[marandaneto]

💡 Motivation and Context

Python's regex engine can return a different result from the flags service even when a pattern compiles. Python matches ^foo$ against "foo\n", while the service does not. Python also treats ^[a&&b]$ differently and rejects Unicode property escapes that the service supports. Definitive local results prevent the existing remote fallback from running.

This adapts the agreed approach in posthog-js#5354 and its cross-SDK discussion to Python re. It does not depend on that PR being merged or released.

• Keep compatible patterns local, including supported Unicode letters, decimal digits, whitespace, scalar literals and dot matching.
• Make unescaped end anchors strict. Preserve escaped dollar signs and dollar signs inside character classes.
• Raise the existing inconclusive result for unsupported syntax and reviewed word-boundary budget cases. Callers fall back to the server when allowed. Local-only evaluation stays unresolved.
• Keep invalid patterns within the supported subset false for both regex and not_regex. Do not invert unsupported patterns or failures into matches, including through cohort negation.

Review feedback updates

Tightening $ can turn a fast successful native search into expensive rejection. The reviewed ^(a+)+$ case with 27 a characters and a final newline took about 2.86 seconds before these updates. Quantified groups containing another quantifier now defer before native matching.

A related ^(a|aa)+$ case exceeded an owned four-second subprocess limit with 50 a characters and a final newline. Before native search, patterns using an unescaped outside-class $ plus quantification or alternation now defer when the input ends in LF. Compilation still happens first, preserving invalid-pattern false/false. Ordinary bounded and quantified alternation controls remain local, as does simple ^foo$ rejection on a final newline.

ASCII inputs now use built-in checks instead of Python-level surrogate and word-class loops. Built-in string membership still catches Python-only ASCII whitespace controls. Unicode safety scans and the empty-pattern early return remain intact.

Pattern classification, anchor translation and compilation now use a private 256-entry standard-library LRU cache. Entries contain immutable pattern metadata or the known-invalid marker, never property values or match results. Unsupported or transient compilation exceptions are not retained. Every per-value LF, Unicode, surrogate and whitespace check still runs outside the cache.

This classifier is deliberately conservative. Some patterns and final-newline inputs now require server evaluation even when a particular local search would be safe. This is not a complete ReDoS detector or a guarantee of equivalence across all Unicode versions, compiler limits or resource budgets. No new engine, timeout architecture, dependency or public option is added.

💚 How did you test it?

Regression tests preceded production edits. The original port had 104 failures before its fix. Review-feedback tests produced 26 missing-deferral/fallback failures before the nested-quantifier change, then 20 before the approved input-sensitive anchor policy. The same commands pass afterward. No wall-clock assertions or regex-engine mocks were added.

On the final signed code, CPython 3.13.13:

• uv run --no-sync pytest posthog/test/test_feature_flags_regex.py --timeout=30 -q: 333 passed.
• Feature flags, snapshots, utilities, tracking metadata and property-matching-version suites: 1,126 passed, plus 415 subtests. Existing deprecation warnings remain.
• Root Ruff lint/format, baseline mypy, public API snapshot, strict import and whitespace checks passed.
• Before this cache update, a warmed valid 1,000-alternative allowlist rejecting unlisted took a median 2.18ms per call. The same bounded workload on final committed code took about 0.20µs per call (five samples of 100 calls), with the same false result. This is a measured observation, not a timing assertion.
• Deterministic tests verify metadata reuse, bounded LRU eviction, immutable metadata, cached invalid results, fresh uncached exceptions and the same pattern across differing safe/unsafe values. The 14 new per-value controls passed before production edits as preserved baseline behavior. No artificial failing timing test or missing-helper setup failure is claimed as RED.
• The two long reviewed cases now become inconclusive quickly in an owned, bounded subprocess for both operators. No unsafe child was left running.
• A bounded 100KB ASCII ^z$ benchmark improved from a five-sample median of about 2.42ms per call to about 1.10µs, with the same false result. These are local observations, not throughput guarantees or unit-test thresholds.
• New targeted controls were compared with the actual frozen service. The earlier accepted candidate also had zero wrong definitive results against its frozen 4,868-vector corpus. Unchanged baseline corpora were not needlessly replayed, and inconclusive is not counted as a local match.

The required isolated review on the final signed HEAD against origin/main reported no actionable findings. An earlier class-union claim was disproved using actual service results: [a||b] matches a pipe character in the service as well as the committed SDK. The final review was required after integrating new main signing documentation. Earlier review feedback and backtracking safeguards are preserved.

The full repository suite, other Python runtimes locally, OpenFeature provider, application integration, mirror/release builds and live-service tests were not run. GitHub CI is not claimed green by these local checks.

📝 Checklist

• All commits are signed (see Commit signing).

• I reviewed the submitted code.

• I added tests to verify the changes.

• I updated the docs if needed. No new public configuration requires documentation.

• No breaking change or entry added to the changelog.

If releasing new changes

• Ran sampo add to generate a changeset file.

The existing .sampo/changesets/local-regex-compatibility.md patch metadata is retained. The generator was not rerun and no duplicate changeset was added.

