gradle: 8.4 and parser: lazy tokenization

build.gradle.kts

@@ -3,5 +3,5 @@ plugins {
 }
 
 tasks.withType<Wrapper> {
-  gradleVersion = "8.3"
+  gradleVersion = "8.4"
 }

gradle/wrapper/gradle-wrapper.properties

@@ -1,6 +1,6 @@
 distributionBase=GRADLE_USER_HOME
 distributionPath=wrapper/dists
-distributionUrl=https\://services.gradle.org/distributions/gradle-8.3-bin.zip
+distributionUrl=https\://services.gradle.org/distributions/gradle-8.4-bin.zip
 networkTimeout=10000
 validateDistributionUrl=true
 zipStoreBase=GRADLE_USER_HOME

gradlew

@@ -83,7 +83,8 @@ done
 # This is normally unused
 # shellcheck disable=SC2034
 APP_BASE_NAME=${0##*/}
-APP_HOME=$( cd "${APP_HOME:-./}" && pwd -P ) || exit
+# Discard cd standard output in case $CDPATH is set (https://github.com/gradle/gradle/issues/25036)
+APP_HOME=$( cd "${APP_HOME:-./}" > /dev/null && pwd -P ) || exit
 
 # Use the maximum available, or set MAX_FD != -1 to use that value.
 MAX_FD=maximum

minimal/src/main/kotlin/gay/pizza/pork/minimal/Tool.kt

@@ -22,14 +22,14 @@ abstract class Tool {
   val rootImportLocator: ImportLocator
     get() = ImportLocator("local", rootFilePath())
 
-  fun tokenize(): TokenStream =
+  fun tokenize(): LazyTokenSource =
-    Tokenizer(createCharSource()).stream()
+    LazyTokenSource(Tokenizer(createCharSource()))
 
   fun parse(attribution: NodeAttribution = DiscardNodeAttribution): CompilationUnit =
-    Parser(TokenStreamSource(tokenize()), attribution).parseCompilationUnit()
+    Parser(TokenStreamSource(tokenize().streamAllRemainingTokens()), attribution).parseCompilationUnit()
 
   fun highlight(scheme: HighlightScheme): List<Highlight> =
-    Highlighter(scheme).highlight(tokenize())
+    Highlighter(scheme).highlight(tokenize().streamAllRemainingTokens())
 
   fun reprint(): String = buildString { visit(Printer(this)) }
 

parser/src/main/kotlin/gay/pizza/pork/parser/LazyTokenSource.kt

@@ -0,0 +1,38 @@
+package gay.pizza.pork.parser
+
+class LazyTokenSource(val tokenizer: Tokenizer) : TokenSource {
+  private val queue = mutableListOf<Token>()
+  private var index = 0
+  override val currentIndex: Int
+    get() = index
+
+  override fun next(): Token {
+    index++
+    if (queue.isNotEmpty()) {
+      return queue.removeFirst()
+    }
+    return tokenizer.next()
+  }
+
+  override fun peek(): Token {
+    if (queue.isNotEmpty()) {
+      return queue.first()
+    }
+    val token = tokenizer.next()
+    queue.add(token)
+    return token
+  }
+
+  override fun peekTypeAhead(ahead: Int): TokenType {
+    wantAtLeast(ahead + 1)
+    return queue[ahead].type
+  }
+
+  private fun wantAtLeast(ahead: Int) {
+    if (queue.size < ahead) {
+      for (i in 1..ahead) {
+        queue.add(tokenizer.next())
+      }
+    }
+  }
+}

parser/src/main/kotlin/gay/pizza/pork/parser/TokenSource.kt

@@ -14,6 +14,9 @@ interface TokenSource : PeekableSource<Token> {
     return tokens
   }
 
+  fun streamAllRemainingTokens(): TokenStream =
+    TokenStream(consumeAllRemainingTokens().filter { !TokenType.ParserIgnoredTypes.contains(it.type) })
+
   fun ignoringParserIgnoredTypes(): TokenSource =
-    TokenStreamSource(TokenStream(consumeAllRemainingTokens().filter { !TokenType.ParserIgnoredTypes.contains(it.type) }))
+    TokenStreamSource(streamAllRemainingTokens())
 }

tool/src/main/kotlin/gay/pizza/pork/tool/TokenizeCommand.kt

@@ -4,15 +4,21 @@ import com.github.ajalt.clikt.core.CliktCommand
 import com.github.ajalt.clikt.parameters.arguments.argument
 import gay.pizza.dough.fs.PlatformFsProvider
 import gay.pizza.pork.minimal.FileTool
+import gay.pizza.pork.parser.TokenType
 
 class TokenizeCommand : CliktCommand(help = "Tokenize Compilation Unit", name = "tokenize") {
   val path by argument("file")
 
   override fun run() {
     val tool = FileTool(PlatformFsProvider.resolve(path))
-    val tokenStream = tool.tokenize()
+    val tokenSource = tool.tokenize()
-    for (token in tokenStream.tokens) {
+    while (true) {
-      println("${token.sourceIndex.index} ${token.type.name} '${sanitize(token.text)}'")
+      val token = tokenSource.next()
+      println("${token.sourceIndex} ${token.type.name} '${sanitize(token.text)}'")
+      tokenSource.peekTypeAhead(5)
+      if (token.type == TokenType.EndOfFile) {
+        break
+      }
     }
   }