Repository navigation
Prelude surface: classes, instances, and the value functions #94
Description
Part of #76 (Phase 3). The highest-leverage library slice: 163 passing files are blocked on Prelude alone, more than every other missing module combined.
stdlib/lib/Prelude.purs is currently six declarations — pure, bind, discard, map, apply, runEffect over foreign import data Effect. The corpus expects the standard classes and instances.
Classes and instances
Show, Eq, Ord, Semigroup, Monoid, Functor, Apply, Applicative, Foldable, plus Boolean combinators, folds, and the String/Char/Number conversions the corpus uses.
Prerequisites this exposes
Each of these is Phase 2 work, and writing the library is the forcing function that finds the rest:
- Instances must resolve — Resolve instance declarations and instance signatures #90. Every class above needs instances, and instances are not in the resolver yet.
- Records and rows on the type spine — class methods take and return records;
ShowneedsRecord.toList-shaped display helpers that use open rows. donotation with multiple statements andlet— already lowered, but the library will stress it.- Guards —
Ord/Eqinstance bodies are guard-shaped (Lower guarded equations, guarded case alternatives, and multi-scrutinee case #88). - Operator sections —
(<>),flip, and friends are written as sections (Lower operator, constructor-operator and type-operator aliases #86).
Constraint
Every addition compiles and executes under Wasmtime, not only type-checks. A Prelude that type-checks but cannot run does not recover a single passing file.
Acceptance
Prelude carries the classes and instances above, executes under Wasmtime, and the passing files blocked on Prelude fall toward zero.
Activity
biuld commentedon Oct 3, 2026
First slice landed in #128:
Data.Functionand the$/#application operators.What landed
Data.Function:apply,applyFlipped,const,flip,on, plusinfixr 0 apply as $andinfixl 1 applyFlipped as #.Preludere-exportsconst,flip,$,#(the official list) with an explicit export list.pursrule), and CC's partial-application adapter now converts the parameters it leaves to the caller to the declaration's shapes.identityis deliberately not defined here: official re-exports it fromControl.Category, so it belongs with the class hierarchy rather than as a privateData.Function.identity.Measured
PSRS_ORACLE=annotations cargo test -p psrs-driver --test suite -- --ignored --nocapturepassingresolutionBoard field
The
Corpus casesfield is now 80: the measured number ofpassingfiles whose first blocker is aPreludeclass or value name (the convention counts first blocking stage, so this is the latent recovery the issue's surface owns, not a change in the first-blocker total). The largest groups are<>15,Monad9,identity/show/Eq7 each,negate5,Functor5,otherwise/&&/<$>/add3 each, and singles forSemiring,Show,when,append,mod,between, and>>>.Remaining
The class hierarchy and its instances (
Semigroup/<>needs a string- and array-concatenation primitive),Show/logShow(needs number formatting andData.Show),Eq/Ordand the arithmetic classes whose surface operators are currently Int-monomorphic intrinsics, andidentity/Control.Category.Merge pull request #129 from biuld/docs/phase3-corpus-cases
biuld commentedon Oct 3, 2026
Landed the first slice:
Data.Semigroupand<>(PR #130).stdlib/lib/Data/Semigroup.pursdeclaresclass Semigroup a where append :: a -> a -> a,infixr 5 append as <>, and theString,Unit, andArray ainstances;Preludere-exportsclass Semigroup,append,<>.arrayAppendprimitive (HIRIntrinsic::ArrayAppend→ CoreExprKind::ArrayAppend→ CCAssignmentKind::ArrayAppend). Because the MIRArrayNewDefaultverifier enforces array-conversion invariants, a generalArrayNewSizedallocator plus two element-copy loops lower it.PSRS_REQUIRE_WASMTIME=1 cargo test -p psrs-driver --lib tests::semigroup(4 pass; Wasmtime execution for string/array append including a non-ASCII string, and right-associativity).Measured with
PSRS_ORACLE=annotations cargo test -p psrs-driver --test suite -- --ignored --nocapture(2026-10-04):NoInstanceFound<>was the largest single P3 name (15 cases) andappendblocked 2 more; all 17 now stop on a later stage, and 10 cases resolve past L2 (5 P10, 3 P5 typecheck, 1 P8, 1 P6). The boardCorpus casesfor this issue drops 80 → 70.Still open here:
show/Show,Monad,Eq,identity,Functor/<$>,negate, and the rest of the class/value surface;assertEqualandlogShow(shared with #95) still needEq/Show.biuld commentedon Oct 4, 2026
Landed
Data.Functorin #142 (merge commit77c8745).class Functor,map, and<$>live inData.Functor, withArray,Maybe, andEitherinstances.Preludere-exports them and adds theEffectinstance. That instance is the old effect-onlymap: creating the action does not run it.One
PSRS_REQUIRE_WASMTIME=1 PSRS_ORACLE=annotationsscoreboard on 2026-10-04, Wasmtime 49.0.2,purs0.15.16:All 125 runtime files exit 0. The added file is
passing/3549.purs.failing/3549.pursnow matchesKindsDoNotUnify.failing/NewtypeInstance5.pursleaves the L5 board and fails in the newtype checker. L1 stays 904/908 and L2 failing stays 72/72.Functorand<$>are gone from the first-blocker names. Corpus cases is now 67:Monad14,identity9,_8,negate6,Eq14,<<<4,otherwise3,&&3,when2,mod2,compare2, and one each ofbetween,P.identity,>>>,<*>,not,Apply,||,Applicative,-, andFoo.Bar.#94 stays open.
Correct the first-stage blocker decomposition and close out Phase 2
Land Effect and Effect.Console, dropping 259 blocked files
Land the unit value and Test.Assert's assertion surface
Land Data.Function and the $/# application operators
Record #94's measured corpus count in the Phase 3 row
Merge pull request #129 from biuld/docs/phase3-corpus-cases
Add Data.Functor and re-export map from Prelude
Vendor the v0.15.16 core libraries into the standard library
Vendor the v0.15.16 core libraries into the standard library
biuld commentedon Oct 16, 2026
The 2026-10-17 scoreboard no longer has a
passingfile whose first blocker is a missingPreludename.Corpus casesmoves from 67 to 0.L2
passingresolution is 403/413, with 0 missing-library blockers, reconfirmed on 2026-10-17 against the vendored corpus (PSRS_ORACLE=annotations, compiler886cd82c). The 6 P3 files areDeclConflict,UnknownExport,ScopeConflict, andExportConflict. This issue stays open; the field tracks the first-blocker count, and the class surface is not re-audited by this comment.Evidence: L2 and L6/M7 in
docs/design/D-04-suite-roadmap.md, recorded by #145.biuld commentedon Oct 18, 2026
The class surface this issue owns is measured complete on the corpus: the 2026-10-17 L2 board first reported no
passingfile blocked on aPreludename (Corpus cases67 → 0), reconfirmed 2026-10-18 on masterf5abccda(PR #148): passing resolution 403/413 with 0 missing-library blockers, and L6/M7 287/340 with no in-scope program's first blocker insidePrelude's surface. The corpus-facing classes landed across the phase's slices:Data.Function($/#),Data.Semigroup(<>),Data.Show,Data.Eq/Data.Ord/Data.Semiring(intrinsic layer),Data.Functor,Data.Monoid,Data.Foldable,Data.Tuple,Data.Traversable, plus the official module surface in the vendored package (e556bf2f, 313 trusted modules). Residual library-foreign work (Test.Assert.checkThrows) is #95's Blocked row; broader class/fundep solver acceptance is #97 (closed) / #98 / #100's scope. Closing on the measured evidence.