[flang-commits] [flang] [flang][parser] Better error recovery for SUBROUTINE/FUNCTION statements (PR #100664)

[Peter Klausler via flang-commits]

https://github.com/klausler updated https://github.com/llvm/llvm-project/pull/100664

>From 05dec22e2434af3b2fb5bba7b1a689ac3b5ac206 Mon Sep 17 00:00:00 2001
From: Peter Klausler <pklausler at nvidia.com>
Date: Thu, 25 Jul 2024 15:35:52 -0700
Subject: [PATCH] [flang][parser] Better error recovery for SUBROUTINE/FUNCTION
 statements

When there's an error in a SUBROUTINE or FUNCTION statement,
errors cascade quickly because the body of the subprogram or
interface isn't in the right context.  So, if a SUBROUTINE or
FUNCTION statement is expected, and contains a SUBROUTINE or
FUNCTION keyword, it counts as one -- retain and emit any errors
pertaining to the arguments or suffix, recover to the end of the
line if needed, and proceed.
---
 flang/lib/Parser/program-parsers.cpp | 35 +++++++++++++++++-----------
 flang/lib/Parser/stmt-parser.h       |  6 +++--
 flang/test/Parser/recovery04.f90     | 24 +++++++++++++++++++
 3 files changed, 50 insertions(+), 15 deletions(-)
 create mode 100644 flang/test/Parser/recovery04.f90

diff --git a/flang/lib/Parser/program-parsers.cpp b/flang/lib/Parser/program-parsers.cpp
index b51b60157f39c..c12b6de01bf03 100644
--- a/flang/lib/Parser/program-parsers.cpp
+++ b/flang/lib/Parser/program-parsers.cpp
@@ -36,10 +36,12 @@ namespace Fortran::parser {
 // Enforcing C1547 is done in semantics.
 static constexpr auto programUnit{
     construct<ProgramUnit>(indirect(Parser<Module>{})) ||
-    construct<ProgramUnit>(indirect(functionSubprogram)) ||
     construct<ProgramUnit>(indirect(subroutineSubprogram)) ||
     construct<ProgramUnit>(indirect(Parser<Submodule>{})) ||
     construct<ProgramUnit>(indirect(Parser<BlockData>{})) ||
+    // ambiguities: REALFUNCTIONF, REALFUNCTIONF(10) if "error" recovered
+    !statement(typeDeclarationStmt) >>
+        construct<ProgramUnit>(indirect(functionSubprogram)) ||
     construct<ProgramUnit>(indirect(Parser<MainProgram>{}))};
 static constexpr auto normalProgramUnit{StartNewSubprogram{} >> programUnit /
         skipMany(";"_tok) / space / recovery(endOfLine, SkipPast<'\n'>{})};
@@ -532,15 +534,20 @@ TYPE_CONTEXT_PARSER("FUNCTION subprogram"_en_US,
 //         [prefix] FUNCTION function-name ( [dummy-arg-name-list] ) [suffix]
 // R1526 prefix -> prefix-spec [prefix-spec]...
 // R1531 dummy-arg-name -> name
-TYPE_CONTEXT_PARSER("FUNCTION statement"_en_US,
-    construct<FunctionStmt>(many(prefixSpec), "FUNCTION" >> name,
-        parenthesized(optionalList(name)), maybe(suffix)) ||
+
+TYPE_PARSER(construct<FunctionStmt>(many(prefixSpec), "FUNCTION" >> name,
+                parenthesized(optionalList(name)), maybe(suffix)) /
+        atEndOfStmt ||
+    construct<FunctionStmt>(many(prefixSpec), "FUNCTION" >> name / atEndOfStmt,
+        // PGI & Intel accept "FUNCTION F"
         extension<LanguageFeature::OmitFunctionDummies>(
             "nonstandard usage: FUNCTION statement without dummy argument list"_port_en_US,
-            construct<FunctionStmt>( // PGI & Intel accept "FUNCTION F"
-                many(prefixSpec), "FUNCTION" >> name,
-                construct<std::list<Name>>(),
-                construct<std::optional<Suffix>>())))
+            pure<std::list<Name>>()),
+        pure<std::optional<Suffix>>()) ||
+    // error recovery
+    construct<FunctionStmt>(many(prefixSpec), "FUNCTION" >> name,
+        defaulted(parenthesized(optionalList(name))), maybe(suffix)) /
+        checkEndOfKnownStmt)
 
