feat(capture): add capture_mode config scaffolding (capture v1, 1/6)

[eli-r-ph]

💡 Motivation and Context

First PR in a stacked series adding Capture V1 (POST /i/v1/analytics/events) support to the SDK, at parity with posthog-go's CaptureMode and posthog-rs's capture-v1.

This PR is pure, inert scaffolding: it adds the capture-mode selector and threads it everywhere it will be needed, but changes no runtime behavior — V0 (legacy /batch/) still runs on every path. Later PRs add the v1 serializer, transport, wiring, harness suite, and docs.

• New CaptureMode enum: V0 (legacy /batch/, the default) and V1 (/i/v1/analytics/events).
• resolve_capture_mode() with precedence explicit kwarg > POSTHOG_CAPTURE_MODE env > V0. Env accepts v0/legacy and v1/analytics_v1 (mirrors posthog-go's vocabulary); an unrecognized env value warns and defaults to V0 so a typo never silently flips the wire protocol. An invalid explicit kwarg raises ValueError (explicit misconfig should be loud).
• Plumbed through Client.__init__ (resolved + stored as self.capture_mode), both Consumer(...) constructions (initial + fork-reinit), and the module-level default client. Consumer stores it but does not yet act on it.

💚 How did you test it?

• New posthog/test/test_capture_mode.py (25 cases): resolution precedence, env aliases/case/whitespace, blank-env default, unrecognized-env warn+default, invalid-kwarg raises, and plumbing to Client.capture_mode + per-Consumer propagation (incl. multi-thread).
• Regression: ran test_consumer, test_client, test_module, test_client_fork, test_dedicated_ai_endpoint + the new file — 213 passed.
• ruff format/ruff check clean; mypy clean on new modules; regenerated references/public_api_snapshot.txt.

📝 Checklist

• I reviewed the submitted code.
• I added tests to verify the changes.
• I updated the docs if needed.
• No breaking change (additive, default V0 keeps upgrades transparent).

🤖 Agent context

Autonomy: Human-driven (agent-assisted)

Authored with Cursor (Claude Opus 4.8) following a pre-agreed, hardened implementation plan. Design decisions: enum members are Pythonic V0/V1 (not posthog-go's Legacy/AnalyticsV1) but the env var accepts both spellings for cross-SDK consistency; the env typo path defaults defensively while an explicit bad kwarg fails fast. Mode is fully inert here by design to keep this PR a no-op for existing users and easy to review. Changeset (sampo add) and docs are intentionally deferred to the final PR so the feature releases as one entry.

[greptile-apps]

_{Reviews (1): Last reviewed commit: "feat(capture): add capture_mode config s..." | Re-trigger Greptile}

[greptile-apps]

In test_explicit_kwarg_takes_precedence_and_coerces, the env is set to "v1" for every sub-case. For the four cases where the kwarg also resolves to V1 (enum_v1, str_v1, str_analytics_v1_alias, str_upper_and_padded), the result is V1 regardless of whether the kwarg or the env won — so these cases never actually exercise the precedence claim. Setting the env to "v0" for those cases would give the test its full demonstrative value.

Suggested change

	@parameterized.expand(
	[
	("enum_v0", CaptureMode.V0, CaptureMode.V0),
	("enum_v1", CaptureMode.V1, CaptureMode.V1),
	("str_v0", "v0", CaptureMode.V0),
	("str_v1", "v1", CaptureMode.V1),
	("str_legacy_alias", "legacy", CaptureMode.V0),
	("str_analytics_v1_alias", "analytics_v1", CaptureMode.V1),
	("str_upper_and_padded", " V1 ", CaptureMode.V1),
	]
	)
	def test_explicit_kwarg_takes_precedence_and_coerces(
	self, _name, kwarg, expected
	) -> None:
	# Env is set to the opposite mode to prove the kwarg wins.
	with mock.patch.dict(os.environ, {CAPTURE_MODE_ENV_VAR: "v1"}):
	self.assertIs(resolve_capture_mode(kwarg), expected)
	@parameterized.expand(
	[
	# env=v1, kwarg resolves to V0: kwarg must win
	("enum_v0", CaptureMode.V0, CaptureMode.V0, "v1"),
	# env=v0, kwarg resolves to V1: kwarg must win
	("enum_v1", CaptureMode.V1, CaptureMode.V1, "v0"),
	("str_v0", "v0", CaptureMode.V0, "v1"),
	("str_v1", "v1", CaptureMode.V1, "v0"),
	("str_legacy_alias", "legacy", CaptureMode.V0, "v1"),
	("str_analytics_v1_alias", "analytics_v1", CaptureMode.V1, "v0"),
	("str_upper_and_padded", " V1 ", CaptureMode.V1, "v0"),
	]
	)
	def test_explicit_kwarg_takes_precedence_and_coerces(
	self, _name, kwarg, expected, env_value
	) -> None:
	# Env is set to the opposite mode to prove the kwarg wins.
	with mock.patch.dict(os.environ, {CAPTURE_MODE_ENV_VAR: env_value}):
	self.assertIs(resolve_capture_mode(kwarg), expected)

