Readme

pyA2L is an ASAM MCD-2MC processing library written in Python.

If you work with ECUs, ASAP2/A2L is the contract that describes what and how to measure or calibrate. pyA2L helps you parse once, inspect and validate programmatically, automate checks, and export back when needed — all from Python.

Contents

• About ASAM MCD-2 MC (ASAP2)
• What pyA2L offers
• Installation
• Getting Started (Quickstart)
• Command-line usage
• Tips
• Examples
• Compatibility
• Project links
• Contributing
• Code of Conduct
• License
• Changelog / Release notes
• Acknowledgements

About ASAM MCD-2 MC (ASAP2)

ASAM MCD-2 MC (ASAP2) is the ASAM standard that defines the A2L description format for ECU measurement signals and calibration parameters. In practice, the A2L acts as a contract between ECU software and tools so different vendors can consistently locate data in memory and convert raw values into physical engineering units. Runtime transport (e.g., CCP/XCP) is out of scope of the standard.

For an authoritative overview of the standard, see the ASAM page: https://www.asam.net/standards/detail/mcd-2-mc/

What pyA2L offers

• Parse .a2l files and persist them as compact, queryable SQLite databases (.a2ldb) to avoid repeated parse costs.
• Programmatic access to ASAP2 entities via SQLAlchemy ORM models (MODULE, MEASUREMENT, CHARACTERISTIC, AXIS_DESCR, RECORD_LAYOUT, COMPU_METHOD/COMPU_TAB, UNIT, FUNCTION, GROUP, VARIANT_CODING, etc.).
• Rich inspection helpers in pya2l.api.inspect (e.g., Characteristic, Measurement, AxisDescr, ModPar, ModCommon) to compute shapes, axis info, allocated memory, conversions, and more.
• Creator API in pya2l.api.create to programmatically build or augment A2L content (PROJECT, MODULE, MEASUREMENT, CHARACTERISTIC, COMPU_METHOD, AXIS_PTS, RECORD_LAYOUT, GROUP, FUNCTION, etc.).
• Validation utilities (pya2l.api.validate) to check common ASAP2 rules and project-specific consistency.
• Export .a2ldb content back to A2L text or JSON when needed.
• Building blocks for automation: reporting, quality gates, CI checks, and integration with CCP/XCP workflows.

Supported ASAP2 version: 1.7.1

Why pyA2L?

• Parse once, query fast: Avoid repeated parser runs by working from SQLite.
• Powerful model: Use SQLAlchemy ORM to navigate ASAP2 entities naturally.
• Beyond parsing: Inspect derived properties, validate consistency, and export back to A2L.
• Automate: Integrate with CI to enforce quality gates on A2L content.

Design highlights

• SQLite-backed storage (.a2ldb) with SQLAlchemy models
• C++ parser (ANTLR4-based) for fast parsing (2.5 MB/s)
• Adaptive flush strategy automatically tunes performance (10% speedup for large files)
• High-level inspection helpers in pya2l.api.inspect
• Creator API in pya2l.api.create for programmatic A2L generation
• Validator framework in pya2l.api.validate yielding structured diagnostics
• Export to A2L text or JSON format with complete roundtrip fidelity
• Optional CLI (a2ldb-imex) for import/export tasks
• Concurrent access via SQLite WAL mode (multiple readers during export)

Learn more about the standard at the ASAM website: https://www.asam.net/standards/detail/mcd-2-mc/wiki/

Installation

• Via pip: shell $ pip install pya2ldb IMPORTANT: Package-name is pya2ldb NOT pya2l!!!
• From Github:
  • Clone / fork / download pyA2Ldb repository.

Getting Started (Quickstart)

Parse once, work from SQLite thereafter.

