Using Pain001

CLI workflow

Run this in a fresh writable directory using synthetic bundled data:

pain001 init pain.001.001.03 -o payments.csv
pain001 validate -t pain.001.001.03 -d payments.csv
pain001 generate -t pain.001.001.03 -d payments.csv -o output

The generate command’s -o flag names an output directory. Template and schema paths resolve from the bundled registry unless explicitly overridden. Exit codes are 0 for success, 1 for processing/validation failure, and 2 for invalid arguments or configuration. Dry-run validates data without writing XML; it is not a bank-acceptance guarantee.

Use generated help rather than a duplicate flag table:

pain001 generate --help
pain001 versions
pain001 inspect pain.001.001.03

Inputs and validation

CSV, SQLite, JSON and JSONL are supported, as are Python records and optional Parquet input. Installed plugins can add formats such as XLSX. See Plugin architecture for privileges, discovery and disable controls, and Input columns: the flat vocabulary of the CSV pipeline for schema-derived field mappings.

Generation recomputes control totals and validates rendered XML against the bundled XSD before writing. Scheme validation, including anti-duplicate, adds domain rules but does not certify compliance with a bank’s private rules. Streaming creates one output per chunk.

Python: in-memory output

This example reads synthetic data shipped in the installed package:

from pain001 import generate_xml_string
from pain001.constants import TEMPLATES_DIR
from pain001.csv.load_csv_data import load_csv_data

message_type = "pain.001.001.03"
bundle = TEMPLATES_DIR / message_type
rows = load_csv_data(str(bundle / "template.csv"))
xml = generate_xml_string(
    rows,
    message_type,
    str(bundle / "template.xml"),
    str(bundle / f"{message_type}.xsd"),
)

Python: explicit file output

from pain001.core.core import process_files
from pain001.constants import TEMPLATES_DIR

message_type = "pain.001.001.03"
bundle = TEMPLATES_DIR / message_type
output = process_files(
    xml_message_type=message_type,
    xml_template_file_path=str(bundle / "template.xml"),
    xsd_schema_file_path=str(bundle / f"{message_type}.xsd"),
    data_file_path=str(bundle / "template.csv"),
    output_path="payment.xml",
)

Unlike the CLI directory flag, the Python output_path argument names a file. Always provide it so an installed template directory is not used as output.

Optional and unreleased integrations

REST is started with pain001 serve after installing the api extra. The feature branch also supports Request-local payment policies through CLI and REST, review-only MCP corrections, and explicit Explicit SFTP delivery. Generation never uploads a payment automatically. See Open-issue audit for feat/v0.0.71 for branch availability, companion work and unmet acceptance criteria.