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!

[github-actions]

posthog-python Compliance Report

Date: 2026-06-27 19:33:51 UTC
Duration: 530130ms

✅ All Tests Passed!

45/45 tests passed

---

Capture Tests

✅ 29/29 tests passed

View Details

Test	Status	Duration
Format Validation.Event Has Required Fields	✅	517ms
Format Validation.Event Has Uuid	✅	10007ms
Format Validation.Event Has Lib Properties	✅	10007ms
Format Validation.Distinct Id Is String	✅	10007ms
Format Validation.Token Is Present	✅	10007ms
Format Validation.Custom Properties Preserved	✅	10007ms
Format Validation.Event Has Timestamp	✅	10007ms
Retry Behavior.Retries On 503	✅	18018ms
Retry Behavior.Does Not Retry On 400	✅	12005ms
Retry Behavior.Does Not Retry On 401	✅	10006ms
Retry Behavior.Respects Retry After Header	✅	16014ms
Retry Behavior.Implements Backoff	✅	30017ms
Retry Behavior.Retries On 500	✅	13011ms
Retry Behavior.Retries On 502	✅	16010ms
Retry Behavior.Retries On 504	✅	16011ms
Retry Behavior.Max Retries Respected	✅	30018ms
Deduplication.Generates Unique Uuids	✅	7002ms
Deduplication.Preserves Uuid On Retry	✅	16016ms
Deduplication.Preserves Uuid And Timestamp On Retry	✅	23019ms
Deduplication.Preserves Uuid And Timestamp On Batch Retry	✅	16006ms
Deduplication.No Duplicate Events In Batch	✅	10002ms
Deduplication.Different Events Have Different Uuids	✅	10007ms
Compression.Sends Gzip When Enabled	✅	10007ms
Batch Format.Uses Proper Batch Structure	✅	10007ms
Batch Format.Flush With No Events Sends Nothing	✅	5005ms
Batch Format.Multiple Events Batched Together	✅	10005ms
Error Handling.Does Not Retry On 403	✅	12008ms
Error Handling.Does Not Retry On 413	✅	10007ms
Error Handling.Retries On 408	✅	14013ms

Feature_Flags Tests

✅ 16/16 tests passed

View Details

Test	Status	Duration
Request Payload.Request With Person Properties Device Id	✅	9502ms
Request Payload.Flags Request Uses V2 Query Param	✅	10006ms
Request Payload.Flags Request Hits Flags Path Not Decide	✅	10007ms
Request Payload.Flags Request Omits Authorization Header	✅	10007ms
Request Payload.Token In Flags Body Matches Init	✅	10007ms
Request Payload.Groups Round Trip	✅	10007ms
Request Payload.Groups Default To Empty Object	✅	10007ms
Request Payload.Person Properties Distinct Id Auto Populated When Caller Omits It	✅	10007ms
Request Payload.Disable Geoip False Propagates As Geoip Disable False	✅	10006ms
Request Payload.Disable Geoip Omitted Defaults To False	✅	10007ms
Request Payload.Flag Keys To Evaluate Contains Only Requested Key	✅	10014ms
Request Lifecycle.No Flags Request On Init Alone	✅	5003ms
Request Lifecycle.No Flags Request On Normal Capture	✅	10508ms
Request Lifecycle.Two Flag Calls Produce Two Remote Requests	✅	9510ms
Request Lifecycle.Mock Response Value Is Returned To Caller	✅	10003ms
Side Effect Events.Get Feature Flag Captures Feature Flag Called Event	✅	10509ms