[00xDisaster]

What

Fill the exported collate<char> vtable with the facet methods that already back g_collateVtable. #9979 added the nine-slot _ZTVSt7collateIcE_nid_postfix data object (NID ruZtIwbCFjk) with the console's 72-byte layout, but every slot was zero, so a title that called compare, transform or hash through the exported vtable faulted. Slots 2 through 8 now hold DestroyCollate, DeleteCollate, RetainFacet, ReleaseFacet, CollateCompare, CollateTransform and CollateHash; slots 0 and 1 stay zero for offset-to-top and typeinfo, as they were. _Getcat still installs g_collateVtable, so both the exported object and facets built by _Getcat call the same implementations.

The exported object is still writable data and supplies no RTTI typeinfo; the debt entry records both. This does not touch time_put or num_put: open #11044 and #11054 implement those different virtuals, and no open PR currently fills this array. #10158 and #10200 add the distinct adjacent messages and numpunct vtable declarations in the same LocaleSupport.cpp block; both are currently CONFLICTING, so this PR merges first and they rebase afterwards, keeping all three declarations. No dependency is required.

Tested

Linux CachyOS 7.3.0-4-cachyos, GCC 16.2.1. Base upstream/main 74ac0e7912a5eaa453fff872f4d4fe803154b08c; head c66f61ff5259e3d05eab0e35c99b948d2b62903b.

• cmake --build build --target guest_locale_tests --parallel: passed.
• ctest --test-dir build -R '^guest_locale$' --output-on-failure: 1/1 passed.
• guest_locale now calls compare, transform, hash, retain and release through &_ZTVSt7collateIcE_nid_postfix[2], checks the compare ordering and FNV hash result, and checks that slots 0 and 1 remain zero while slots 2 through 8 are non-null.
• tools/check_conventions.py --base upstream/main --head HEAD and git diff --check: clean.
• GPUs tested: n/a.

Checklist

• Based on current main; no other open PR implements the same functions
• One topic per PR; follow-ups go in a new PR
• No comments in code except technical debt
• Unimplemented paths throw (NotImplemented_nid_no_patch); silent stubs are listed under Silent stubs in their library's technical debt file
• No notes, investigation .md files or images added to the repository (attach them to this PR)
• New third-party code is a submodule built from source
• Generic behaviour, not specific to one title
• Depends on: none
• Fixes: none
• Unblocks: none
• GPUs tested: n/a
• AI-assisted: no

[MarcZX10]

Reviewed exact head c66f61ff5259e3d05eab0e35c99b948d2b62903b.

The exported nine-slot object points at the same collate methods as the existing facet vtable, leaves offset-to-top/typeinfo zero as documented, and the test invokes all populated slots through the exported object. The debt update correctly records the remaining RTTI limitation and the evidence boundary.

Checks: git diff --check upstream/main...HEAD passes and the conventions, eligibility, intake, labels and report checks pass. The author reports guest_locale_tests and guest_locale passing locally. The current hosted relinker and register-context jobs pass; macOS relinker/register-context and write-watch jobs are still queued, while the general Linux/Windows jobs are skipped by the path policy. I did not run a local build in this review.

[github-actions]

Other open pull requests touch the same code. Merging one will break the other, or both implement the same thing. If this one builds on them, list them in Depends on.

Pull request	Conflicting files (lines on main)	Same exports	Same new files
PR 10158	core/libs/prx/libc/src/LocaleSupport.cpp L374
PR 10200	core/libs/prx/libc/src/LocaleSupport.cpp L374
PR 10749	docs/dev/debt/libc.md L190-191

Some comparisons could not be completed because a pull request conflicts with the base branch. Review these shared files manually before concluding that there is no overlap.

