| 1 |
//===-- DILLexerTests.cpp --------------------------------------------===// |
| 2 |
// |
| 3 |
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. |
| 4 |
// See https://llvm.org/LICENSE.txt for license information. |
| 5 |
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception |
| 6 |
// |
| 7 |
//===----------------------------------------------------------------------===// |
| 8 |
|
| 9 |
#include "lldb/ValueObject/DILLexer.h" |
| 10 |
#include "llvm/ADT/StringRef.h" |
| 11 |
#include "llvm/Testing/Support/Error.h" |
| 12 |
#include "gtest/gtest.h" |
| 13 |
#include <string> |
| 14 |
|
| 15 |
using llvm::StringRef; |
| 16 |
|
| 17 |
using namespace lldb_private::dil; |
| 18 |
|
| 19 |
llvm::Expected<std::vector<std::pair<Token::Kind, std::string>>> |
| 20 |
ExtractTokenData(llvm::StringRef input_expr) { |
| 21 |
|
| 22 |
llvm::Expected<DILLexer> maybe_lexer = DILLexer::Create(input_expr); |
| 23 |
if (!maybe_lexer) |
| 24 |
return maybe_lexer.takeError(); |
| 25 |
DILLexer lexer(*maybe_lexer); |
| 26 |
|
| 27 |
std::vector<std::pair<Token::Kind, std::string>> data; |
| 28 |
do { |
| 29 |
Token tok = lexer.GetCurrentToken(); |
| 30 |
data.push_back(std::make_pair(tok.GetKind(), tok.GetSpelling())); |
| 31 |
lexer.Advance(); |
| 32 |
} while (data.back().first != Token::eof); |
| 33 |
// Don't return the eof token. |
| 34 |
data.pop_back(); |
| 35 |
return data; |
| 36 |
} |
| 37 |
|
| 38 |
TEST(DILLexerTests, SimpleTest) { |
| 39 |
StringRef input_expr("simple_var"); |
| 40 |
llvm::Expected<DILLexer> maybe_lexer = DILLexer::Create(input_expr); |
| 41 |
ASSERT_THAT_EXPECTED(maybe_lexer, llvm::Succeeded()); |
| 42 |
DILLexer lexer(*maybe_lexer); |
| 43 |
Token token = lexer.GetCurrentToken(); |
| 44 |
|
| 45 |
EXPECT_EQ(token.GetKind(), Token::identifier); |
| 46 |
EXPECT_EQ(token.GetSpelling(), "simple_var"); |
| 47 |
lexer.Advance(); |
| 48 |
token = lexer.GetCurrentToken(); |
| 49 |
EXPECT_EQ(token.GetKind(), Token::eof); |
| 50 |
} |
| 51 |
|
| 52 |
TEST(DILLexerTests, TokenKindTest) { |
| 53 |
Token token = Token(Token::identifier, "ident", 0); |
| 54 |
|
| 55 |
EXPECT_TRUE(token.Is(Token::identifier)); |
| 56 |
EXPECT_FALSE(token.Is(Token::l_paren)); |
| 57 |
EXPECT_TRUE(token.IsOneOf({Token::eof, Token::identifier})); |
| 58 |
EXPECT_FALSE(token.IsOneOf( |
| 59 |
{Token::l_paren, Token::r_paren, Token::coloncolon, Token::eof})); |
| 60 |
} |
| 61 |
|
| 62 |
TEST(DILLexerTests, LookAheadTest) { |
| 63 |
StringRef input_expr("(anonymous namespace)::some_var"); |
| 64 |
llvm::Expected<DILLexer> maybe_lexer = DILLexer::Create(input_expr); |
| 65 |
ASSERT_THAT_EXPECTED(maybe_lexer, llvm::Succeeded()); |
| 66 |
DILLexer lexer(*maybe_lexer); |
| 67 |
Token token = lexer.GetCurrentToken(); |
| 68 |
|
| 69 |
// Current token is '('; check the next 4 tokens, to make |
| 70 |
// sure they are the identifier 'anonymous', the identifier 'namespace' |
| 71 |
// ')' and '::', in that order. |
| 72 |
EXPECT_EQ(token.GetKind(), Token::l_paren); |
| 73 |
EXPECT_EQ(lexer.LookAhead(1).GetKind(), Token::identifier); |
| 74 |
EXPECT_EQ(lexer.LookAhead(1).GetSpelling(), "anonymous"); |
| 75 |
EXPECT_EQ(lexer.LookAhead(2).GetKind(), Token::identifier); |
| 76 |
EXPECT_EQ(lexer.LookAhead(2).GetSpelling(), "namespace"); |
| 77 |
EXPECT_EQ(lexer.LookAhead(3).GetKind(), Token::r_paren); |
| 78 |
EXPECT_EQ(lexer.LookAhead(4).GetKind(), Token::coloncolon); |
| 79 |
|
| 80 |
// Our current index should still be 0, as we only looked ahead; we are still |
| 81 |
// officially on the '('. |
| 82 |
EXPECT_EQ(lexer.GetCurrentTokenIdx(), 0u); |
| 83 |
|
| 84 |
// Accept the 'lookahead', so our current token is '::', which has the index |
| 85 |
// 4 in our vector of tokens (which starts at zero). |
| 86 |
lexer.Advance(4); |
| 87 |
token = lexer.GetCurrentToken(); |
| 88 |
EXPECT_EQ(token.GetKind(), Token::coloncolon); |
| 89 |
EXPECT_EQ(lexer.GetCurrentTokenIdx(), 4u); |
| 90 |
|
| 91 |
lexer.Advance(); |
| 92 |
token = lexer.GetCurrentToken(); |
| 93 |
EXPECT_EQ(token.GetKind(), Token::identifier); |
