feat(error-tracking): emit stack frames in canonical bottom-up order

[cat-ph]

Problem

PostHog's error tracking expects a single canonical stack-frame order across every SDK so the ingestion pipeline can fingerprint, group, and display exceptions consistently. The canonical wire order for $exception_list[].stacktrace.frames is bottom-up: frames[0] is the outermost entry point and the last frame is the crash site.

The Elixir SDK's logger handler built frames directly from __STACKTRACE__, which is innermost-first (crash site first). This is the reverse of the canonical order, so Elixir exceptions were emitted upside-down relative to the other SDKs.

Part of cross-SDK stack-frame-order standardization: PostHog/sdk-specs#11

Change

• In lib/posthog/handler.ex, do_stacktrace/3 now reverses the frame list it builds from __STACKTRACE__, so the emitted frames are bottom-up (entry point first, crash site last).
• Arity-based frame formatting inside do_stacktrace/3 is untouched; only list order flips.
• Source-context fields (context_line / pre_context / post_context) stay attached to their own frames; only the order of the list changes.
• $exception_list ordering is unchanged (already canonical: [0] = outermost/reported exception).
• The synthetic single-frame logger path (the stacktrace/3 clause that emits one synthetic: true frame) is unaffected: a single-frame list is trivially canonical, and nothing in that path assumes crash-first order.

Tests

• Updated all order-dependent assertions in test/posthog/handler_test.exs to expect bottom-up frames. Head-matches that pinned the crash site as the first frame were converted to tail assertions (List.last/1 / Enum.take(frames, -3)).
• Added an explicit regression test: the first frame is the entry point (Task.Supervised.invoke_mfa/2) and the last frame is the crash site (anonymous fn/0 in LoggerHandlerKit.Act.task_error/1).
• mix test: 308 passed, 18 excluded. mix format --check-formatted: clean.

Coordination

This is a breaking wire-order change for the frames array. It should be merged/released only after the pipeline normalization gate (cymbal) is live, so that exceptions ingested during the transition are normalized to a single canonical order regardless of which SDK version produced them.

Released as a MINOR version bump so the pipeline can gate normalization on $lib_version.

Note: the pipeline currently has no dedicated Elixir stack handler; that gap is tracked in the spec PR above.

[greptile-apps]

_{Reviews (1): Last reviewed commit: "feat(error-tracking): emit stack frames ..." | Re-trigger Greptile}

[github-actions]

posthog-elixir Compliance Report

Date: 2026-07-23 18:03:54 UTC
Duration: 117477ms

✅ All Tests Passed!

46/46 tests passed

---

Capture Tests

✅ 29/29 tests passed

View Details

Test	Status	Duration
Format Validation.Event Has Required Fields	✅	610ms
Format Validation.Event Has Uuid	✅	608ms
Format Validation.Event Has Lib Properties	✅	608ms
Format Validation.Distinct Id Is String	✅	608ms
Format Validation.Token Is Present	✅	608ms
Format Validation.Custom Properties Preserved	✅	608ms
Format Validation.Event Has Timestamp	✅	608ms
Retry Behavior.Retries On 503	✅	5613ms
Retry Behavior.Does Not Retry On 400	✅	2610ms
Retry Behavior.Does Not Retry On 401	✅	2611ms
Retry Behavior.Respects Retry After Header	✅	5614ms
Retry Behavior.Implements Backoff	✅	15617ms
Retry Behavior.Retries On 500	✅	5614ms
Retry Behavior.Retries On 502	✅	5613ms
Retry Behavior.Retries On 504	✅	5614ms
Retry Behavior.Max Retries Respected	✅	15624ms
Deduplication.Generates Unique Uuids	✅	618ms
Deduplication.Preserves Uuid On Retry	✅	5613ms
Deduplication.Preserves Uuid And Timestamp On Retry	✅	10609ms
Deduplication.Preserves Uuid And Timestamp On Batch Retry	✅	5615ms
Deduplication.No Duplicate Events In Batch	✅	613ms
Deduplication.Different Events Have Different Uuids	✅	610ms
Compression.Sends Gzip When Enabled	✅	608ms
Batch Format.Uses Proper Batch Structure	✅	609ms
Batch Format.Flush With No Events Sends Nothing	✅	606ms
Batch Format.Multiple Events Batched Together	✅	612ms
Error Handling.Does Not Retry On 403	✅	2607ms
Error Handling.Does Not Retry On 413	✅	2611ms
Error Handling.Retries On 408	✅	5614ms

Feature_Flags Tests

✅ 17/17 tests passed

View Details

Test	Status	Duration
Request Payload.Request With Person Properties Device Id	✅	609ms
Request Payload.Flags Request Uses V2 Query Param	✅	608ms
Request Payload.Flags Request Hits Flags Path Not Decide	✅	608ms
Request Payload.Flags Request Omits Authorization Header	✅	608ms
Request Payload.Token In Flags Body Matches Init	✅	608ms
Request Payload.Groups Round Trip	✅	608ms
Request Payload.Groups Default To Empty Object	✅	608ms
Request Payload.Disable Geoip False Propagates As Geoip Disable False	✅	608ms
Request Payload.Disable Geoip Omitted Defaults To False	✅	608ms
Request Payload.Flag Keys To Evaluate Contains Only Requested Key	✅	608ms
Request Lifecycle.No Flags Request On Init Alone	✅	3ms
Request Lifecycle.No Flags Request On Normal Capture	✅	608ms
Request Lifecycle.Two Flag Calls Produce Two Remote Requests	✅	1213ms
Request Lifecycle.Mock Response Value Is Returned To Caller	✅	607ms
Retry Behavior.Retries Flags On 502	✅	910ms
Retry Behavior.Retries Flags On 504	✅	910ms
Side Effect Events.Get Feature Flag Captures Feature Flag Called Event	✅	1211ms

[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.

[github-actions]

This PR was closed due to lack of activity. Feel free to reopen if it's still relevant.

[greptile-apps]

_{Reviews (2): Last reviewed commit: "feat(error-tracking): emit stack frames ..." | Re-trigger Greptile}