🤖 Agent context

Autonomy: Human-driven (agent-assisted)

Pi assisted with implementation, test auditing and publication using Git, GitHub CLI, pytest, Ruff, mypy and isolated autoreview. Bounded subprocess probes and actual service results were used to verify performance concerns and review claims. The existing inconclusive/fallback path was retained instead of adding an engine or public API. Human review is required before merge.

[github-actions]

posthog-python Compliance Report

Date: 2026-10-23T10:00:01.483498+00:00
Duration: 259899ms

✅ All Tests Passed!

121/121 tests passed

---

Capture_V1 Tests

✅ 95/95 tests passed

View Details

Test	Status	Duration
Endpoint And Method.Targets V1 Endpoint	✅	517ms
Endpoint And Method.Does Not Use Legacy Endpoints	✅	512ms
Required Headers.Has Authorization Bearer Header	✅	511ms
Required Headers.Has Content Type Json	✅	510ms
Required Headers.Has Posthog Sdk Info Format	✅	510ms
Required Headers.Has Posthog Attempt Header	✅	511ms
Required Headers.Has Posthog Request Id	✅	510ms
Required Headers.Has Posthog Request Timestamp	✅	510ms
Required Headers.Has User Agent	✅	510ms
Body Format.Body Has Created At And Batch	✅	510ms
Body Format.No Api Key In Body	✅	510ms
Body Format.No Sent At In Body	✅	510ms
Event Format.Event Has Required Root Fields	✅	510ms
Event Format.Event Uuid Is Valid	✅	511ms
Event Format.Event Timestamp Is Rfc3339	✅	510ms
Event Format.Non Utc Event Timestamp Is Converted To Utc	✅	522ms
Event Format.Distinct Id Is String	✅	510ms
Event Format.Distinct Id At Root Not Properties	✅	510ms
Event Format.Custom Properties Preserved	✅	509ms
Event Format.Set Properties Preserved	✅	510ms
Event Format.Set Once Properties Preserved	✅	510ms
Event Format.Groups Properties Preserved	✅	510ms
Event Format.Sdk Generates Uuid If Not Provided	✅	510ms
Event Format.Event Has Required Root Fields Batch	✅	514ms
Event Format.Event Uuid Is Valid Batch	✅	514ms
Event Format.Event Timestamp Is Rfc3339 Batch	✅	513ms
Event Format.Distinct Id Is String Batch	✅	514ms
Event Format.Distinct Id At Root Not Properties Batch	✅	514ms
Event Format.Custom Properties Preserved Batch	✅	513ms
Event Format.Set Properties Preserved Batch	✅	514ms
Event Format.Set Once Properties Preserved Batch	✅	514ms
Event Format.Groups Properties Preserved Batch	✅	513ms
Event Format.Sdk Generates Uuid If Not Provided Batch	✅	514ms
Batch Behavior.Multiple Events In Single Batch	✅	518ms
Batch Behavior.Batch Envelope Smoke	✅	515ms
Batch Behavior.Flush With No Events Sends Nothing	✅	507ms
Batch Behavior.Flush At Triggers Batch	✅	1011ms
Batch Behavior.Created At Reflects Batch Creation Time	✅	510ms
Deduplication.Generates Unique Uuids	✅	518ms
Deduplication.Different Events Same Content Different Uuids	✅	512ms
Deduplication.Preserves Uuid On Retry	✅	6519ms
Deduplication.Preserves Timestamp On Retry	✅	6521ms
Deduplication.Preserves Uuid And Timestamp On Batch Retry	✅	6519ms
Deduplication.No Duplicate Events In Batch	✅	518ms
Header Behavior On Retry.Attempt Header Starts At One	✅	510ms
Header Behavior On Retry.Attempt Header Increments On Retry	✅	13528ms
Header Behavior On Retry.Request Id Preserved On Retry	✅	6520ms
Header Behavior On Retry.Different Requests Have Different Request Ids	✅	3020ms
Header Behavior On Retry.Request Timestamp Changes On Retry	✅	6520ms
Response Format Validation.Success Response Has Uuid Keyed Results	✅	511ms
Response Format Validation.Success Response Has Ok For Each Event	✅	515ms
Response Format Validation.Success No Retry After When All Ok	✅	512ms
Response Format Validation.Success Retry After Present When Retry Events	✅	1516ms
Response Format Validation.Success No Retry After When Drop Only	✅	512ms
Response Format Validation.Response Echoes Request Id	✅	511ms
Retry Behavior.Retries On 408	✅	6519ms
Retry Behavior.Retries On 500	✅	6520ms
Retry Behavior.Retries On 503	✅	8522ms
Retry Behavior.Retries On 504	✅	6519ms
Retry Behavior.Retryable Errors Have Retry After	✅	3518ms
Retry Behavior.Respects Retry After On Retryable Error	✅	11523ms
Retry Behavior.Does Not Retry On 400	✅	2513ms
Retry Behavior.Does Not Retry On 401	✅	2513ms
Retry Behavior.Does Not Retry On 402	✅	2513ms
Retry Behavior.Does Not Retry On 413	✅	2513ms
Retry Behavior.Does Not Retry On 415	✅	2513ms
Retry Behavior.Non Retryable Errors Have No Retry After	✅	2514ms
Retry Behavior.Implements Backoff	✅	22525ms
Retry Behavior.Max Retries Respected	✅	22536ms
Partial Batch Handling.Handles 200 Full Success	✅	2514ms
Partial Batch Handling.Handles 200 With All Ok	✅	3514ms
Partial Batch Handling.Does Not Retry Dropped Events	✅	3517ms
Partial Batch Handling.Does Not Retry Limited Events	✅	3516ms
Partial Batch Handling.Prunes Ok Events On Partial Retry	✅	6521ms
Partial Batch Handling.Prunes Dropped Events On Partial Retry	✅	6521ms
Partial Batch Handling.Retries Only Retry Events From Partial	✅	6523ms
Partial Batch Handling.Partial Retry Preserves Uuids	✅	6522ms
Partial Batch Handling.Partial Retry Attempt Header Increments	✅	6519ms
Partial Batch Handling.Partial Retry Request Id Preserved	✅	6517ms
Partial Batch Handling.Respects Retry After On Partial	✅	8521ms
Partial Batch Handling.Unknown Result Treated As Terminal	✅	3514ms
Partial Batch Handling.Mixed Ok Drop Limited No Retry	✅	3520ms
Compression.Sends Gzip Content Encoding	✅	511ms
Compression.No Content Encoding When Disabled	✅	510ms
Compression.Compressed Body Is Decompressible	✅	510ms
Error Handling.Does Not Retry On Unknown 4Xx	✅	2513ms
Event Options.Cookieless Mode Override	✅	510ms
Event Options.Disable Skew Correction Override	✅	511ms
Event Options.Process Person Profile Override	✅	510ms
Event Options.Product Tour Id Override	✅	510ms
Event Options.Unset Options Omitted	✅	510ms
Event Options.Options Override In Batch	✅	514ms
Geoip And Historical Migration.Geoip Disable Injected Into Properties	✅	510ms
Geoip And Historical Migration.Historical Migration Set In Body	✅	510ms
Geoip And Historical Migration.Historical Migration Absent By Default	✅	510ms

Capture_Ai Tests

✅ 5/5 tests passed

View Details

Test	Status	Duration
Routing.Capture Ai Posts To Ai Endpoint	✅	511ms
Routing.Capture Does Not Reroute Ai Named Events	✅	510ms
Identity.Capture Ai Event Has Uuid	✅	510ms
Identity.Capture Ai Keeps Supplied Uuid	✅	511ms
Timestamp Format.Non Utc Event Timestamp Is Converted To Utc	✅	510ms

Feature_Flags Tests

✅ 17/17 tests passed

View Details

Test	Status	Duration
Request Payload.Request With Person Properties Device Id	✅	11ms
Request Payload.Flags Request Uses V2 Query Param	✅	9ms
Request Payload.Flags Request Hits Flags Path Not Decide	✅	10ms
Request Payload.Flags Request Omits Authorization Header	✅	10ms
Request Payload.Token In Flags Body Matches Init	✅	9ms
Request Payload.Groups Round Trip	✅	10ms
Request Payload.Groups Default To Empty Object	✅	9ms
Request Payload.Disable Geoip False Propagates As Geoip Disable False	✅	9ms
Request Payload.Disable Geoip Omitted Defaults To False	✅	10ms
Request Payload.Flag Keys To Evaluate Contains Only Requested Key	✅	10ms
Request Lifecycle.No Flags Request On Init Alone	✅	3ms
Request Lifecycle.No Flags Request On Normal Capture	✅	510ms
Request Lifecycle.Two Flag Calls Produce Two Remote Requests	✅	14ms
Request Lifecycle.Mock Response Value Is Returned To Caller	✅	10ms
Retry Behavior.Retries Flags On 502	✅	313ms
Retry Behavior.Retries Flags On 504	✅	313ms
Side Effect Events.Get Feature Flag Captures Feature Flag Called Event	✅	512ms

Feature_Flags_Local_Evaluation Tests

✅ 4/4 tests passed

View Details

Test	Status	Duration
Versioned Boolean Matching.Matching Version Missing	✅	70ms
Versioned Boolean Matching.Matching Version 1	✅	68ms
Versioned Boolean Matching.Matching Version 2	✅	71ms
Versioned Boolean Matching.Version Only Reload 1 2 1 2 Missing	✅	35ms

[posthog]

PostHog Review · Standard · reviewed 7915272: 2 Should fix.

This is the only automatic review while this pull request is a draft. Comment @posthog review to review new changes sooner. Otherwise, a review of the new changes runs when you mark it ready for review.

[posthog]

PostHog Review · Standard · reviewed 7f99653: 1 Should fix.

This Standard review started from the posthog-review label. Comment @posthog review --deep for a Deep review, or @posthog help to see all commands.