zzy/scc
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor(lex_parser): 移除旧的词法解析器实现并更新依赖

Browse Source 
移除了 libs/lex_parser 目录下的所有头文件和源文件，包括：
- lex_parser.h 和 lex_parser.c 核心解析功能
- 所有测试文件（test_char.c, test_identifier.c, test_number.c,
  test_skip_block_comment.c, test_skip_line.c, test_string.c）

更新了 lexer 模块的依赖配置，将 lex_parser 替换为 sstream，
同时更新了 lexer.h 中的相关包含头文件和数据结构定义，
简化了 scc_lexer_t 结构体的字段。
This commit is contained in:
zzy
2026-02-16 16:56:40 +08:00
parent 088050c903
commit 0e7dec202a
30 changed files with 1840 additions and 1979 deletions
Download Patch File Download Diff File Expand all files Collapse all files

403
libs/lexer/tests/test_lexer.c Normal file
View File

@@ -0,0 +1,403 @@
// test_lexer.c
#include <lexer.h>
#include <string.h>
#include <utest/acutest.h>
// 辅助函数：释放 token 的 lexeme
static void free_token(scc_lexer_tok_t *tok) { scc_cstring_free(&tok->lexeme); }
// 单 token 测试宏（检查类型）
#define TEST_TOKEN(input, expected_type) \
do { \
scc_lexer_t lexer; \
scc_lexer_tok_t token; \
scc_sstream_t stream; \
scc_sstream_init_by_buffer(&stream, input, strlen(input), 0, 16); \
scc_sstream_ring_t *ref = scc_sstream_ref_ring(&stream); \
scc_lexer_init(&lexer, ref); \
scc_lexer_get_token(&lexer, &token); \
\
TEST_CHECK(token.type == expected_type); \
TEST_MSG("Input: '%s'", input); \
TEST_MSG("Expected: %s", scc_get_tok_name(expected_type)); \
TEST_MSG("Got: %s", scc_get_tok_name(token.type)); \
\
free_token(&token); \
scc_sstream_drop_ring(ref); \
scc_sstream_drop(&stream); \
} while (0)
// 多 token 序列测试宏（接受类型数组）
#define TEST_SEQUENCE(input, ...) \
do { \
scc_lexer_t lexer; \
scc_lexer_tok_t token; \
scc_sstream_t stream; \
scc_sstream_init_by_buffer(&stream, input, strlen(input), 0, 16); \
scc_sstream_ring_t *ref = scc_sstream_ref_ring(&stream); \
scc_lexer_init(&lexer, ref); \
\
scc_tok_type_t expected[] = {__VA_ARGS__}; \
size_t count = sizeof(expected) / sizeof(expected[0]); \
for (size_t i = 0; i < count; i++) { \
scc_lexer_get_token(&lexer, &token); \
TEST_CHECK(token.type == expected[i]); \
TEST_MSG("Token %zu: input '%s'", i, input); \
TEST_MSG("Expected: %s", scc_get_tok_name(expected[i])); \
TEST_MSG("Got: %s", scc_get_tok_name(token.type)); \
free_token(&token); \
} \
\
scc_sstream_drop_ring(ref); \
scc_sstream_drop(&stream); \
} while (0)
// ============================ 测试用例 ============================
void test_operators() {
TEST_CASE("Arithmetic operators");
TEST_TOKEN("+", SCC_TOK_ADD);
TEST_TOKEN("++", SCC_TOK_ADD_ADD);
TEST_TOKEN("+=", SCC_TOK_ASSIGN_ADD);
TEST_TOKEN("-", SCC_TOK_SUB);
TEST_TOKEN("--", SCC_TOK_SUB_SUB);
TEST_TOKEN("-=", SCC_TOK_ASSIGN_SUB);
TEST_TOKEN("*", SCC_TOK_MUL);
TEST_TOKEN("*=", SCC_TOK_ASSIGN_MUL);
TEST_TOKEN("/", SCC_TOK_DIV);
TEST_TOKEN("/=", SCC_TOK_ASSIGN_DIV);
TEST_TOKEN("%", SCC_TOK_MOD);
TEST_TOKEN("%=", SCC_TOK_ASSIGN_MOD);
TEST_CASE("Bitwise operators");
TEST_TOKEN("&", SCC_TOK_AND);
