Files
carbon-lang/toolchain/diagnostics/emitted_diagnostics_test.cpp
T
Richard SmithandJon Ross-Perkins e68e54dae4 Issue a diagnostic if we try to parse a source file that is too large. (#4429)
Previously in an optimized build we'd produce bogus tokens, such as
tokens with incorrect IdentifierIds, and in a debug build we would try
to CHECK-fail -- but actually wouldn't, because we're incorrectly
checking for `2 << bits` instead of `1 << bits`. I hit this while I was
trying to do some profiling and was seeing some very strange
diagnostics.

The diagnostic is pointed at the first token that is beyond the limit to
help people determine where to split their files.

---------

Co-authored-by: Jon Ross-Perkins <jperkins@google.com>
2024-10-22 23:21:35 +00:00

127 lines
4.5 KiB
C++

// Part of the Carbon Language project, under the Apache License v2.0 with LLVM
// Exceptions. See /LICENSE for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
#include <gtest/gtest.h>
#include <fstream>
#include <string>
#include "common/set.h"
#include "llvm/ADT/StringExtras.h"
#include "re2/re2.h"
#include "toolchain/diagnostics/diagnostic_kind.h"
namespace Carbon {
namespace {
constexpr DiagnosticKind Diagnostics[] = {
#define CARBON_DIAGNOSTIC_KIND(Name) DiagnosticKind::Name,
#include "toolchain/diagnostics/diagnostic_kind.def"
};
// Returns true for diagnostics which have no tests. In general, diagnostics
// should be tested.
static auto IsUntestedDiagnostic(DiagnosticKind diagnostic_kind) -> bool {
switch (diagnostic_kind) {
case DiagnosticKind::TestDiagnostic:
case DiagnosticKind::TestDiagnosticNote:
// These exist only for unit tests.
return true;
case DiagnosticKind::ErrorReadingFile:
case DiagnosticKind::ErrorStattingFile:
case DiagnosticKind::FileTooLarge:
// These diagnose filesystem issues that are hard to unit test.
return true;
case DiagnosticKind::ArrayBoundTooLarge:
// Int literals are currently limited to i32. Once that's fixed, this
// should be tested.
return true;
case DiagnosticKind::ExternLibraryInImporter:
case DiagnosticKind::ExternLibraryOnDefinition:
case DiagnosticKind::HexadecimalEscapeMissingDigits:
case DiagnosticKind::ImplOfUndefinedInterface:
case DiagnosticKind::IncompleteTypeInFunctionParam:
case DiagnosticKind::InvalidDigit:
case DiagnosticKind::InvalidDigitSeparator:
case DiagnosticKind::InvalidHorizontalWhitespaceInString:
case DiagnosticKind::MismatchedIndentInString:
case DiagnosticKind::ModifierPrivateNotAllowed:
case DiagnosticKind::MultiLineStringWithDoubleQuotes:
case DiagnosticKind::NameAmbiguousDueToExtend:
case DiagnosticKind::TooManyDigits:
case DiagnosticKind::UnaryOperatorRequiresWhitespace:
case DiagnosticKind::UnicodeEscapeSurrogate:
case DiagnosticKind::UnicodeEscapeTooLarge:
case DiagnosticKind::UnknownBaseSpecifier:
case DiagnosticKind::UnsupportedCRLineEnding:
case DiagnosticKind::UnsupportedLFCRLineEnding:
// TODO: Should look closer at these, but adding tests is a high risk of
// loss in merge conflicts due to the amount of tests being changed right
// now.
return true;
case DiagnosticKind::TooManyTokens:
// This isn't feasible to test with a normal testcase, but is tested in
// lex/tokenized_buffer_test.cpp.
return true;
default:
return false;
}
}
TEST(EmittedDiagnostics, Verify) {
std::ifstream manifest_in("toolchain/diagnostics/all_testdata.txt");
ASSERT_TRUE(manifest_in.good());
RE2 diagnostic_re(R"(\w\((\w+)\): )");
ASSERT_TRUE(diagnostic_re.ok()) << diagnostic_re.error();
Set<std::string> emitted_diagnostics;
std::string test_filename;
while (std::getline(manifest_in, test_filename)) {
std::ifstream test_in(test_filename);
ASSERT_TRUE(test_in.good());
std::string line;
while (std::getline(test_in, line)) {
std::string diagnostic;
if (RE2::PartialMatch(line, diagnostic_re, &diagnostic)) {
emitted_diagnostics.Insert(diagnostic);
}
}
}
llvm::SmallVector<llvm::StringRef> missing_diagnostics;
for (auto diagnostic_kind : Diagnostics) {
if (IsUntestedDiagnostic(diagnostic_kind)) {
EXPECT_FALSE(emitted_diagnostics.Erase(diagnostic_kind.name()))
<< diagnostic_kind
<< " was previously untested, and is now tested. That's good, but "
"please remove it from IsUntestedDiagnostic.";
continue;
}
if (!emitted_diagnostics.Erase(diagnostic_kind.name())) {
missing_diagnostics.push_back(diagnostic_kind.name());
}
}
constexpr llvm::StringLiteral Bullet = "\n - ";
std::sort(missing_diagnostics.begin(), missing_diagnostics.end());
EXPECT_TRUE(missing_diagnostics.empty())
<< "Some diagnostics have no tests:" << Bullet
<< llvm::join(missing_diagnostics, Bullet);
llvm::SmallVector<std::string> unexpected_matches;
emitted_diagnostics.ForEach(
[&](const std::string& match) { unexpected_matches.push_back(match); });
std::sort(unexpected_matches.begin(), unexpected_matches.end());
EXPECT_TRUE(unexpected_matches.empty())
<< "Matched things that don't appear to be diagnostics:" << Bullet
<< llvm::join(unexpected_matches, Bullet);
}
} // namespace
} // namespace Carbon