is used as the print-history source, while InvenTree remains the stock-control system of record.

What It Does

For each successful Bambuddy archive, the service:

finds or creates an InvenTree Part;
creates one InvenTree StockItem for the Bambuddy archive;
writes Weight and PrintTime part parameters when those templates exist in InvenTree;
uploads the Bambuddy archive thumbnail to the InvenTree part image;
writes an InvenTree part URL back into Bambuddy archive external_url;
stores sync state in SQLite to prevent duplicate stock items.

Failed, stopped, running, or still-printing archives are skipped when SYNC_SUCCESS_ONLY=true.

Data Flow

Bambuddy Archives
        |
        | webhook or polling
        v
bambuddy-inventree-sync
        |
        | InvenTree API
        v
Part + StockItem + Parameters + Image
        |
        | Bambuddy API
        v
Archive external_url -> InvenTree Part page

InvenTree Mapping

Bambuddy data	InvenTree target
archive/model identity	`Part`
successful print archive	`StockItem`
target category	`Part.category`
target storage	`StockItem.location`
archive id	`StockItem.batch = bambuddy-<archive_id>`
filament weight	`Part` parameter `Weight`
print time	`Part` parameter `PrintTime`
archive thumbnail	`Part.image`
InvenTree part page	Bambuddy `external_url`

Duplicate Protection

The service is idempotent:

each Bambuddy archive ID is stored in data/sync.sqlite3;
each StockItem gets batch bambuddy-<archive_id>;
rerunning backfill does not create duplicate stock items;
repeat prints of the same file/model reuse the same Part and create new StockItem rows.

The Part identity key is controlled by:

PART_KEY_FIELDS=content_hash,filename,print_name,name

The generated key becomes an InvenTree IPN:

BMB-<first-12-sha1-chars>

Delete Policy

If an archive is deleted in Bambuddy after it was synced, the corresponding InvenTree data is not deleted.

This is intentional. InvenTree is the inventory record, and deleting print history in Bambuddy should not silently remove stock records. The synced Part, StockItem, parameters, and image remain in InvenTree.

Requirements

Docker with Linux containers enabled.
Network access from the sync container to Bambuddy API and InvenTree API.
Bambuddy API key.
InvenTree API token.
Existing InvenTree part category ID.
Existing InvenTree stock location ID.
Optional InvenTree parameter templates: Weight and PrintTime.

On Windows Server 2022, verify Docker with:

docker run --rm hello-world
docker run --rm python:3.12-slim python --version

Docker Deployment

Clone the repository:

git clone https://git.tcom.space/tcom/Lab8DATAPROCESSOR.git
cd Lab8DATAPROCESSOR

Create the runtime config:

Copy-Item .env.example .env
notepad .env

Start the service:

docker compose up -d --build

View logs:

docker compose logs --tail=100

Stop the service:

docker compose down

Update to the latest version:

git pull
docker compose up -d --build

Configuration

Example .env:

BAMBUDDY_BASE_URL=http://192.168.1.5:8000/api/v1
BAMBUDDY_API_KEY=replace-with-bambuddy-api-key

INVENTREE_BASE_URL=http://192.168.0.3:1337
INVENTREE_WEB_URL=
INVENTREE_TOKEN=replace-with-inventree-token
INVENTREE_PART_CATEGORY_ID=26
INVENTREE_STOCK_LOCATION_ID=98

SERVICE_API_TOKEN=change-me
WEBHOOK_SHARED_SECRET=

SYNC_SUCCESS_ONLY=true
SYNC_PART_IMAGES=true
OVERWRITE_PART_IMAGES=false
SYNC_ARCHIVE_EXTERNAL_LINK=true
OVERWRITE_ARCHIVE_EXTERNAL_LINK=false

DEFAULT_STOCK_QUANTITY=1
INVENTREE_STOCK_STATUS=10
PART_IPN_PREFIX=BMB
PART_KEY_FIELDS=content_hash,filename,print_name,name

BACKFILL_PAGE_SIZE=50
POLL_INTERVAL_SECONDS=300
SYNC_ON_STARTUP=false
HTTP_TIMEOUT_SECONDS=30
DATA_DIR=/data

Do not commit .env. It contains API tokens and is ignored by git.