TEST_TOKEN("&&", SCC_TOK_AND_AND);
TEST_TOKEN("&=", SCC_TOK_ASSIGN_AND);
TEST_TOKEN("|", SCC_TOK_OR);
TEST_TOKEN("||", SCC_TOK_OR_OR);
TEST_TOKEN("|=", SCC_TOK_ASSIGN_OR);
TEST_TOKEN("^", SCC_TOK_XOR);
TEST_TOKEN("^=", SCC_TOK_ASSIGN_XOR);
TEST_TOKEN("~", SCC_TOK_BIT_NOT);
TEST_TOKEN("<<", SCC_TOK_L_SH);
TEST_TOKEN("<<=", SCC_TOK_ASSIGN_L_SH);
TEST_TOKEN(">>", SCC_TOK_R_SH);
TEST_TOKEN(">>=", SCC_TOK_ASSIGN_R_SH);
TEST_CASE("Comparison operators");
TEST_TOKEN("==", SCC_TOK_EQ);
TEST_TOKEN("!=", SCC_TOK_NEQ);
TEST_TOKEN("<", SCC_TOK_LT);
TEST_TOKEN("<=", SCC_TOK_LE);
TEST_TOKEN(">", SCC_TOK_GT);
TEST_TOKEN(">=", SCC_TOK_GE);
TEST_CASE("Special symbols");
TEST_TOKEN("(", SCC_TOK_L_PAREN);
TEST_TOKEN(")", SCC_TOK_R_PAREN);
TEST_TOKEN("[", SCC_TOK_L_BRACKET);
TEST_TOKEN("]", SCC_TOK_R_BRACKET);
TEST_TOKEN("{", SCC_TOK_L_BRACE);
TEST_TOKEN("}", SCC_TOK_R_BRACE);
TEST_TOKEN(";", SCC_TOK_SEMICOLON);
TEST_TOKEN(",", SCC_TOK_COMMA);
TEST_TOKEN(":", SCC_TOK_COLON);
TEST_TOKEN(".", SCC_TOK_DOT);
TEST_TOKEN("...", SCC_TOK_ELLIPSIS);
TEST_TOKEN("->", SCC_TOK_DEREF);
TEST_TOKEN("?", SCC_TOK_COND);
}
void test_keywords() {
TEST_CASE("C89 keywords");
TEST_TOKEN("while", SCC_TOK_WHILE);
TEST_TOKEN("sizeof", SCC_TOK_SIZEOF);
TEST_TOKEN("if", SCC_TOK_IF);
TEST_TOKEN("else", SCC_TOK_ELSE);
TEST_TOKEN("for", SCC_TOK_FOR);
TEST_TOKEN("do", SCC_TOK_DO);
TEST_TOKEN("switch", SCC_TOK_SWITCH);
TEST_TOKEN("case", SCC_TOK_CASE);
TEST_TOKEN("default", SCC_TOK_DEFAULT);
TEST_TOKEN("break", SCC_TOK_BREAK);
TEST_TOKEN("continue", SCC_TOK_CONTINUE);
TEST_TOKEN("return", SCC_TOK_RETURN);
TEST_TOKEN("goto", SCC_TOK_GOTO);
TEST_TOKEN("auto", SCC_TOK_AUTO);
TEST_TOKEN("register", SCC_TOK_REGISTER);
TEST_TOKEN("static", SCC_TOK_STATIC);
TEST_TOKEN("extern", SCC_TOK_EXTERN);
TEST_TOKEN("typedef", SCC_TOK_TYPEDEF);
TEST_TOKEN("const", SCC_TOK_CONST);
TEST_TOKEN("volatile", SCC_TOK_VOLATILE);
TEST_TOKEN("signed", SCC_TOK_SIGNED);
TEST_TOKEN("unsigned", SCC_TOK_UNSIGNED);
TEST_TOKEN("short", SCC_TOK_SHORT);
TEST_TOKEN("long", SCC_TOK_LONG);
TEST_TOKEN("int", SCC_TOK_INT);
TEST_TOKEN("char", SCC_TOK_CHAR);
TEST_TOKEN("float", SCC_TOK_FLOAT);
TEST_TOKEN("double", SCC_TOK_DOUBLE);
TEST_TOKEN("void", SCC_TOK_VOID);
TEST_TOKEN("struct", SCC_TOK_STRUCT);
TEST_TOKEN("union", SCC_TOK_UNION);
TEST_TOKEN("enum", SCC_TOK_ENUM);
TEST_CASE("C99 keywords");
TEST_TOKEN("inline", SCC_TOK_INLINE);
TEST_TOKEN("restrict", SCC_TOK_RESTRICT);
// _Bool, _Complex, _Imaginary 可根据需要添加
TEST_CASE("SCC extensions (if enabled)");
TEST_TOKEN("asm", SCC_TOK_ASM);
TEST_TOKEN("atomic", SCC_TOK_ATOMIC);
TEST_TOKEN("bool", SCC_TOK_BOOL);
TEST_TOKEN("complex", SCC_TOK_COMPLEX);
}
void test_literals() {
TEST_CASE("Integer literals - decimal");
TEST_TOKEN("0", SCC_TOK_INT_LITERAL);
TEST_TOKEN("123", SCC_TOK_INT_LITERAL);
TEST_TOKEN("2147483647", SCC_TOK_INT_LITERAL);
TEST_TOKEN("4294967295", SCC_TOK_INT_LITERAL);
