|
1 | | -# Target for the push-triggered oracle (.github/workflows/oracle.yml, push path). |
2 | | -# Format = mine-target.txt: first non-comment line is "owner/repo", an https URL, |
3 | | -# or "local:<in-repo path>" (a fixture copied into the oracle's target/). |
4 | | -# Optional lines: ref=, paths=, build=, include_tests=. Dev-branch only. |
5 | | -# |
6 | | -# Cross-tool oracle on a Linux-buildable fixture, so ALL THREE tools run (Infer# |
7 | | -# included — ScreenToGif's WPF won't build on Linux). Re-run after the OWN014 |
8 | | -# migration (PR #35): the fixture's SUBSCRIPTION leak (#1 — SystemEvents.Display |
9 | | -# SettingsChanged += with no -=) is now lowered by the extractor to a `capture` |
10 | | -# fact and reported as OWN014 (region escape), not OWN001; OWN014 was added to the |
11 | | -# comparator's OWN_LEAK set. So #1 must STILL land "Own.NET-only" (CodeQL / Infer# |
12 | | -# have no event-subscription-leak query) — this run confirms the migration kept the |
13 | | -# differentiation, on real cross-tool output rather than the selftest's fixtures. |
14 | | -# Expect: #1 Own.NET-only (now OWN014), #2/#3/#4 dispose leaks in "Agree", 0 |
15 | | -# oracle-only. |
16 | | -local:corpus/fixtures/systemevents-console |
17 | | -build=SystemEventsLeak.csproj |
| 1 | +# Oracle target: SixLabors/ImageSharp — the repo we just mined. Validate our post-fix |
| 2 | +# leak findings against CodeQL (cs/local-not-disposed family) + Infer# (RESOURCE_LEAK) |
| 3 | +# on real PRODUCT code (comparator excludes tests by default). Expectation: the pure |
| 4 | +# IDisposable leaks (DeflateCompressor's empty Dispose) AGREE; the ArrayPool buffer |
| 5 | +# (SharedArrayPoolBuffer) and any subscription escapes land Own.NET-only — defect |
| 6 | +# classes the oracle leak queries can't express, not FPs. |
| 7 | +SixLabors/ImageSharp |
| 8 | +build=src/ImageSharp/ImageSharp.csproj |
0 commit comments