java.g grammar: parse underscore in pointcut as identifier

On source level 21 (preview), '_' is no longer scanned as an identifier but as a token UNDERSCORE. To avoid a compile error, explicitly parse the token as an identifier. See upstream commit 8398f6c. Signed-off-by: Alexander Kriegisch <[email protected]>
eclipse-aspectj · Feb 13, 2024 · 5d2f2ae · 5d2f2ae
1 parent 4ea8e1c
commit 5d2f2ae
Show file tree

Hide file tree

Showing 17 changed files with 585 additions and 575 deletions.
diff --git a/org.eclipse.jdt.core.compiler.batch/grammar/java.g b/org.eclipse.jdt.core.compiler.batch/grammar/java.g
@@ -1172,6 +1172,11 @@ ColonPseudoToken ::= ':'
 /.$putCase consumePseudoToken(":"); $break ./
 /:$readableName any allowable token in pointcut or type pattern, except ':':/
 
+-- On source level 21 (preview), '_' is no longer scanned as an identifier but as a token UNDERSCORE.
+-- To avoid a compile error, explicitly parse the token as an identifier. See upstream commit 8398f6c121.
+PseudoToken ::= '_'
+/.$putCase consumePseudoTokenIdentifier(); $break ./
+
 PseudoToken ::= JavaIdentifier
 /.$putCase consumePseudoTokenIdentifier(); $break ./
 /:$readableName allowable token in pointcut or type pattern:/