TEST_CASE("Integer literals - hexadecimal");
TEST_TOKEN("0x0", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0x1A3F", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0XABCDEF", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0x123abc", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0XFF", SCC_TOK_INT_LITERAL);
TEST_CASE("Integer literals - octal");
TEST_TOKEN("0123", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0777", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0", SCC_TOK_INT_LITERAL); // 0 既是十进制也是八进制
TEST_CASE("Integer literals - binary (C23 extension)");
TEST_TOKEN("0b1010", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0B1100", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0b0", SCC_TOK_INT_LITERAL);
TEST_CASE("Integer literals with suffixes");
TEST_TOKEN("123U", SCC_TOK_INT_LITERAL);
TEST_TOKEN("456L", SCC_TOK_INT_LITERAL);
TEST_TOKEN("789UL", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0x1FFLL", SCC_TOK_INT_LITERAL);
TEST_TOKEN("0b1010ULL", SCC_TOK_INT_LITERAL);
TEST_CASE("Floating literals - decimal");
TEST_TOKEN("0.0", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("3.14", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN(".5", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("0.", SCC_TOK_FLOAT_LITERAL);
TEST_CASE("Floating literals - scientific");
TEST_TOKEN("1e10", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("1E-5", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("2.5e+3", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN(".1e2", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("1.e3", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("123.456e-7", SCC_TOK_FLOAT_LITERAL);
TEST_CASE("Floating literals - hexadecimal (C99)");
TEST_TOKEN("0x1.2p3", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("0x1p-2", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("0x0.1p10", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("0X1.2P3", SCC_TOK_FLOAT_LITERAL);
TEST_CASE("Floating literals with suffixes");
TEST_TOKEN("1.0f", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("2.0F", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("3.0l", SCC_TOK_FLOAT_LITERAL);
TEST_TOKEN("4.0L", SCC_TOK_FLOAT_LITERAL);
TEST_CASE("Character literals - simple");
TEST_TOKEN("'a'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'0'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("' '", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\t'", SCC_TOK_CHAR_LITERAL); // 制表符在单引号内
TEST_CASE("Character literals - escape sequences");
TEST_TOKEN("'\\n'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\\t'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\\\\'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\\''", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\\\"'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\\?'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\\0'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\\123'", SCC_TOK_CHAR_LITERAL); // 八进制