| 94 |
EXPECT_EQ(token.GetSpelling(), "some_var"); |
| 95 |
EXPECT_EQ(lexer.GetCurrentTokenIdx(), 5u); |
| 96 |
EXPECT_EQ(token.GetLocation(), strlen("(anonymous namespace)::")); |
| 97 |
|
| 98 |
lexer.Advance(); |
| 99 |
token = lexer.GetCurrentToken(); |
| 100 |
EXPECT_EQ(token.GetKind(), Token::eof); |
| 101 |
} |
| 102 |
|
| 103 |
TEST(DILLexerTests, MultiTokenLexTest) { |
| 104 |
EXPECT_THAT_EXPECTED( |
| 105 |
ExtractTokenData("This string has (several ) ::identifiers"), |
| 106 |
llvm::HasValue(testing::ElementsAre( |
| 107 |
testing::Pair(Token::identifier, "This"), |
| 108 |
testing::Pair(Token::identifier, "string"), |
| 109 |
testing::Pair(Token::identifier, "has"), |
| 110 |
testing::Pair(Token::l_paren, "("), |
| 111 |
testing::Pair(Token::identifier, "several"), |
| 112 |
testing::Pair(Token::r_paren, ")"), |
| 113 |
testing::Pair(Token::coloncolon, "::"), |
| 114 |
testing::Pair(Token::identifier, "identifiers")))); |
| 115 |
} |
| 116 |
|
| 117 |
TEST(DILLexerTests, IdentifiersTest) { |
| 118 |
// These strings should lex into identifier tokens. |
| 119 |
std::vector<std::string> valid_identifiers = { |
| 120 |
"$My_name1", "$pc", "abcd", "_", "_a", "_a_", "$", |
| 121 |
"a_b", "this", "self", "a", "MyName", "namespace"}; |
| 122 |
|
| 123 |
// The lexer can lex these strings, but they should not be identifiers. |
| 124 |
std::vector<std::string> invalid_identifiers = {"", "::", "(", ")", "0abc"}; |
| 125 |
|
| 126 |
// The lexer is expected to fail attempting to lex these strings (it cannot |
| 127 |
// create valid tokens out of them). |
| 128 |
std::vector<std::string> invalid_tok_strings = {"#include", "a@a"}; |
| 129 |
|
| 130 |
// Verify that all of the valid identifiers come out as identifier tokens. |
| 131 |
for (auto &str : valid_identifiers) { |
| 132 |
SCOPED_TRACE(str); |
| 133 |
EXPECT_THAT_EXPECTED(ExtractTokenData(str), |
| 134 |
llvm::HasValue(testing::ElementsAre( |
| 135 |
testing::Pair(Token::identifier, str)))); |
| 136 |
} |
| 137 |
|
| 138 |
// Verify that the lexer fails on invalid token strings. |
| 139 |
for (auto &str : invalid_tok_strings) { |
| 140 |
SCOPED_TRACE(str); |
| 141 |
auto maybe_lexer = DILLexer::Create(str); |
| 142 |
EXPECT_THAT_EXPECTED(maybe_lexer, llvm::Failed()); |
| 143 |
} |
| 144 |
|
| 145 |
// Verify that none of the invalid identifiers come out as identifier tokens. |
| 146 |
for (auto &str : invalid_identifiers) { |
| 147 |
SCOPED_TRACE(str); |
| 148 |
llvm::Expected<DILLexer> maybe_lexer = DILLexer::Create(str); |
| 149 |
EXPECT_THAT_EXPECTED(maybe_lexer, llvm::Succeeded()); |
| 150 |
DILLexer lexer(*maybe_lexer); |
| 151 |
Token token = lexer.GetCurrentToken(); |
| 152 |
EXPECT_TRUE(token.IsNot(Token::identifier)); |
| 153 |
EXPECT_TRUE(token.IsOneOf({Token::eof, Token::coloncolon, Token::l_paren, |
| 154 |
Token::r_paren, Token::numeric_constant})); |
| 155 |
} |
| 156 |
} |
| 157 |
|
| 158 |
TEST(DILLexerTests, NumbersTest) { |
| 159 |
// These strings should lex into number tokens. |
| 160 |
std::vector<std::string> valid_numbers = {"123", "0x123", "0123", "0b101"}; |
| 161 |
|
| 162 |
// The lexer can lex these strings, but they should not be numbers. |
| 163 |
std::vector<std::string> invalid_numbers = {"", "x123", "b123"}; |
| 164 |
|
| 165 |
for (auto &str : valid_numbers) { |
| 166 |
SCOPED_TRACE(str); |
| 167 |
EXPECT_THAT_EXPECTED(ExtractTokenData(str), |
| 168 |
llvm::HasValue(testing::ElementsAre( |
| 169 |
testing::Pair(Token::numeric_constant, str)))); |
| 170 |
} |
| 171 |
// Verify that none of the invalid numbers come out as numeric tokens. |
| 172 |
for (auto &str : invalid_numbers) { |
| 173 |
SCOPED_TRACE(str); |
| 174 |
llvm::Expected<DILLexer> maybe_lexer = DILLexer::Create(str); |
| 175 |
EXPECT_THAT_EXPECTED(maybe_lexer, llvm::Succeeded()); |
| 176 |
DILLexer lexer(*maybe_lexer); |
| 177 |
Token token = lexer.GetCurrentToken(); |
| 178 |
EXPECT_TRUE(token.IsNot(Token::numeric_constant)); |
| 179 |
EXPECT_TRUE(token.IsOneOf({Token::eof, Token::identifier})); |
| 180 |
} |
| 181 |
} |