API ​

pacs008 offers a REST API, CLI, and Python library for payment data checks and XML generation.

Implementation notes ​

  • Use sync generation for quick checks.
  • Use async generation for large files.
  • Keep the input payload and validation report together.

Installation ​

Install the package from PyPI. You need Python 3.9.2 or later.

bash
python -m pip install pacs008

REST API ​

Start the FastAPI server for HTTP validation and XML generation.

Start the server ​

bash
uvicorn pacs008.api.app:app --reload --host 0.0.0.0 --port 8000

Endpoints ​

EndpointDescription
GET /healthHealth check that returns service status
POST /validateValidate payment data without generating XML
POST /generateGenerate XML now and return the file
POST /generate/asyncSubmit an async generation job
GET /status/{job_id}Check job status by ID
GET /download/{job_id}Download XML after the job completes
DELETE /jobs/{job_id}Cancel a pending or running job
GET /docsSwagger UI for testing all endpoints

See Message Types for the full list of supported message pages.

Validation example ​

Validate payment data before you generate XML.

bash
curl -X POST http://localhost:8000/api/validate \
  -H "Content-Type: application/json" \
  -d '{
    "message_type": "pacs.008.001.13",
    "data": [{
      "msg_id": "MSG-2026-001",
      "creation_date_time": "2026-01-15T10:30:00",
      "nb_of_txs": "1",
      "settlement_method": "CLRG",
      "interbank_settlement_date": "2026-01-15",
      "end_to_end_id": "E2E-INV-2026-001",
      "interbank_settlement_amount": "25000.00",
      "interbank_settlement_currency": "EUR",
      "charge_bearer": "SHAR",
      "debtor_name": "Acme Corp GmbH",
      "debtor_agent_bic": "DEUTDEFF",
      "creditor_agent_bic": "COBADEFF",
      "creditor_name": "Widget Industries SA"
    }]
  }'
json
{
  "valid": true,
  "message_type": "pacs.008.001.13",
  "errors": [],
  "warnings": []
}

Synchronous generation example ​

Generate a pacs.008.001.13 XML file from JSON.

bash
curl -X POST http://localhost:8000/api/generate \
  -H "Content-Type: application/json" \
  -d '{
    "message_type": "pacs.008.001.13",
    "template": "pacs008/templates/pacs.008.001.13/template.xml",
    "schema": "pacs008/templates/pacs.008.001.13/pacs.008.001.13.xsd",
    "data": [{
      "msg_id": "MSG-2026-001",
      "creation_date_time": "2026-01-15T10:30:00",
      "nb_of_txs": "1",
      "settlement_method": "CLRG",
      "interbank_settlement_date": "2026-01-15",
      "end_to_end_id": "E2E-INV-2026-001",
      "tx_id": "TX-001",
      "interbank_settlement_amount": "25000.00",
      "interbank_settlement_currency": "EUR",
      "charge_bearer": "SHAR",
      "debtor_name": "Acme Corp GmbH",
      "debtor_agent_bic": "DEUTDEFF",
      "creditor_agent_bic": "COBADEFF",
      "creditor_name": "Widget Industries SA"
    }]
  }' --output pacs008_output.xml

Asynchronous generation ​

For large files, submit an async job and poll until it finishes.

bash
# Submit the job
JOB=$(curl -s -X POST http://localhost:8000/api/generate/async \
  -H "Content-Type: application/json" \
  -d '{"message_type":"pacs.008.001.13","data":[...]}')

JOB_ID=$(echo $JOB | jq -r '.job_id')

# Poll for completion
curl http://localhost:8000/api/status/$JOB_ID

# Download the result
curl http://localhost:8000/api/download/$JOB_ID --output result.xml
json
{
  "job_id": "8f7f0d4b-7df9-4d1a-8d47-19f4f28b6d38",
  "status": "completed",
  "message_type": "pacs.008.001.13",
  "download_url": "/api/download/8f7f0d4b-7df9-4d1a-8d47-19f4f28b6d38"
}

CLI ​

The CLI takes a data file, message version, template, and schema. It validates the input and writes XML to the output directory.

Basic usage ​

bash
pacs008 -t <message_type> \
  -m <template_file> \
  -s <schema_file> \
  -d <data_file>

Example ​

bash
pacs008 -t pacs.008.001.13 \
  -m pacs008/templates/pacs.008.001.13/template.xml \
  -s pacs008/templates/pacs.008.001.13/pacs.008.001.13.xsd \
  -d payments.csv

Dry-run mode ​

Use --dry-run to validate input data without generating XML. The exit code shows whether validation passed (0) or failed (1).

bash
pacs008 -t pacs.008.001.13 \
  -m pacs008/templates/pacs.008.001.13/template.xml \
  -s pacs008/templates/pacs.008.001.13/pacs.008.001.13.xsd \
  -d payments.csv \
  --dry-run

Add --verbose for detailed output during generation.

