Track DBIx::Class acceptance for reachability PR #1647
Maintainer antworten meist innerhalb von 1 Tag
Dieses Issue hat noch niemand übernommen.
Bewertung
- Schwierigkeit
- 5/5
- Geschätzter Aufwand
- Über eine Woche
- Anfängerfreundlichkeit
- 12/100
Rechercherichtung
This is an umbrella tracker, not a single change. Start by reproducing the Storage object-collection failure in t/100extra_source.t from the DBIx::Class test suite, comparing system Perl with PerlOnJava. Trace which runtime-held references keep the blessed Storage object alive (refcount 131). Done means that test's two assertions pass and the fix has its own project regression test. Link the work to #1649.
Vom Indexierungsmodell aus dem Issue-Text verfasst.
Beschreibung
Purpose
Track DBIx::Class CPAN acceptance for the lifecycle work around PR #1647 and ownership semantics in #1649. Run DBIx::Class unchanged so failures expose PerlOnJava behavior directly.
Relationship to #1649 and PR #1647
The initial failures included premature cleanup of returned DBIx::Class references, which is part of the ownership and lifetime behavior described by #1649. The nested-query bind failure also exposed stale $#array proxies after a packed query passed through returned references. The runtime fix is in PerlOnJava; DBIx::Class does not need a compatibility patch.
PR #1647 remains the current independent reachability/lifecycle PR. Use the DBIx::Class acceptance results to identify remaining lifecycle regressions and link focused core fixes to #1649.
Plan
- Fix PerlOnJava ownership and array-size proxy behavior
- Preserve arrays strongly referenced by a subroutine return until the caller receives that value.
- Keep a fresh
$#arrayaccess valid when the array is reached through a live Perl reference, even if earlier scope cleanup orphaned old size proxies.
- Keep permanent regression coverage
- Validate the core returned-query range behavior with the unit suite and both backends.
- Validate grouped counts and nested
FROMbind values with upstream DBIx::Class, unchanged.
- Fix the remaining object-lifecycle regressions
- Reproduce the Storage object retention from
t/100extra_source.t; system Perl passes while PerlOnJava fails both object-collection assertions. - Trace which PerlOnJava-owned references keep that blessed Storage object reachable and correct their release/weak-reference handling in the runtime.
- Add focused project-owned coverage for the reproduced ownership defect.
- Reproduce the Storage object retention from
- Complete DBIx::Class acceptance
- Re-run the full
nice -n 19 ./jcpan --jobs 5 -t DBIx::Classsuite with a timeout and captured output after the runtime fix. - Triage every remaining failing test program against system Perl; separately diagnose the late SQLite test-lock collision seen in the parallel run.
- Re-run the full
- Update the handoff
- Report DBIx lifecycle acceptance separately from unrelated failures in PR #1647.
- Keep #1649 open until its broader ownership criteria are met.
Current evidence
- The focused grouped-count regression passes all eight assertions on system Perl, JVM, and interpreter with upstream DBIx::Class 0.082844 and no DBIx patch.
- The core returned-query range regression passes on system Perl, JVM, and interpreter.
nice -n 19 makeandmake check-linkspass with the current core changes.- The latest full run completed 325 test files / 43,363 tests: 183 test programs and 362 subtests failed. The earlier run had 184 failing programs / 447 failing subtests. These are PerlOnJava defects to fix, not acceptable compatibility failures.
- In a focused comparison, system Perl passes
t/100extra_source.t(11 assertions); PerlOnJava fails its two DBIx::Class Storage object-collection assertions. The object remains reported at refcount 131. - The parallel full run also ended with a SQLite test-lock collision in
xt/whitespace.t; establish whether that is runner coordination or an interpreter lock bug.
Completion criteria
- The focused regressions pass on system Perl, JVM, and interpreter.
- The Storage ownership defect is fixed in PerlOnJava with permanent regression coverage.
- The full DBIx::Class run completes and every remaining failure is categorized and addressed; lifecycle fixes are linked to #1649.
- PR #1647's DBIx lifecycle acceptance is recorded without treating any PerlOnJava defect as expected.
- Vorherrschende Sprache
- Perl
- Sterne
- 64
- Forks
- 8
- Ø Merge
- 5 Std. 43 Min.
- Gemergte PRs (30 T.)
- 204
Entwicklungsumgebung
- Enthält ein Dockerfile oder eine Docker-Compose-Datei
- Keine Pull-Request-Vorlage
- Beitragsleitfaden lesen
Erste Schritte
- Lesen Sie das ganze Issue und danach den Beitragsleitfaden des Projekts.
- Schreiben Sie ins Issue, dass Sie es übernehmen — das erspart doppelte Arbeit.
- Forken Sie das Repository und arbeiten Sie in einem Branch.
- Öffnen Sie einen Pull Request, der die Issue-Nummer nennt.
Mehr aus fglock/PerlOnJava
-
bug
Schwierigkeit 2/5 1-3 Stunden Anfängerfreundlichkeit 72/100
fglock/PerlOnJava#1579 ·
Maintainer antworten meist innerhalb von 1 Tag
-
Schwierigkeit 3/5 1-2 Tage Anfängerfreundlichkeit 56/100
fglock/PerlOnJava#1851 ·
Maintainer antworten meist innerhalb von 1 Tag
-
perf: close method-call throughput gap to blead PerlEvtl. vergeben Ein verknüpfter Pull Request ist offen oder bereits gemergt. Offenarea:performance enhancement
Schwierigkeit 5/5 Über eine Woche Anfängerfreundlichkeit 25/100
fglock/PerlOnJava#1846 · 1 Kommentar ·
Maintainer antworten meist innerhalb von 1 Tag
-
Schwierigkeit 3/5 1-2 Tage Anfängerfreundlichkeit 48/100
fglock/PerlOnJava#1835 ·
Maintainer antworten meist innerhalb von 1 Tag
-
area:cpan-port bug
Schwierigkeit 3/5 1-2 Tage Anfängerfreundlichkeit 68/100
fglock/PerlOnJava#1820 ·
Maintainer antworten meist innerhalb von 1 Tag
Alle Issues in fglock/PerlOnJava
Ähnliche Issues
-
fix(api): API v3 Product Read ignores request language (lc) for sample products and image URLsEvtl. vergeben @muntasir-396 hat das vor 1 Tag übernommen. Offen
Schwierigkeit 2/5 1-3 Stunden Anfängerfreundlichkeit 88/100
openfoodfacts/openfoodfacts-server#15042 · 1 Kommentar ·
Maintainer antworten meist innerhalb von 1 Tag
-
Schwierigkeit 2/5 1-3 Stunden Anfängerfreundlichkeit 75/100
Maintainer antworten meist innerhalb von 1 Tag
-
Schwierigkeit 1/5 Unter einer Stunde Anfängerfreundlichkeit 78/100
libwww-perl/URI#214 ·
-
bug
Schwierigkeit 2/5 1-3 Stunden Anfängerfreundlichkeit 78/100
dekobon/big-code-analysis#1644 ·
Maintainer antworten meist innerhalb von 1 Tag
-
Type: Bug
Schwierigkeit 2/5 1-3 Stunden Anfängerfreundlichkeit 82/100
inverse-inc/packetfence#9549 ·
Maintainer antworten meist innerhalb von 1 Tag