Import a .a2l file and persist it as .a2ldb (SQLite):

from pya2l import import_a2l

session = import_a2l(
    "ASAP2_Demo_V161.a2l",
    # Optional:
    # encoding="utf-8",        # default is latin-1
    # progress_bar=False,      # silence the progress meter
    # loglevel="ERROR",        # also suppresses progress
    # force_overwrite=True,    # overwrite existing .a2ldb without prompting
)
# Creates ASAP2_Demo_V161.a2ldb in the working directory

• Open an existing .a2ldb without re-parsing:

from pya2l import open_existing

session = open_existing("ASAP2_Demo_V161")  # extension .a2ldb is implied

Or use a context manager so resources are closed deterministically:

from pya2l import open_existing
import pya2l.model as model

with open_existing("ASAP2_Demo_V161") as session:
    measurements = session.query(model.Measurement).limit(5).all()

Query with SQLAlchemy ORM - List all measurements ordered by name with address and data type:

from pya2l import open_existing
import pya2l.model as model

session = open_existing("ASAP2_Demo_V161")
measurements = (
    session.query(model.Measurement)
    .order_by(model.Measurement.name)
    .all()
)
for m in measurements:
    print(f"{m.name:48} {m.datatype:12} 0x{m.ecu_address.address:08x}")

When not using with ... as ..., call session.close() when finished so SQLite WAL data is checkpointed and file handles are released.

High-level inspection helpers - Use convenience wrappers from pya2l.api.inspect to access derived info:

from pya2l import open_existing
from pya2l.api.inspect import Characteristic, Measurement, AxisDescr

session = open_existing("ASAP2_Demo_V161")
ch = Characteristic(session, "ASAM.C.MAP.UBYTE.IDENTICAL")
print("shape:", ch.dim().shape)
print("element size:", ch.fnc_element_size(), "bytes")
print("num axes:", ch.num_axes())

me = Measurement(session, "ASAM.M.SCALAR.UBYTE.IDENTICAL")
print("is virtual:", me.is_virtual())

axis = ch.axisDescription("X")
print("axis info:", axis.axisDescription("X"))

Create A2L content programmatically - Use the Creator API to build or augment A2L databases:

from pya2l import open_create
from pya2l.api.create import (
    ProjectCreator, ModuleCreator, MeasurementCreator,
    CharacteristicCreator, CompuMethodCreator
)

session = open_create("MyProject.a2ldb")

# Create project and module
pc = ProjectCreator(session)
project = pc.create_project("MyProject", "Demo ECU Project")

mc = ModuleCreator(session)
module = mc.create_module("MyModule", "Demo Module", project=project)

# Add a conversion method
cmc = CompuMethodCreator(session)
cm = cmc.create_compu_method(
    "CM_Voltage", "Voltage conversion", "LINEAR",
    "%6.3", "V", module_name="MyModule"
)
cmc.add_coeffs_linear(cm, offset=0.0, factor=0.01)  # y = 0.01*x + 0

# Add a measurement
mec = MeasurementCreator(session)
meas = mec.create_measurement(
    "BatteryVoltage", "Battery voltage in Volts",
    "UWORD", "CM_Voltage", resolution=1, accuracy=0.1,
    lower_limit=0.0, upper_limit=20.0,
    module_name="MyModule"
)
mec.add_ecu_address(meas, 0x1000)

# Add a characteristic (calibration parameter)
cc = CharacteristicCreator(session)
char = cc.create_characteristic(
    "InjectionMap", "Fuel injection map",
    "MAP", 0x2000, "RL_InjMap", 0.0, "CM_Voltage",
    0.0, 100.0, module_name="MyModule"
)

mc.commit()
db.close()

Validate your database

from pya2l import open_existing
from pya2l.api.validate import Validator

session = open_existing("ASAP2_Demo_V161")
vd = Validator(session)
for msg in vd():  # iterate diagnostics
    # msg has fields: type (Level), category (Category), diag_code (Diagnostics), text (str)
    print(msg.type.name, msg.category.name, msg.diag_code.name, "-", msg.text)

Export back to A2L or JSON

from pya2l import export_a2l

# Export to A2L text format
export_a2l("ASAP2_Demo_V161", "exported.a2l")

# Or export to JSON for further processing
from pya2l.imex.json_exporter import export_json
export_json("ASAP2_Demo_V161.a2ldb", "exported.json")

Tips

• Default import encoding is latin-1; override encoding= if your file differs.
• Silence the progress meter via progress_bar=False or loglevel="ERROR".
• Python package name is pya2ldb (not pya2l).
• See howto for Excel export and other short recipes.

Examples

• See pya2l/examples for sample A2L files and scripts.
• The Sphinx docs contain a fuller tutorial and HOWTO guides.

Create API and coverage parity

pyA2L offers a Creator API in pya2l.api.create to programmatically build or augment A2L content. The project’s goal is coverage parity: everything you can query via pya2l.api.inspect is intended to be creatable via pya2l.api.create.

Example: creating common entities

from pya2l import open_create
from pya2l.api.create import ModuleCreator
from pya2l.api.inspect import Module

# Open or create a database
session = open_create("MyProject.a2l")  # or .a2ldb

mc = ModuleCreator(session)
# Create a module
mod = mc.create_module("DEMO", "Demo ECU module")

# Units and conversions
temp_unit = mc.add_unit(mod, name="degC", long_identifier="Celsius",
                        display="°C", type_str="TEMPERATURE")
ct = mc.add_compu_tab(mod, name="TAB_NOINTP_DEMO", long_identifier="Demo Tab",
                      conversion_type="TAB_NOINTP",
                      pairs=[(0, 0.0), (100, 1.0)], default_numeric=0.0)

# Frames and transformers
fr = mc.add_frame(mod, name="FRAME1", long_identifier="Demo frame",
                  scaling_unit=1, rate=10, measurements=["ENGINE_SPEED"])
tr = mc.add_transformer(mod, name="TR1", version="1.0",
                        executable32="tr32.dll", executable64="tr64.dll",
                        timeout=1000, trigger="ON_CHANGE", inverse="NONE",
                        in_objects=["ENGINE_SPEED"], out_objects=["SPEED_PHYS"])

# Typedefs and instances
ts = mc.add_typedef_structure(mod, name="TSig", long_identifier="Signal",
                              size=8)
mc.add_structure_component(ts, name="raw", typedefName="UWORD", addressOffset=0)
inst = mc.add_instance(mod, name="S1", long_identifier="Inst of TSig",
                       type_name="TSig", address=0x1000)

# Verify with inspect helpers
mi = Module(session)
print("#frames:", len(list(mi.frame.query())))
print("#compu tabs:", len(list(mi.compu_tab.query())))

See pya2l/examples/create_quickstart.py for a more complete example.

Command-line usage

A small CLI is provided as a console script named a2ldb-imex:

# Show version (exit 0)
$ a2ldb-imex -V

# Import an A2L (creates .a2ldb next to the input or in CWD with -L)
$ a2ldb-imex -i path/to/file.a2l

# Import with explicit encoding and create DB in current directory
$ a2ldb-imex -i path/to/file.a2l -E latin-1 -L

# Overwrite existing .a2ldb without prompting
$ a2ldb-imex -i path/to/file.a2l -f

# Suppress all output (useful in scripts/CI)
$ a2ldb-imex -i path/to/file.a2l -q

# Export an .a2ldb back to A2L text (stdout by default or -o file)
$ a2ldb-imex -e path/to/file.a2ldb -o exported.a2l

Compatibility

• Python: 3.10 – 3.14
• Platforms: Prebuilt wheels are published for selected platforms. From source, Windows/macOS are supported; Linux may require building native extensions.

Project links

• Source code: https://github.com/christoph2/pyA2L
• Issue tracker: https://github.com/christoph2/pyA2L/issues
• PyPI: https://pypi.org/project/pya2ldb/
• Documentation
• PDF Manual

Contributing

Contributions are welcome! Please open an issue to discuss significant changes before submitting a PR. See the existing tests under pya2l/tests and examples under pya2l/examples to get started. Contributors should use pre-commit to run formatting and lint checks before committing; see https://pre-commit.com/ for installation and usage.

Code of Conduct

This project follows a Code of Conduct to foster an open and welcoming community. Please read and abide by it when interacting in issues, discussions, and pull requests.

See CODE_OF_CONDUCT for full details.

Changelog / Release notes

See GitHub Releases: https://github.com/christoph2/pyA2L/releases