Repository navigation
Conversation
There was a problem hiding this comment.
Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.
An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing.
|
debt-tools at 9e60565, every finding: |
|
debt-src at 9e60565, every finding: |
trissim
force-pushed
the
batch/submission-core-nov1
branch
from
November 2, 2026 04:57
fb15048 to
156ba49
Compare
trissim
force-pushed
the
batch/submission-core-nov1
branch
from
November 2, 2026 05:10
156ba49 to
4b9e180
Compare
Contributor
Author
|
Closing as superseded. Three of this batch's commits are already applied on main. The other two match main's 5981823 and f6b61f3 (landed with #3848, merge e9ae799) apart from test-only wording, and main holds the newer versions. Generated by Claude Code |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Submission/session integration:18files,+221/-59
Integrates original Core#3848 (123967) over9c3f94b2 main (production038c plus two rule-only commits), retaining original provenance, fixture identity/validated launcher corrections, fa7d96's witnessed-plan reload construction,95fe3e82's canonical history/ACP destination separation and77d2cd15's nominal fixture producer. Current integrated head4b9e180e3019fd23c879a6c469781f1fe23c87eb. Paired Toad continuation is batch/submission-ui-nov1/04afb568b9364eb205a1c313af9a69448c1428ec (original#1389/ec73 plus9c3d999c/304e54e52 corrections). DO NOT merge/install this API half independently.
Before/after ownership
Before: project mutation discarded captured birth at ingress, and the live binding reconstructed a plan through a reverse import. After: ThreadManagement.set_project accepts ThreadIncarnation while preserving the atomic original-project guard; live binding dispatches to the witnessed SessionLoadPlan that constructs the reload. Both families survive. No compatibility overload, cast, decoder widening or local import.
Historical evidence and resolved identity defect
Actual combined Core run atf31d4798:84collected,82passed,2failed,154.30s,exit1. Corrections at06fb770 use nominal ACP identity internally, encode the foreign SDK argument, derive Message sender from the session-to-thread owner, and use the existing validated native fixture instead of /bin/echo.
Affected3-case rerun:2passed,1failed,21.00s,exit1. Remaining PRODUCT defect: session_lifecycle.py:248 passes AcpSessionId to TranscriptReplay.replay(session_id:str), which fills TranscriptPageRead.thread with the wrong identity. Observer dies during TranscriptReadIdentity.same_content/FieldCodec.project: Expected str, received AcpSessionId. Continuing author Errors owns the same3848 repair separating canonical history target from typed ACP destination, including both replay sites, subscription and consumers. No author pytest requested. Logs: /home/ts/.cache/agent-scratch/pi-commander/submission-nov1/{core-tests,core-fixture-tests}.log. Earlier failures are retained, not relabeled as green.
Current correction and executed selection
TranscriptReplay.replay/subscribe now take canonical thread separately from nominal AcpSessionId destination. SDK publication encodes at its foreign boundary; the strict decoder is unchanged. Historical integration156ba49f2:8affected cases executed,7passed/1failed37.98s,rc1. All replay/subscription/rename/project cases passed; the pre-replay fixture passed raw id=native_id[:8] into MessageEntry.id. That exact failure remains in core156-tests.log/.rc.
77d2cd1 corrects the producer with existing FreshNativeIdText.decode, preserving spelling. Composed4b9e180e: the entire affected test_transcript_input_display.py collected8cases and passed8/8 in6.25s/rc0. Kernel TestLease, INT300, explicit src/tools/tests/dependencies, installed038c interpreter and existing durableTMPDIR; core4b9-tests.log/.rc. This is a different eight-case selection, not a relabeling or rerun of the original156 selection.
Toad916's three-path selection finished20passed/2failed286.56s/rc1. The callable fixture correction is composed at04afb568b; its two shared-close variants then failed2/2 in18.08s/rc1 because the scripted peer advertises/implements unstable session/close while its base run_agent launch leaves that SDK feature disabled. Errors owns that exact existing peer-launch correction on1389; no product fallback or weaker assertions. Neither API half is installed.
Delivery boundary
This PR documents real integrated work; source/import checks are not app/native acceptance. Finish the exact identity repair and affected consumer qualification, integrate/pin the paired merged Toad in this PR, then merge/install together and verify the actual workflow. Current production receipts/scroll/export/native pins remain intact. No3850 runtime overhaul,3829 retention work, uncertain input replay or authority relaxation is included. Assumed: physical phone/performance and provider acceptance remain separate and unproved.