Important Settings

BAMBUDDY_BASE_URL: Bambuddy API base URL. Use /api/v1.
INVENTREE_BASE_URL: InvenTree root URL or /api URL. The service normalizes it internally.
INVENTREE_WEB_URL: Browser-facing InvenTree URL for Bambuddy external links. If empty, INVENTREE_BASE_URL is used.
INVENTREE_PART_CATEGORY_ID: Existing InvenTree category where auto-created printed parts are placed.
INVENTREE_STOCK_LOCATION_ID: Existing InvenTree stock location where printed stock items are placed.
SYNC_SUCCESS_ONLY: When true, only successful/completed prints are imported.
POLL_INTERVAL_SECONDS: Enables automatic periodic backfill. 300 means every 5 minutes. 0 disables polling.
SYNC_PART_IMAGES: Downloads Bambuddy thumbnail and uploads it to InvenTree Part.image.
OVERWRITE_PART_IMAGES: When false, manually set InvenTree part images are preserved.
SYNC_ARCHIVE_EXTERNAL_LINK: Writes the InvenTree part page URL into Bambuddy external_url.
OVERWRITE_ARCHIVE_EXTERNAL_LINK: When false, existing non-InvenTree external links in Bambuddy are preserved.

InvenTree IDs

Use numeric IDs for target category and stock location. You can find them from the InvenTree UI URL or API.

Example:

curl.exe -H "Authorization: Token YOUR_TOKEN" http://192.168.0.3:1337/api/part/category/26/
curl.exe -H "Authorization: Token YOUR_TOKEN" http://192.168.0.3:1337/api/stock/location/98/

Current known example:

INVENTREE_PART_CATEGORY_ID=26
INVENTREE_STOCK_LOCATION_ID=98

Validate Connectivity

After the container starts:

curl.exe http://localhost:8088/health
curl.exe -H "X-Service-Token: change-me" http://localhost:8088/validate

Expected result:

/health returns status: ok;
/validate confirms Bambuddy, InvenTree category, and InvenTree location are reachable.

Initial Backfill

Test one successful archive first:

curl.exe -X POST -H "X-Service-Token: change-me" "http://localhost:8088/sync/backfill?max_archives=1"

Run full backfill:

curl.exe -X POST -H "X-Service-Token: change-me" http://localhost:8088/sync/backfill

Check status:

curl.exe -H "X-Service-Token: change-me" http://localhost:8088/sync/status

When SYNC_SUCCESS_ONLY=true, still-printing or failed archives are skipped and do not consume max_archives.

Automatic Operation

Recommended production mode:

SYNC_SUCCESS_ONLY=true
POLL_INTERVAL_SECONDS=300
SYNC_ON_STARTUP=false

This makes the service check Bambuddy every 5 minutes and import all newly completed prints.

For near-real-time syncing, also configure a Bambuddy webhook:

http://WINDOWS-SERVER-IP:8088/webhooks/bambuddy

Use the print_complete event. If WEBHOOK_SHARED_SECRET is set, Bambuddy must send:

X-Sync-Secret: your-secret

Running both webhook and polling is recommended. Webhook gives fast sync, polling catches missed events after restarts or temporary failures.

Manual Sync

Sync one archive:

curl.exe -X POST -H "X-Service-Token: change-me" http://localhost:8088/sync/archive/11

Force re-sync one archive:

curl.exe -X POST -H "X-Service-Token: change-me" "http://localhost:8088/sync/archive/11?force=true"

Force sync is useful after changing image, parameter, or external link behavior.

API Endpoints

GET  /health
GET  /validate
GET  /sync/status
POST /sync/archive/{archive_id}
POST /sync/backfill
POST /webhooks/bambuddy

Manual sync endpoints require X-Service-Token when SERVICE_API_TOKEN is set.

Troubleshooting

View recent logs:

docker compose logs --tail=100

Rebuild after pulling code changes:

git pull
docker compose up -d --build

Common issues:

401 from service endpoints: missing or wrong X-Service-Token.
401 from Bambuddy: wrong BAMBUDDY_API_KEY.
401 from InvenTree: wrong INVENTREE_TOKEN.
InvenTree description length errors: update to the latest service version; descriptions are capped at 250 characters.
No items imported: check if Bambuddy archives are still printing or failed.
Duplicate protection prevents repeated imports: use force=true only when you want to refresh parameters/images/links for an existing archive.

Backups

Persistent sync state is stored in:

./data/sync.sqlite3

Back up the data directory together with .env if you move the service to another server.