[llvm-branch-commits] [llvm] [Support] Allow restricting verify directives to comments (PR #233698)

[Alexander Richardson via llvm-branch-commits]

https://github.com/arichardson updated https://github.com/llvm/llvm-project/pull/233698

>From c8500b13268809067ef5dbf1682f385a087b9293 Mon Sep 17 00:00:00 2001
From: Alex Richardson <alexrichardson at google.com>
Date: Tue, 20 Oct 2026 14:56:05 -0700
Subject: [PATCH] [nspr] initial commit

---
 .../llvm/Support/SourceMgrDiagnosticVerifier.h    | 12 ++++++++++--
 llvm/lib/Support/SourceMgrDiagnosticVerifier.cpp  | 15 ++++++++++++++-
 .../Support/SourceMgrDiagnosticVerifierTest.cpp   |  7 ++++---
 3 files changed, 28 insertions(+), 6 deletions(-)

diff --git a/llvm/include/llvm/Support/SourceMgrDiagnosticVerifier.h b/llvm/include/llvm/Support/SourceMgrDiagnosticVerifier.h
index bcd1426e9548f..5ac34ad38a504 100644
--- a/llvm/include/llvm/Support/SourceMgrDiagnosticVerifier.h
+++ b/llvm/include/llvm/Support/SourceMgrDiagnosticVerifier.h
@@ -23,6 +23,7 @@
 #include "llvm/Support/Regex.h"
 #include "llvm/Support/SourceMgr.h"
 #include <optional>
+#include <string>
 
 namespace llvm {
 
@@ -66,9 +67,13 @@ class LLVM_ABI SourceMgrDiagnosticVerifier {
   /// bypassing \p mgr's diagnostic handler.
   ///
   /// \p prefixes are the directive prefixes to recognize (e.g. "expected"
-  /// for 'expected-error'); they must be non-empty.
+  /// for 'expected-error'); they must be non-empty. If \p commentPrefixes is
+  /// non-empty, directives are only recognized after the first occurrence of
+  /// one of these strings on a line, so that text outside of comments is
+  /// ignored.
   SourceMgrDiagnosticVerifier(const SourceMgr &mgr, raw_ostream &os,
-                              ArrayRef<StringRef> prefixes = {"expected"});
+                              ArrayRef<StringRef> prefixes = {"expected"},
+                              ArrayRef<StringRef> commentPrefixes = {});
 
   /// Matches a diagnostic against the expected diagnostics for \p loc, or
   /// against the '@unknown' expected diagnostics if \p loc is std::nullopt.
@@ -134,6 +139,9 @@ class LLVM_ABI SourceMgrDiagnosticVerifier {
   /// Regex used to recognize '<prefix>-<kind>' comments.
   Regex expected;
 
+  /// The strings that start a comment, see the constructor.
+  SmallVector<std::string, 0> commentPrefixes;
+
   /// The expected diagnostics for each buffer that has been scanned so far,
   /// keyed by buffer identifier (i.e. file name).
   StringMap<SmallVector<ExpectedDiag, 2>> expectedDiagsPerFile;
diff --git a/llvm/lib/Support/SourceMgrDiagnosticVerifier.cpp b/llvm/lib/Support/SourceMgrDiagnosticVerifier.cpp
index fac95397e9ecb..1275016e9bfc0 100644
--- a/llvm/lib/Support/SourceMgrDiagnosticVerifier.cpp
+++ b/llvm/lib/Support/SourceMgrDiagnosticVerifier.cpp
@@ -10,11 +10,13 @@
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/Support/MemoryBuffer.h"
 #include "llvm/Support/raw_ostream.h"
+#include <algorithm>
 
 using namespace llvm;
 
 SourceMgrDiagnosticVerifier::SourceMgrDiagnosticVerifier(
-    const SourceMgr &mgr, raw_ostream &os, ArrayRef<StringRef> prefixes)
+    const SourceMgr &mgr, raw_ostream &os, ArrayRef<StringRef> prefixes,
+    ArrayRef<StringRef> commentPrefixes)
     : mgr(mgr), os(os) {
   assert(!prefixes.empty() && "must have at least one prefix");
   std::string prefixAlt;
@@ -29,6 +31,11 @@ SourceMgrDiagnosticVerifier::SourceMgrDiagnosticVerifier(
                    ")-(error|note|remark|warning)(-re)?"
                    " *(@([+-][0-9]+|above|below|unknown))? *{{(.*)}}$");
 
+  for (StringRef prefix : commentPrefixes) {
+    assert(!prefix.empty() && "comment prefixes must be non-empty");
+    this->commentPrefixes.push_back(prefix.str());
+  }
+
   for (unsigned i = 0, e = mgr.getNumBuffers(); i != e; ++i)
     (void)computeExpectedDiags(mgr.getMemoryBuffer(i + 1));
 }
@@ -126,6 +133,12 @@ SourceMgrDiagnosticVerifier::computeExpectedDiags(const MemoryBuffer *buf) {
   buf->getBuffer().split(lines, '\n');
   for (unsigned lineNo = 0, e = lines.size(); lineNo < e; ++lineNo) {
     StringRef line = lines[lineNo].rtrim();
+    if (!commentPrefixes.empty()) {
+      size_t commentStart = StringRef::npos;
+      for (StringRef prefix : commentPrefixes)
+        commentStart = std::min(commentStart, line.find(prefix));
+      line = line.substr(commentStart);
+    }
 
     // Every directive contains '{{'; check that first to avoid running the
     // regex on most lines.
diff --git a/llvm/unittests/Support/SourceMgrDiagnosticVerifierTest.cpp b/llvm/unittests/Support/SourceMgrDiagnosticVerifierTest.cpp
index 19beb0fd98988..034d181ac379f 100644
--- a/llvm/unittests/Support/SourceMgrDiagnosticVerifierTest.cpp
+++ b/llvm/unittests/Support/SourceMgrDiagnosticVerifierTest.cpp
@@ -122,15 +122,16 @@ TEST_F(SourceMgrDiagnosticVerifierTest, InstalledAsDiagHandler) {
   EXPECT_FALSE(Verifier.verify());
 }
 
-TEST_F(SourceMgrDiagnosticVerifierTest, CustomPrefixes) {
+TEST_F(SourceMgrDiagnosticVerifierTest, CustomPrefixesAndCommentScoping) {
   const MemoryBuffer *Buf =
       addBuffer("op\n"
                 "// check-error at -1 {{first prefix}}\n"
                 "# alt-warning at -2 {{second prefix}}\n"
                 "// uncheck-error at -3 {{longer prefix ignored}}\n"
                 "// other-check-error at -4 {{hyphenated prefix ignored}}\n"
-                "// expected-error at -5 {{default prefix ignored}}\n");
-  SourceMgrDiagnosticVerifier Verifier(SM, OS, {"check", "alt"});
+                "// expected-error at -5 {{default prefix ignored}}\n"
+                "op check-error {{outside comment ignored}}\n");
+  SourceMgrDiagnosticVerifier Verifier(SM, OS, {"check", "alt"}, {"//", "#"});
 
   EXPECT_EQ(
       Verifier.process(getMessage(Buf, SourceMgr::DK_Error, "first prefix")),