feat: Add os_distro information to events

[parinporecha]

Bringing to parity with posthog-go (PostHog/posthog-go#167)

Context

Currently events carry $os and $os_version properties. While that's fine for Windows and Mac OS, it needs distro level information for Linux to make the system context more diagnostic.

Changes

• $os_distro property is now added to system_context() if os is Linux.

Tests

• Added this property in existing unit test.
• Cleaned up unused distro_info param in the same test.

Tests are successful locally.

Screenshots (created through example script)

• Linux
  

• Mac OS (property not added)
  

[greptile-apps]

Comments Outside Diff (1)

1. posthog/test/test_client.py, line 2234-2246 (link)

   New distro logic untested — mock bypasses the code under test

   The Linux test case mocks get_os_info entirely, which means the two new lines that actually call distro.name() (lines 498-500 in utils.py) are never exercised. Any regression in that logic would go undetected.

   A more effective approach is to keep the Linux branch in get_os_info() under test by mocking the distro module directly (the same pattern used for platform and sys), rather than short-circuiting the whole function:

   # Instead of patching get_os_info entirely, patch the distro module:
   with mock.patch("posthog.utils.distro") as mock_distro:
       mock_distro.info.return_value = {"version": expected_os_version}
       mock_distro.name.return_value = expected_os_distro or ""
       context = system_context()

   This ensures the new distro.name() call and the conditional guard are actually covered by the test.

   Prompt To Fix With AI

   This is a comment left during a code review.
   Path: posthog/test/test_client.py
   Line: 2234-2246
   
   Comment:
   **New distro logic untested — mock bypasses the code under test**
   
   The Linux test case mocks `get_os_info` entirely, which means the two new lines that actually call `distro.name()` (lines 498-500 in `utils.py`) are never exercised. Any regression in that logic would go undetected.
   
   A more effective approach is to keep the Linux branch in `get_os_info()` under test by mocking the `distro` module directly (the same pattern used for `platform` and `sys`), rather than short-circuiting the whole function:
   
   ```python
   # Instead of patching get_os_info entirely, patch the distro module:
   with mock.patch("posthog.utils.distro") as mock_distro:
       mock_distro.info.return_value = {"version": expected_os_version}
       mock_distro.name.return_value = expected_os_distro or ""
       context = system_context()
   ```
   
   This ensures the new `distro.name()` call and the conditional guard are actually covered by the test.
   
   How can I resolve this? If you propose a fix, please make it concise.

Prompt To Fix All With AI

This is a comment left during a code review.
Path: posthog/utils.py
Line: 516-517

Comment:
**Empty `$os_distro` sent when distro is unknown**

`os_distro` is initialized to `""` and only overwritten if `distro.name()` returns a truthy value. However, the condition that appends it to `info` does not guard against the empty-string case, so Linux systems where `distro.name()` returns `""` will emit `$os_distro: ""`. This is misleading — the property should only be present when there is a meaningful value.

```suggestion
    if os_name == "Linux" and os_distro:
        info["$os_distro"] = os_distro
```

How can I resolve this? If you propose a fix, please make it concise.

---

This is a comment left during a code review.
Path: posthog/test/test_client.py
Line: 2234-2246

Comment:
**New distro logic untested — mock bypasses the code under test**

The Linux test case mocks `get_os_info` entirely, which means the two new lines that actually call `distro.name()` (lines 498-500 in `utils.py`) are never exercised. Any regression in that logic would go undetected.

A more effective approach is to keep the Linux branch in `get_os_info()` under test by mocking the `distro` module directly (the same pattern used for `platform` and `sys`), rather than short-circuiting the whole function:

```python
# Instead of patching get_os_info entirely, patch the distro module:
with mock.patch("posthog.utils.distro") as mock_distro:
    mock_distro.info.return_value = {"version": expected_os_version}
    mock_distro.name.return_value = expected_os_distro or ""
    context = system_context()
```

This ensures the new `distro.name()` call and the conditional guard are actually covered by the test.

How can I resolve this? If you propose a fix, please make it concise.

---

This is a comment left during a code review.
Path: posthog/test/test_client.py
Line: 2238-2242

Comment:
**Mock always injects `$os_distro`, making a "Linux, no distro" case untestable**

The mock return value unconditionally includes `"$os_distro": expected_os_distro`. If a future test case covers Linux with an unknown/empty distro (`expected_os_distro=None` or `""`), `system_context()` will receive `$os_distro: None` (or `""`) from the mock and include it in its output, but `expected_context` won't have that key (because of the `if expected_os_distro:` guard). The assertion would then fail spuriously.

The mock should mirror the behaviour of the real function — only include `$os_distro` when it has a value:

```suggestion
                    with mock.patch(
                        "posthog.utils.get_os_info",
                        return_value={
                            "$os": expected_os,
                            "$os_version": expected_os_version,
                            **( {"$os_distro": expected_os_distro} if expected_os_distro else {} ),
                        },
                    ):
```

How can I resolve this? If you propose a fix, please make it concise.

_{Reviews (1): Last reviewed commit: "Add os_distro information to events" | Re-trigger Greptile}

[parinporecha]

@marandaneto can you maybe take a look?

[marandaneto]

left a suggestion otherwise lgtm

[parinporecha]

CI is having the same issue as posthog-go

[github-actions]

This PR hasn't seen activity in a week! Should it be merged, closed, or further worked on? If you want to keep it open, post a comment or remove the stale label – otherwise this will be closed in another week.