Pull request	Shared files not compared
PR 5021	docs/dev/debt/libc.md
PR 5303	docs/dev/debt/libc.md
PR 6713	docs/dev/debt/libc.md
PR 7081	core/libs/prx/libc/src/LocaleSupport.cpp docs/dev/debt/libc.md
PR 7386	docs/dev/debt/libc.md
PR 7447	docs/dev/debt/libc.md
PR 7544	docs/dev/debt/libc.md
PR 7648	docs/dev/debt/libc.md
PR 7715	docs/dev/debt/libc.md
PR 8150	docs/dev/debt/libc.md
PR 9371	docs/dev/debt/libc.md
PR 9447	docs/dev/debt/libc.md
PR 9526	core/libs/prx/libc/src/LocaleSupport.cpp core/libs/tests/GuestLocale.cpp docs/dev/debt/libc.md
PR 9772	docs/dev/debt/libc.md
PR 9863	docs/dev/debt/libc.md
PR 9929	docs/dev/debt/libc.md
PR 9953	docs/dev/debt/libc.md
PR 9964	docs/dev/debt/libc.md
PR 10002	docs/dev/debt/libc.md
PR 10007	docs/dev/debt/libc.md
PR 10234	docs/dev/debt/libc.md
PR 10315	docs/dev/debt/libc.md
PR 10317	docs/dev/debt/libc.md
PR 10321	docs/dev/debt/libc.md
PR 10325	docs/dev/debt/libc.md
PR 10328	docs/dev/debt/libc.md
PR 10345	docs/dev/debt/libc.md
PR 10349	docs/dev/debt/libc.md
PR 10405	docs/dev/debt/libc.md
PR 10467	docs/dev/debt/libc.md
PR 10506	docs/dev/debt/libc.md
PR 10510	docs/dev/debt/libc.md
PR 10512	docs/dev/debt/libc.md
PR 10529	docs/dev/debt/libc.md
PR 10565	docs/dev/debt/libc.md
PR 11332	docs/dev/debt/libc.md
PR 11341	docs/dev/debt/libc.md

[MarcZX10]

Follow-up to my review: the remaining hosted checks have completed. Linux/macOS/Windows relinker, ASan, Linux Debug/Release and macOS Intel Debug/Release register-context, and macOS write-watch jobs pass. Full Linux and Windows product jobs are skipped by the repository path policy.

[MarcZX10]

I manually checked the current overlap warnings. #10158 adds _ZTVSt8messagesIcE_nid_postfix and #10200 adds _ZTVSt8numpunctIcE_nid_postfix in the same LocaleSupport.cpp declaration block where this PR implements _ZTVSt7collateIcE_nid_postfix; these are distinct vtables, not duplicate implementations. A three-way merge with each current head reports a content conflict in LocaleSupport.cpp at that insertion point, while the shared GuestLocale.cpp edits merge cleanly. Please coordinate the insertion/rebase resolution before combining them.

[wachi35676]

Windows run of head c66f61ff5259 merged onto main 90f918fa7. Windows 11 10.0.26300, WinLibs GCC 15.2.0, Release.

• guest_locale passes, 2 runs. That covers compare, transform, hash, retain and release through the exported vtable.
• With main's LocaleSupport.cpp, it aborts with Guest locale check failed.
• It merges cleanly together with feat(libc): wire the codecvt<wchar_t,char> facet vtable #11172 and feat(libc): format num_put<char>::do_put as the console's libc does #11054, which also edit LocaleSupport.cpp and GuestLocale.cpp.

AI-assisted: yes (Claude Code); results above were run on this machine.

[00xDisaster]

Merge-order follow-up on the overlap: #10158 and #10200 add the distinct adjacent messages and numpunct vtable declarations in the same LocaleSupport.cpp block, not the collate export. A three-way merge with their current heads conflicts only at that insertion point. Both of those PRs are currently CONFLICTING, while this head is CLEAN and fully reviewed, so I propose this PR merge first and #10158/#10200 rebase afterwards, keeping all three declarations and tests. The dependency line stays none. I did not modify either other branch; if you or mazen320 prefer the opposite order, I will rebase this one instead.

[00xDisaster]

All checks are now green on c66f61ff52 (18 pass, 2 path-skipped), including the macOS relinker, register-context and write-watch jobs. The merge-order note for #10158/#10200 is in the body and comment above; no code changed since your review. Could you approve this head if you have no further concerns?

[oneandonlydean]

@00xDisaster Thanks for checking the merge with #10158 and #10200. That order works for us: this one first, then those two rebase and keep all three declarations and their tests. The Linux and Windows product jobs were path-skipped on this head, so we'll still run those before it goes in.