TEST_TOKEN("'\\xAB'", SCC_TOK_CHAR_LITERAL); // 十六进制
TEST_CASE("Character literals - multi-byte (implementation defined)");
TEST_TOKEN("'ab'", SCC_TOK_CHAR_LITERAL);
TEST_TOKEN("'\\x41\\x42'", SCC_TOK_CHAR_LITERAL); // 多个转义
TEST_CASE("String literals - basic");
TEST_TOKEN("\"hello\"", SCC_TOK_STRING_LITERAL);
TEST_TOKEN("\"\"", SCC_TOK_STRING_LITERAL);
TEST_TOKEN("\"a b c\"", SCC_TOK_STRING_LITERAL);
TEST_CASE("String literals - escape sequences");
TEST_TOKEN("\"a\\nb\\tc\"", SCC_TOK_STRING_LITERAL);
TEST_TOKEN("\"\\\\ \\\" \\' \\?\"", SCC_TOK_STRING_LITERAL);
TEST_TOKEN("\"\\123\\xAB\"", SCC_TOK_STRING_LITERAL);
TEST_CASE("String literals - wide and UTF-8 prefixes (C11)");
TEST_TOKEN("L\"wide\"", SCC_TOK_STRING_LITERAL);
TEST_TOKEN("u\"utf16\"", SCC_TOK_STRING_LITERAL);
TEST_TOKEN("U\"utf32\"", SCC_TOK_STRING_LITERAL);
TEST_TOKEN("u8\"utf8\"", SCC_TOK_STRING_LITERAL);
}
void test_whitespace() {
TEST_CASE("Whitespace characters");
TEST_TOKEN(" ", SCC_TOK_BLANK);
TEST_TOKEN("\t", SCC_TOK_BLANK);
TEST_TOKEN("\v", SCC_TOK_BLANK);
TEST_TOKEN("\f", SCC_TOK_BLANK);
TEST_TOKEN(" \t\v\f", SCC_TOK_BLANK); // 连续空白应为一个 token
}
void test_newlines() {
TEST_CASE("Newline characters");
TEST_TOKEN("\n", SCC_TOK_ENDLINE);
TEST_TOKEN("\r", SCC_TOK_ENDLINE);
TEST_TOKEN("\r\n", SCC_TOK_ENDLINE); // 应视为单个换行符
}
void test_comments() {
TEST_CASE("Line comments");
TEST_TOKEN("// single line comment", SCC_TOK_LINE_COMMENT);
TEST_TOKEN("// comment with // inside", SCC_TOK_LINE_COMMENT);
TEST_TOKEN("// comment at end", SCC_TOK_LINE_COMMENT);
TEST_CASE("Block comments");
TEST_TOKEN("/* simple */", SCC_TOK_BLOCK_COMMENT);
TEST_TOKEN("/* multi\nline */", SCC_TOK_BLOCK_COMMENT);
TEST_TOKEN("/**/", SCC_TOK_BLOCK_COMMENT); // 空注释
TEST_TOKEN("/* with * inside */", SCC_TOK_BLOCK_COMMENT);
TEST_TOKEN("/* nested /* not allowed in C */",
SCC_TOK_BLOCK_COMMENT); // 词法上不会嵌套
}
void test_identifiers() {
TEST_CASE("Valid identifiers");
TEST_TOKEN("foo", SCC_TOK_IDENT);
TEST_TOKEN("_foo", SCC_TOK_IDENT);
TEST_TOKEN("foo123", SCC_TOK_IDENT);
TEST_TOKEN("foo_bar", SCC_TOK_IDENT);
TEST_TOKEN("FOO", SCC_TOK_IDENT);
TEST_TOKEN("_", SCC_TOK_IDENT);
TEST_TOKEN("__LINE__", SCC_TOK_IDENT); // 预处理宏名也是标识符
// 超长标识符（假设缓冲区足够）
char long_id[1024];
memset(long_id, 'a', sizeof(long_id) - 1);
long_id[sizeof(long_id) - 1] = '\0';
TEST_TOKEN(long_id, SCC_TOK_IDENT);
}
void test_preprocessor() {
TEST_CASE("Preprocessor directives - just the # token");
TEST_TOKEN("#", SCC_TOK_SHARP);
TEST_TOKEN("##", SCC_TOK_SHARP); // 第一个 # 是 token，第二个 # 将是下一个
// token（在序列测试中验证）
// 多 token 序列测试 #include 等
TEST_SEQUENCE("#include <stdio.h>", SCC_TOK_SHARP, SCC_TOK_IDENT,
SCC_TOK_BLANK, SCC_TOK_LT, SCC_TOK_IDENT, SCC_TOK_DOT,
SCC_TOK_IDENT, SCC_TOK_GT);
