Skip to content

10.x: translation-support inventory, every public query member classified and probed (port of #384) #430

Closed

Description

@CaffeinatedCoder

Problem

The 10.3.0 additions (#405, #406, #191) are done only when test/oracles/translation-support.json classifies them server and the translation-support probe passes (#408). That inventory is #384, and it doesn't exist on release/10.x. So none of the 10.3.0 cherry-picks can meet its own "done when", and a 10.x public member whose translation appears or disappears is still noticed only if a hand-written test calls it.

The inventory has already been run on this line. On 2026-10-25, PR #386's classification and probe ran without changes on a scratch release/10.x checkout (03c04e4, not committed), on EF Core 10.0.0 and on 10.0.13 with Npgsql 10.0.3. They gave the same 2,256 cells as main, and the 10.x public API was fully classified. The only failures were #385's 13 cells (26 lines), which 10.x's candidates.json doesn't register (comment on #384).

Change

Once #384 is on main, cherry-pick it with -x into a PR against release/10.x:

Done when

TranslationSupportInventoryTests and TranslationSupportProbeTests pass on release/10.x on the floor and newest-patch legs. The seed fails the probe (dotnet run tools/oracles.cs seeds). The port lands before #405, #406 and #191 are picked, so each of those moves its own entry to server.

Release

Tests only, with no package change and no changed answer. Fixed only on release/10.x, so close it by hand.

Activity

  1. added this to the 10.3.0 milestone on Oct 25, 2026
  2. added
    testsAdds or changes tests only (oracles, harnesses); no library code
    area/rangesRange unions, variants and the shape engines
    area/range-setRangeSet and multirange algebra
    area/value-setsValue set families and wrapper arities
    package/efcoreAffects CodoMetis.ValueRanges.EFCore.PostgreSQL
    package/efcore-nodatimeAffects CodoMetis.ValueRanges.EFCore.PostgreSQL.NodaTime
    on Oct 25, 2026
  3. CaffeinatedCoder commented on Oct 25, 2026

    @CaffeinatedCoder
    OwnerAuthor

    PR #431. One change from the plan above: the probe's composition helpers come over as main's four Composition/ files, unchanged from 30a0e24, not as a trimmed copy. The probe uses Projected, CompositionRunner.IsNull and FirstLine, and IsNull relies on NullRule. A trimmed copy would differ between the lines, and a later port of the composition sweeps can now add only the sweep classes. Verified on all four 10.x legs (PostgreSQL 16 and 18 × EF floor and 10.0.13/10.0.3), and the seed is caught.

  4. CaffeinatedCoder commented on Oct 25, 2026

    @CaffeinatedCoder
    OwnerAuthor

    Fixed on release/10.x by PR #431 (merged as d1a5041). The inventory and the probe now run on this line, so #405, #406 and #191 can each move their entry to server when they are cherry-picked. Closed by hand, because Closes #n acts only on main.

  5. added a commit that references this issue on Oct 25, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    area/range-setRangeSet and multirange algebraarea/rangesRange unions, variants and the shape enginesarea/value-setsValue set families and wrapper aritiespackage/efcoreAffects CodoMetis.ValueRanges.EFCore.PostgreSQLpackage/efcore-nodatimeAffects CodoMetis.ValueRanges.EFCore.PostgreSQL.NodaTimetestsAdds or changes tests only (oracles, harnesses); no library code

    Projects

    No projects

      Milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions