Files
carbon-lang/toolchain/lex/tokenized_buffer_fuzzer.cpp
T
Jon Ross-Perkins 9ac92ad71b Add support for compiling multiple files at once. (#3182)
Rearranges driver logic into CompilationUnits in order to associate
artifacts from the various stages of compilation.

Note, I'm not totally sure what the right thing to do is for
lower/codegen, so I'm just doing a rote change there for now that
mirrors prior phases (this is all the code supports anyways, so is
probably right for now regardless).

SourceBuffer error output is moved local for consistency with other
steps, and so that it's less ambiguous whether the error should be
expected to already include a filename.
2023-09-06 21:34:59 +00:00

57 lines
2.2 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 <cstring>
#include "common/check.h"
#include "llvm/ADT/StringRef.h"
#include "toolchain/diagnostics/null_diagnostics.h"
#include "toolchain/lex/tokenized_buffer.h"
namespace Carbon::Testing {
// NOLINTNEXTLINE: Match the documented fuzzer entry point declaration style.
extern "C" int LLVMFuzzerTestOneInput(const unsigned char* data,
std::size_t size) {
// Ignore large inputs.
// TODO: Investigate replacement with an error limit. Content with errors on
// escaped quotes (`\"` repeated) have O(M * N) behavior for M errors in a
// file length N, so either that will need to also be fixed or M will need to
// shrink for large (1MB+) inputs.
// This also affects parse_tree_fuzzer.cpp.
if (size > 100000) {
return 0;
}
static constexpr llvm::StringLiteral TestFileName = "test.carbon";
llvm::vfs::InMemoryFileSystem fs;
llvm::StringRef data_ref(reinterpret_cast<const char*>(data), size);
CARBON_CHECK(fs.addFile(
TestFileName, /*ModificationTime=*/0,
llvm::MemoryBuffer::getMemBuffer(data_ref, /*BufferName=*/TestFileName,
/*RequiresNullTerminator=*/false)));
auto source = SourceBuffer::CreateFromFile(fs, llvm::nulls(), TestFileName);
auto buffer = Lex::TokenizedBuffer::Lex(*source, NullDiagnosticConsumer());
if (buffer.has_errors()) {
return 0;
}
// Walk the lexed and tokenized buffer to ensure it isn't corrupt in some way.
//
// TODO: We should enhance this to do more sanity checks on the resulting
// token stream.
for (Lex::Token token : buffer.tokens()) {
int line_number = buffer.GetLineNumber(token);
CARBON_CHECK(line_number > 0) << "Invalid line number!";
CARBON_CHECK(line_number < INT_MAX) << "Invalid line number!";
int column_number = buffer.GetColumnNumber(token);
CARBON_CHECK(column_number > 0) << "Invalid line number!";
CARBON_CHECK(column_number < INT_MAX) << "Invalid line number!";
}
return 0;
}
} // namespace Carbon::Testing