Official Python client for the BAPP Auto API. Provides a simple, consistent interface for authentication, entity CRUD, and task execution.
pip install bapp-api-clientfrom bapp_api_client import BappApiClient
client = BappApiClient(token="your-api-key")# List with filters
countries = client.list("core.country", page=1, search="Romania")
# Get by ID
country = client.get("core.country", "42")
# Create
new = client.create("core.country", {"name": "Romania", "code": "RO"})
# Update (full)
client.update("core.country", "42", {"name": "Romania", "code": "RO"})
# Patch (partial)
client.patch("core.country", "42", {"code": "RO"})
# Delete
client.delete("core.country", "42")The client supports Token (API key) and Bearer (JWT / OAuth) authentication.
Token auth already includes a tenant binding, so you don't need to specify tenant separately.
# Static API token (tenant is included in the token)
client = BappApiClient(token="your-api-key")
# Bearer (JWT / OAuth)
client = BappApiClient(bearer="eyJhbG...", tenant="1")tenant and app can be changed at any time after construction:
client.tenant = "2"
client.app = "wms"| Option | Description | Default |
|---|---|---|
token |
Static API token (Token <value>) — includes tenant |
— |
bearer |
Bearer / JWT token | — |
host |
API base URL | https://panel.bapp.ro/api |
tenant |
Tenant ID (x-tenant-id header) |
None |
app |
App slug (x-app-slug header) |
"account" |
timeout |
HTTP request timeout (seconds) | 30 |
max_retries |
Max retries on transient errors (5xx, 429, connection) | 3 |
| Method | Description |
|---|---|
me() |
Get current user profile |
get_app(app_slug) |
Get app configuration by slug |
list(content_type, **filters) |
List entities (paginated) |
get(content_type, id) |
Get a single entity |
create(content_type, data) |
Create an entity |
update(content_type, id, data) |
Full update (PUT) |
patch(content_type, id, data) |
Partial update (PATCH) |
delete(content_type, id) |
Delete an entity |
list_introspect(content_type) |
Get list view metadata |
detail_introspect(content_type) |
Get detail view metadata |
get_document_views(record) |
Extract available views from a record |
get_document_url(record, output?, label?, variation?) |
Build a render/download URL |
get_document_content(record, output?, label?, variation?) |
Fetch document bytes (PDF, HTML, JPG) |
download_document(record, dest, output?, label?, variation?) |
Stream document to file (memory-efficient) |
list_tasks() |
List available task codes |
detail_task(code) |
Get task configuration |
run_task(code, payload?) |
Execute a task |
run_task_async(code, payload?) |
Run a long-running task and poll until done |
list() returns the results directly as a list/array. Pagination metadata is
available as extra attributes:
count— total number of items across all pagesnext— URL of the next page (ornull)previous— URL of the previous page (ornull)
When data contains file objects, the client automatically switches from JSON to
multipart/form-data. Mix regular fields and files in the same call:
# File objects, byte strings, or tuples (filename, file) are auto-detected
client.create("myapp.document", {
"name": "Report",
"file": open("report.pdf", "rb"),
})
# Tuple form for explicit filename / content-type
client.create("myapp.document", {
"name": "Report",
"file": ("report.pdf", open("report.pdf", "rb"), "application/pdf"),
})
# Also works with tasks
client.run_task("myapp.import_data", {
"format": "csv",
"file": open("data.csv", "rb"),
})Records may include public_view and/or view_token fields with JWT tokens
for rendering documents (invoices, orders, reports, etc.) as HTML, PDF, or images.
The SDK normalises both formats and builds the correct URL automatically:
order = client.get("company_order.order", "42")
# Get a PDF download URL (auto-detects public_view vs view_token)
url = client.get_document_url(order, output="pdf")
# Pick a specific view by label
url = client.get_document_url(order, output="html", label="Comanda interna")
# Use a variation (e.g. for different report periods)
url = client.get_document_url(order, output="pdf", variation="v4")
# Fetch the actual content as bytes
pdf_bytes = client.get_document_content(order, output="pdf")
with open("order.pdf", "wb") as f:
f.write(pdf_bytes)
# Enumerate all available views
views = client.get_document_views(order)
for v in views:
print(v["label"], v["type"], v.get("variations"))get_document_views() returns a list of normalised view entries with label,
token, type ("public_view" or "view_token"), variations, and
default_variation. Use it to enumerate available views (e.g. for a dropdown).
Tasks are server-side actions identified by a dotted code (e.g. myapp.export_report).
# List all tasks
tasks = client.list_tasks()
# Inspect a task
cfg = client.detail_task("myapp.export_report")
# Run without payload (GET)
result = client.run_task("myapp.export_report")
# Run with payload (POST)
result = client.run_task("myapp.export_report", {"format": "csv"})Some tasks run asynchronously on the server. When triggered, they return an id
that can be polled via bapp_framework.taskdata. Use run_task_async() to
handle this automatically — it polls until finished is true and returns the
final task data (which includes a file URL when the task produces a download).
MIT