Metadata-Version: 2.4
Name: ddtrace-internal
Version: 0.3.0
Summary: Stub internal Datadog Python package (ddtrace-internal).
Author-email: "Datadog, Inc." <dev@datadoghq.com>
License-Expression: BSD-3-Clause OR Apache-2.0
Project-URL: Repository, https://github.com/DataDog/datadog-apm-python
Requires-Python: >=3.9
Description-Content-Type: text/markdown

# ddtrace-internal

Pure-Python code shared by multiple products in dd-trace-py, extracted from
`ddtrace.internal`: the module registry and `codeowners` so far.

Pure Python by design -- this installs and imports with nothing but an
interpreter. The compiled components live in their own distribution, and the
two are peers: neither imports the other, and anything needing both depends on
both.

## Local install (editable)

```bash
python -m pip install -e .
```

## Running tests

```bash
scripts/run_tests.py                        # all supported Python versions
scripts/run_tests.py --python 3.12          # a single version
scripts/run_tests.py --list                 # show the matrix without running it
```

Each (variant, Python) pair in `suitespec.yml` runs as a cached `uv run --no-project`
environment with the package installed editable, so tests import the working tree; run it
from anywhere. In CI, `--wheel-path <dir>` installs the exact built wheel instead of the
working tree.

## Layout

- `src/_ddtrace_internal/` — package root
- `tests/<module_path>/test.py` — tests, mirroring the source layout (e.g. `src/_ddtrace_internal/foo.py` → `tests/foo/test.py`)
- `suitespec.yml` — test matrix definition
