[flang] Fix preprocessor regression

[klausler]

For numeric kind suffixes like 1_foo, the preprocessor should be able to perform macro replacement for macros named either "_foo" or "foo".

Fixes #133399.

[llvmbot]

@llvm/pr-subscribers-flang-parser

Author: Peter Klausler (klausler)

Changes

For numeric kind suffixes like 1_foo, the preprocessor should be able to perform macro replacement for macros named either "_foo" or "foo".

Fixes #133399.

---

Full diff: https://github.com/llvm/llvm-project/pull/134405.diff

3 Files Affected:

• (modified) flang/lib/Parser/prescan.cpp (+17-4)
• (modified) flang/lib/Parser/token-sequence.cpp (+1-1)
• (modified) flang/test/Preprocessing/kind-suffix.F90 (+3)

diff --git a/flang/lib/Parser/prescan.cpp b/flang/lib/Parser/prescan.cpp
index 755cb18cb8caf..91c10b2e48cf7 100644
--- a/flang/lib/Parser/prescan.cpp
+++ b/flang/lib/Parser/prescan.cpp
@@ -875,11 +875,24 @@ bool Prescanner::HandleExponent(TokenSequence &tokens) {
 }
 
 bool Prescanner::HandleKindSuffix(TokenSequence &tokens) {
-  if (*at_ == '_' && IsLegalInIdentifier(*SkipWhiteSpace(at_ + 1))) {
-    EmitCharAndAdvance(tokens, *at_);
-    if (IsLegalIdentifierStart(*at_)) {
-      // The kind specifier might be a macro, so put it into its own token.
+  if (*at_ == '_' && IsLegalInIdentifier(at_[1])) {
+    // The kind specifier might be a macro (with or without its leading
+    // underscore); put it into its own token if it has been defined.
+    const char *p{at_ + 1};
+    while (IsLegalInIdentifier(*++p)) {
+    }
+    if (CharBlock id{at_, static_cast<std::size_t>(p - at_)};
+        preprocessor_.IsNameDefined(id)) {
+      // In 1.0e0_foo, "_foo" is a defined name; retain the
+      // underscore
       tokens.CloseToken();
+    } else {
+      EmitCharAndAdvance(tokens, '_');
+      if (CharBlock id{at_, static_cast<std::size_t>(p - at_)};
+          preprocessor_.IsNameDefined(id)) {
+        // In 1.0e0_foo, "foo" is a defined name
+        tokens.CloseToken();
+      }
     }
     while (IsLegalInIdentifier(*at_)) {
       EmitCharAndAdvance(tokens, *at_);
diff --git a/flang/lib/Parser/token-sequence.cpp b/flang/lib/Parser/token-sequence.cpp
index cdbe89b1eb441..c3016707b8ebf 100644
--- a/flang/lib/Parser/token-sequence.cpp
+++ b/flang/lib/Parser/token-sequence.cpp
@@ -187,7 +187,7 @@ TokenSequence &TokenSequence::ToLowerCase() {
       } else if (*p == 'h' || *p == 'H') {
         // Hollerith
         *p = 'h';
-      } else if (*p == '_') {
+      } else if (*p == '_' && p + 1 < limit && (p[1] == '"' || p[1] == '\'')) {
         // kind-prefixed character literal (e.g., 1_"ABC")
       } else {
         // exponent
diff --git a/flang/test/Preprocessing/kind-suffix.F90 b/flang/test/Preprocessing/kind-suffix.F90
index 36aa323630c6c..ea2a3daaed40d 100644
--- a/flang/test/Preprocessing/kind-suffix.F90
+++ b/flang/test/Preprocessing/kind-suffix.F90
@@ -1,6 +1,9 @@
 ! RUN: %flang -E %s 2>&1 | FileCheck %s
 #define n k
+#define _m _p
 parameter(n=4)
 !CHECK: print *,1_k
 print *,1_n
+!CHECK: print *,1_p
+print *,1_m
 end