TEST_SEQUENCE("#define FOO 123", SCC_TOK_SHARP, SCC_TOK_IDENT,
SCC_TOK_BLANK, SCC_TOK_IDENT, SCC_TOK_BLANK,
SCC_TOK_INT_LITERAL);
}
void test_edge_cases() {
TEST_CASE("Invalid characters");
TEST_TOKEN("@", SCC_TOK_UNKNOWN);
TEST_TOKEN("`", SCC_TOK_UNKNOWN);
TEST_TOKEN("$", SCC_TOK_UNKNOWN); // 在 C 中不是标识符字符
TEST_CASE("Empty input");
TEST_TOKEN("", SCC_TOK_EOF); // 立即 EOF
TEST_CASE("Only whitespace");
TEST_TOKEN(" \t", SCC_TOK_BLANK);
// 之后应该为 EOF，但我们的单 token 测试只取第一个 token
TEST_CASE("Numbers followed by letters (no suffix)");
// 词法上应拆分为数字和标识符
TEST_SEQUENCE("123abc", SCC_TOK_INT_LITERAL, SCC_TOK_IDENT);
TEST_SEQUENCE("0x123xyz", SCC_TOK_INT_LITERAL, SCC_TOK_IDENT);
}
void test_sequences() {
TEST_CASE("Simple expression");
TEST_SEQUENCE("a + b * c", SCC_TOK_IDENT, SCC_TOK_BLANK, SCC_TOK_ADD,
SCC_TOK_BLANK, SCC_TOK_IDENT, SCC_TOK_BLANK, SCC_TOK_MUL,
SCC_TOK_BLANK, SCC_TOK_IDENT);
TEST_CASE("Function call");
TEST_SEQUENCE("func(1, 2);", SCC_TOK_IDENT, SCC_TOK_L_PAREN,
SCC_TOK_INT_LITERAL, SCC_TOK_COMMA, SCC_TOK_BLANK,
SCC_TOK_INT_LITERAL, SCC_TOK_R_PAREN, SCC_TOK_SEMICOLON);
TEST_CASE("Multi-character operators");
TEST_SEQUENCE(">>=", SCC_TOK_ASSIGN_R_SH);
TEST_SEQUENCE("<<=", SCC_TOK_ASSIGN_L_SH);
TEST_SEQUENCE("...", SCC_TOK_ELLIPSIS);
TEST_SEQUENCE("->", SCC_TOK_DEREF);
TEST_SEQUENCE("##", SCC_TOK_SHARP, SCC_TOK_SHARP); // 两个预处理记号
TEST_CASE("Comments and whitespace interleaved");
TEST_SEQUENCE("/* comment */ a // line comment\n b", SCC_TOK_BLOCK_COMMENT,
SCC_TOK_BLANK, SCC_TOK_IDENT, SCC_TOK_BLANK,
SCC_TOK_LINE_COMMENT, SCC_TOK_ENDLINE, SCC_TOK_BLANK,
SCC_TOK_IDENT);
TEST_CASE("String literals with escapes");
TEST_SEQUENCE("\"hello\\nworld\"", SCC_TOK_STRING_LITERAL);
TEST_SEQUENCE(
"L\"wide\"",
SCC_TOK_STRING_LITERAL); // 前缀作为标识符？不，整个是字符串字面量
TEST_CASE("Character literals with escapes");
TEST_SEQUENCE("'\\x41'", SCC_TOK_CHAR_LITERAL);
TEST_SEQUENCE("'\\123'", SCC_TOK_CHAR_LITERAL);
}
void test_error_recovery() {
// 测试未闭合的字符字面量：词法分析器可能继续直到遇到换行或 EOF
// 这里假设它会产生一个 SCC_TOK_CHAR_LITERAL 但包含到结束
// 但标准 C 中未闭合是错误，我们可能返回 UNKNOWN
TEST_CASE("Unterminated character literal");
TEST_TOKEN("'a", SCC_TOK_UNKNOWN); // 取决于实现，可能为 CHAR_LITERAL
// 更可靠的测试：序列中下一个 token 是什么
TEST_SEQUENCE("'a b", SCC_TOK_UNKNOWN,
SCC_TOK_IDENT); // 假设第一个 token 是错误
TEST_CASE("Unterminated string literal");
TEST_TOKEN("\"hello", SCC_TOK_UNKNOWN); // 同样
TEST_CASE("Unterminated block comment");
TEST_SEQUENCE("/* comment",
SCC_TOK_BLOCK_COMMENT); // 直到 EOF，可能仍为注释
}
// ============================ 主测试列表 ============================
TEST_LIST = {
{"operators", test_operators},
{"keywords", test_keywords},
{"literals", test_literals},
{"whitespace", test_whitespace},
{"newlines", test_newlines},
{"comments", test_comments},
{"identifiers", test_identifiers},
{"preprocessor", test_preprocessor},
{"edge_cases", test_edge_cases},
{"sequences", test_sequences},
{"error_recovery", test_error_recovery},
{NULL, NULL},
};