internal/pkg: content-based dependency substitution

This change introduces a new fast path for FloodArtifact. It is taken when a curing artifact has identical-by-content controlled relevant inputs and are otherwise identical to an already-cured artifact.

Signed-off-by: Ophestra <cat@gensokyo.uk>

internal/pkg/ir.go

@@ -76,6 +76,9 @@ type IContext struct {
 	// Written to by various methods, should be zeroed after [Artifact.Params]
 	// returns and must not be exposed directly.
 	w io.Writer
+	// Optional identifier to checksum cache, replaces [IRKindIdent] with
+	// checksum values if non-nil.
+	ident map[unique.Handle[ID]]unique.Handle[Checksum]
 }
 
 // irZero is a zero IR word.
@@ -163,7 +166,15 @@ func (i *IContext) WriteIdent(a Artifact) {
 	defer i.ic.putIdentBuf(buf)
 
 	IRKindIdent.encodeHeader(0).put(buf[:])
-	*(*ID)(buf[wordSize:]) = i.ic.Ident(a).Value()
+	if i.ident != nil {
+		checksum, ok := i.ident[i.ic.Ident(a)]
+		if !ok {
+			panic(InvalidLookupError(checksum.Value()))
+		}
+		*(*ID)(buf[wordSize:]) = checksum.Value()
+	} else {
+		*(*ID)(buf[wordSize:]) = i.ic.Ident(a).Value()
+	}
 	i.mustWrite(buf[:])
 }
 
@@ -207,6 +218,16 @@ func (i *IContext) WriteString(s string) {
 // Encode writes a deterministic, efficient representation of a to w and returns
 // the first non-nil error encountered while writing to w.
 func (ic *irCache) Encode(w io.Writer, a Artifact) (err error) {
+	return ic.encode(w, a, nil)
+}
+
+// encode implements Encode but replaces identifiers with their cured checksums
+// for a non-nil ident. Caller must acquire Cache.identMu.
+func (ic *irCache) encode(
+	w io.Writer,
+	a Artifact,
+	ident map[unique.Handle[ID]]unique.Handle[Checksum],
+) (err error) {
 	deps := a.Dependencies()
 	idents := make([]*extIdent, len(deps))
 	for i, d := range deps {
@@ -228,6 +249,17 @@ func (ic *irCache) Encode(w io.Writer, a Artifact) (err error) {
 		return *a == *b
 	})
 
+	// late substitution preserves order
+	if ident != nil {
+		for _, dbuf := range idents {
+			checksum, ok := ident[unique.Make(ID(dbuf[wordSize:]))]
+			if !ok {
+				return InvalidLookupError(dbuf[wordSize:])
+			}
+			*(*ID)(dbuf[wordSize:]) = checksum.Value()
+		}
+	}
+
 	// kind uint64 | deps_sz uint64
 	var buf [wordSize * 2]byte
 	binary.LittleEndian.PutUint64(buf[:], uint64(a.Kind()))
@@ -244,7 +276,7 @@ func (ic *irCache) Encode(w io.Writer, a Artifact) (err error) {
 	}
 
 	func() {
-		i := IContext{ic, w}
+		i := IContext{ic, w, ident}
 
 		defer panicToError(&err)
 		defer func() { i.ic, i.w = nil, nil }()