Repository navigation
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #189 +/- ##
==========================================
+ Coverage 78.37% 80.70% +2.32%
==========================================
Files 9 9
Lines 1369 1394 +25
==========================================
+ Hits 1073 1125 +52
+ Misses 296 269 -27
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
…drops Coverage analysis of the full CT run showed the uncovered lines that guard real behaviour clustered in four places, now tested: - check_opts_values validates every option enum value no HTTP-level test exercises (ntlm/negotiate auth, the http_version and sslversion values, verbose, explicit stream/stream_window defaults, string sslcert/sslkey, reply_to) plus the check_opts error return, taking katipo_req to full line coverage. - spillover_success and spillover_admission_timeout pin the admission pick (meck on wpool_pool:random_worker) to deterministically drive a request through spillover to the second worker, and through the bounded admission call timing out against a suspended candidate. - port_late_progress injects fabricated port frames for an unregistered request -- a chunk and a terminal error -- proving both are dropped silently and the worker stays healthy; the real race window is too narrow to hit reliably. streaming_cancel_during_flood exercises the same invariant end-to-end under a chunk flood. - pipelining_http1 covers the last accepted pool option value. Also remove katipo_worker:start_link/1: wpool drives the worker callbacks through wpool_process, so nothing ever called it. Its is_list guard moves to init/1.
The 500ms request timer raced admission-to-headers latency: in the loaded parallel async group the timer could fire before the headers arrived, delivering katipo_error instead -- which the selective receive in expect_stream_headers ignored, so the test sat out its full 5s and failed with an uninformative no_stream_headers. Seen once on the CI HTTP/3 job. Widen the budget to 2s (no wall-clock cost: parallel siblings already wait longer), and teach expect_stream_headers to fail immediately with the actual error when a terminal message beats the headers, so any future failure names its cause.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Coverage analysis of the full CT run showed the uncovered source lines that guard real behaviour clustered in four places, and this closes all of them.
check_opts_values validates every option enum value no HTTP-level test exercises -- ntlm/negotiate auth, the http_version and sslversion values, verbose, explicit stream/stream_window defaults, string sslcert/sslkey, reply_to -- plus the check_opts error return, taking katipo_req to full line coverage. Admission picks are random (wpool's fast_rand_uniform is unseedable), so spillover_success and spillover_admission_timeout pin the picked worker with meck on wpool_pool:random_worker to deterministically drive a request through spillover to the second worker, and through the bounded admission call timing out against a suspended candidate. port_late_progress injects fabricated port frames for an unregistered request -- a chunk and a terminal error -- proving both are dropped silently and the worker stays healthy; the real race window (port output in flight while the cancel lands) is too narrow to hit reliably, which is also why the late-terminal branch was previously covered only by timing luck. streaming_cancel_during_flood exercises the same invariant end-to-end under a chunk flood, and pipelining_http1 covers the last accepted pool option value.
Also removes katipo_worker:start_link/1: wpool drives the worker callbacks through wpool_process, so nothing ever called it. Its is_list guard moves to init/1.
The lines still uncovered after this are deliberate: the thirteen no-argument convenience wrappers, two microsecond race windows (await's late-response flush, request timer vs completion), and defensive catch-alls.
🤖 Generated with Claude Code
https://claude.ai/code/session_014cDA7Wx8Pw1xXJ73LnhqKq