 // R1532 suffix ->
 //         proc-language-binding-spec [RESULT ( result-name )] |
@@ -566,11 +573,13 @@ TYPE_CONTEXT_PARSER("SUBROUTINE subprogram"_en_US,
 //         [prefix] SUBROUTINE subroutine-name [( [dummy-arg-list] )
 //         [proc-language-binding-spec]]
 TYPE_PARSER(
-    construct<SubroutineStmt>(many(prefixSpec), "SUBROUTINE" >> name,
-        parenthesized(optionalList(dummyArg)), maybe(languageBindingSpec)) ||
-    construct<SubroutineStmt>(many(prefixSpec), "SUBROUTINE" >> name,
-        pure<std::list<DummyArg>>(),
-        pure<std::optional<LanguageBindingSpec>>()))
+    (construct<SubroutineStmt>(many(prefixSpec), "SUBROUTINE" >> name,
+         !"("_tok >> pure<std::list<DummyArg>>(),
+         pure<std::optional<LanguageBindingSpec>>()) ||
+        construct<SubroutineStmt>(many(prefixSpec), "SUBROUTINE" >> name,
+            defaulted(parenthesized(optionalList(dummyArg))),
+            maybe(languageBindingSpec))) /
+    checkEndOfKnownStmt)
 
 // R1536 dummy-arg -> dummy-arg-name | *
 TYPE_PARSER(construct<DummyArg>(name) || construct<DummyArg>(star))
diff --git a/flang/lib/Parser/stmt-parser.h b/flang/lib/Parser/stmt-parser.h
index ba647fd60d4ae..00bae2bf950c8 100644
--- a/flang/lib/Parser/stmt-parser.h
+++ b/flang/lib/Parser/stmt-parser.h
@@ -30,6 +30,10 @@ inline constexpr auto unterminatedStatement(const PA &p) {
           maybe(label), space >> p));
 }
 
+constexpr auto atEndOfStmt{space >>
+    withMessage("expected end of statement"_err_en_US, lookAhead(";\n"_ch))};
+constexpr auto checkEndOfKnownStmt{recovery(atEndOfStmt, SkipTo<'\n'>{})};
+
 constexpr auto endOfLine{
     "\n"_ch >> ok || fail("expected end of line"_err_en_US)};
 
@@ -86,8 +90,6 @@ constexpr auto executionPartErrorRecovery{stmtErrorRecoveryStart >>
 // END statement error recovery
 constexpr auto missingOptionalName{pure<std::optional<Name>>()};
 constexpr auto noNameEnd{"END" >> missingOptionalName};
-constexpr auto atEndOfStmt{space >>
-    withMessage("expected end of statement"_err_en_US, lookAhead(";\n"_ch))};
 constexpr auto bareEnd{noNameEnd / recovery(atEndOfStmt, SkipTo<'\n'>{})};
 
 // For unrecognizable construct END statements.  Be sure to not consume
diff --git a/flang/test/Parser/recovery04.f90 b/flang/test/Parser/recovery04.f90
new file mode 100644
index 0000000000000..144ebd24f71b5
--- /dev/null
+++ b/flang/test/Parser/recovery04.f90
@@ -0,0 +1,24 @@
+! RUN: not %flang_fc1 -fsyntax-only %s 2>&1 | FileCheck %s
+module m
+ contains
+  !CHECK: expected end of statement
+  !CHECK: subroutine s1(var i, j)
+  subroutine s1(var i, j)
+  end subroutine
+  !CHECK: expected end of statement
+  !CHECK: subroutine s2[b]
+  subroutine s2[b]
+  end subroutine
+  !CHECK: expected end of statement
+  !CHECK: function f1(var i, j)
+  function f1(var i, j)
+  end function
+  !CHECK: expected end of statement
+  !CHECK: function f2[b]
+  function f2[b]
+  end function
+  !CHECK: expected end of statement
+  !CHECK: function f3(a,*)
+  function f3(a,*)
+  end function
+end