Python API ​

Use the library directly in Python scripts or services.

Generate XML from payment records ​

python
from pacs008 import generate_xml_string

payments = [{
    "msg_id": "MSG-2026-001",
    "creation_date_time": "2026-01-15T10:30:00",
    "nb_of_txs": "1",
    "settlement_method": "CLRG",
    "interbank_settlement_date": "2026-01-15",
    "end_to_end_id": "E2E-INV-2026-001",
    "tx_id": "TX-001",
    "interbank_settlement_amount": "25000.00",
    "interbank_settlement_currency": "EUR",
    "charge_bearer": "SHAR",
    "debtor_name": "Acme Corp GmbH",
    "debtor_agent_bic": "DEUTDEFF",
    "creditor_agent_bic": "COBADEFF",
    "creditor_name": "Widget Industries SA",
}]

xml = generate_xml_string(
    payments,
    "pacs.008.001.13",
    "pacs008/templates/pacs.008.001.13/template.xml",
    "pacs008/templates/pacs.008.001.13/pacs.008.001.13.xsd",
)
print(xml)

SWIFT compliance check ​

Check and clean data against SWIFT character and field-length rules before generation.

python
from pacs008.compliance import cleanse_data_with_report

raw = [{"debtor_name": "Müller & Söhne™", "msg_id": "X" * 50}]
clean, report = cleanse_data_with_report(raw)
print(report.summary())

Docker ​

Run the API in a container using the bundled Dockerfile.

bash
docker build -t pacs008:latest .
docker run -p 8000:8000 pacs008:latest
bash
docker run --rm   -e PACS008_LOG_LEVEL=INFO   -v $PWD/examples:/data   -p 8000:8000 pacs008:latest

IBAN and BIC validation ​

Validate financial identifiers independently of XML generation.

python
from pacs008.validation import validate_iban, validate_bic

is_valid, error = validate_iban("DE89370400440532013000", strict=False)
is_valid, error = validate_bic("DEUTDEFF", strict=False)

Streaming ​

Load large datasets in configurable chunks to limit memory usage.

python
from pacs008.data.loader import load_payment_data_streaming

for chunk in load_payment_data_streaming("large_payments.csv", chunk_size=500):
    print(f"Processing {len(chunk)} records")
python
from pacs008.validation import validate_batch

for chunk in load_payment_data_streaming("large_payments.csv", chunk_size=500):
    report = validate_batch(chunk, "pacs.008.001.13")
    print(report.summary())

Validation service ​

Run the full pre-generation validation pipeline programmatically.

python
from pacs008.validation import ValidationService, ValidationConfig

service = ValidationService()
report = service.validate_all(ValidationConfig(
    xml_message_type="pacs.008.001.13",
    xml_template_file_path="pacs008/templates/pacs.008.001.13/template.xml",
    xsd_schema_file_path="pacs008/templates/pacs.008.001.13/pacs.008.001.13.xsd",
    data_file_path="payments.csv",
))
print(report.is_valid, report.errors)

Required data fields ​

Each payment record must include these fields. Version-specific fields are listed below.

FieldDescriptionConstraint
msg_idMessage identifierMax 35 characters
creation_date_timeCreation timestampISO 8601 format
nb_of_txsNumber of transactionsPositive integer
settlement_methodSettlement methodCLRG, INDA, COVE, or INGA
end_to_end_idEnd-to-end identifierMax 35 characters
interbank_settlement_amountInterbank settlement amountDecimal, e.g. `25000.00`
interbank_settlement_currencySettlement currencyISO 4217 code
charge_bearerCharge bearerDEBT, CRED, SHAR, or SLEV
debtor_nameDebtor nameMax 140 characters
debtor_agent_bicDebtor agent BIC8 or 11 characters
creditor_agent_bicCreditor agent BIC8 or 11 characters
creditor_nameCreditor nameMax 140 characters

Version-specific fields ​

FieldDescriptionConstraint
uetrUnique end-to-end transaction referenceUUID format — available from v08
mandate_idMandate identifierAvailable from v10
expiry_date_timeMessage expiry timestampAvailable in v13

Last updated: