Codemagic CLI Tools
Codemagic CLI Tools are a set of command-line utilities for managing Android and iOS app builds,
code signing, and deployment. The tools are used to power mobile app builds at codemagic.io but can be also used in other virtual environments or locally.
Installing
Codemagic CLI Tools are available on PyPI
and can be installed and updated using pip.
python -m pip install codemagic-cli-tools
The package requires Python version 3.8+.
Command line usage
Installing the package adds the following executables to your path:
Online documentation for all installed executables can be found under
docs
.
Alternatively, you could see the documentation by using --help
option from command line:
<command> --help
to see general description and subcommands for the tool, or
<command> <subcommand> --help
to get detailed information of the subcommand.
For example:
$ keychain create --help
usage: keychain create [-h] [--log-stream {stderr,stdout}] [--no-color] [--version] [-s] [-v] [-pw PASSWORD] [-p PATH]
Create a macOS keychain, add it to the search list
optional arguments:
-h, --help show this help message and exit
Optional arguments for command create:
-pw PASSWORD, --password PASSWORD
Keychain password. Alternatively to entering PASSWORD in plaintext, it may also be specified using the "@env:" prefix followed by an environment variable name, or the "@file:" prefix followed by a path to the file containing the value. Example: "@env:<variable>" uses the value in the environment variable named "<variable>", and "@file:<file_path>" uses the value from the file at "<file_path>". [Default: '']
Options:
--log-stream {stderr,stdout}
Log output stream. [Default: stderr]
--no-color Do not use ANSI colors to format terminal output
--version Show tool version and exit
-s, --silent Disable log output for commands
-v, --verbose Enable verbose logging for commands
Optional arguments for keychain:
-p PATH, --path PATH Keychain path. If not provided, the system default keychain will be used instead
Usage from Python
In addition to the command line interface, the package also provides a mirroring Python API.
All utilities that are available as CLI tools are accessible from Python in package
codemagic.tools
.
The CLI actions are instance methods that are decorated by the action
decorator. For example, you can use the Keychain
tool from Python source as follows:
In [1]: from pathlib import Path
In [2]: from codemagic.tools import Keychain
In [3]: Keychain().get_default()
Out[3]: PosixPath('/Users/priit/Library/Keychains/login.keychain-db')
In [4]: keychain = Keychain(Path("/tmp/new.keychain"))
In [5]: keychain.create()
Out[5]: PosixPath('/tmp/new.keychain')
In [6]: keychain.make_default()
In [7]: Keychain().get_default()
Out[7]: PosixPath('/private/tmp/new.keychain')
Development
This project uses Poetry to manage dependencies. Before starting development, please ensure that your
machine has Poetry available. Installation instructions can be found from their
docs.
Assuming you've already cloned the repository
itself, or a fork of it, you can get started by running
poetry install
This will install all required dependencies specified in the poetry.lock
file.
The source code of the project lives inside the src
directory and tests are
implemented in the tests
directory.
Code formatting and linting rules
Automatic code formatting is done with Black.
Invoke Black checks from repository root directory with
poetry run black --check .
Linting rules are enforced using Ruff.
Checks can be started from repository root with
poetry run ruff check .
Static type checks
A huge portion of the Python source code has type hints, and all public methods or functions
are expected to have type hints. Static type checks of the source code are performed using
Mypy from the repository root by running
poetry run mypy .
Running tests
Pytest is used as the framework. As mentioned above,
tests are stored in the tests
directory, separated from package source code. Test code layout
mirrors the structure of the codemagic
package in the source directory.
Tests can be started by running the following command from the repository root:
poetry run pytest
Note that tests that target App Store Connect API or
Google Play Developer API live endpoints
are skipped by default for obvious reasons. They can be enabled (either for TDD or other reasons)
by setting the environment variable RUN_LIVE_API_TESTS
to any non-empty value.
Note that for the tests to run successfully, you'd have to define the following environment variables:
-
For App Store Connect:
export TEST_APPLE_KEY_IDENTIFIER=... # Key ID
export TEST_APPLE_ISSUER_ID=... # Issued ID
And either of the two:
export TEST_APPLE_PRIVATE_KEY_PATH=...
export TEST_APPLE_PRIVATE_KEY_CONTENT=...
Those can be obtained from App Store Connect -> Users and Access -> Keys.
For more information follow Apple's official documentation about Creating API Keys for App Store Connect API.
-
For Google Play:
export TEST_GCLOUD_PACKAGE_NAME=... # Package name (Ex: com.google.example)'
And either of the two:
export TEST_GCLOUD_SERVICE_ACCOUNT_CREDENTIALS_PATH=... # Path to gcloud service account creedentials with `JSON` key type
export TEST_GCLOUD_SERVICE_ACCOUNT_CREDENTIALS_CONTENT=... # Content of gcloud service account creedentials with `JSON` key type
-
For Firebase:
Either of the two:
export TEST_FIREBASE_SERVICE_ACCOUNT_CREDENTIALS_PATH=... # Path to gcloud service account creedentials with `JSON` key type
export TEST_FIREBASE_SERVICE_ACCOUNT_CREDENTIALS_CONTENT=... # Content of gcloud service account creedentials with `JSON` key type
Pre-commit hooks
Optionally, the pre-commit framework can be used to ensure that
the source code updates are compliant with all the rules mentioned above.
Installation instructions are available in their docs.
The repository already contains pre-configured .pre-commit-config.yaml
, so to enable
the hooks, just run
pre-commit install