refactor(lex_parser): 重命名libcore为scc_core并重构头文件包含

- 将依赖项从libcore重命名为scc_core
- 更新头文件包含路径从<libcore.h>到<scc_core.h>
- 保持原有功能不变

refactor(lexer): 重命名libcore为scc_core并添加词法流式解析功能

- 将依赖项从libcore重命名为scc_core
- 移除不再需要的scc_lexer_token结构体定义
- 重命名struct cc_lexer为struct scc_lexer
- 添加scc_lexer_stream_t流式解析器相关定义和实现
- 新增lexer_stream.c文件实现流式token缓冲功能

refactor(lexer_log): 重命名logger变量和头文件定义

- 将头文件保护宏从__SMCC_LEXER_LOG_H__改为__SCC_LEXER_LOG_H__
- 将logger变量从__smcc_lexer_log改为__scc_lexer_log
- 更新头文件包含从<libcore.h>到<scc_core.h>

refactor(lexer_token): 重新组织token头文件结构

- 将头文件保护宏从__SMCC_CC_TOKEN_H__改为__SCC_LEXER_TOKEN_H__
- 更新头文件包含从<libcore.h>到<scc_core.h>
- 将scc_lexer_token结构体定义移至该文件

refactor(lexer): 简化token匹配代码格式

- 移除LCC相关的注释内容
- 优化括号符号的token匹配代码格式，使用clang-format控制

refactor(pprocessor): 更新依赖项名称和头文件包含

- 将libcore重命名为scc_core
- 将libutils重命名为scc_utils
- 更新头文件包含路径

refactor(runtime): 重命名libcore为scc_core并重构目录结构

- 将libcore目录重命名为scc_core
- 将libutils目录重命名为scc_utils
- 更新所有相关的头文件包含路径
- 修改cbuild.toml中的包名称
- 更新core_vec.h中的宏定义以支持标准库模式

libs/format/cbuild.toml

@@ -0,0 +1,8 @@
+[package]
+name = "scc_format"
+version = "0.1.0"
+
+dependencies = [
+    { name = "libcore", path = "../../runtime/libcore" },
+    { name = "libutils", path = "../../runtime/libutils" },
+]

libs/format/include/scf.h

@@ -0,0 +1,150 @@
+/**
+ * @file scf.h
+ * @brief scc format (SMF) 头文件
+ *
+ * SCF是一个极简的可执行可链接文件格式，专注于内部链接处理，
+ * 同时提供外部符号导入/导出的抽象接口。
+ */
+
+#ifndef __SCC_FORMAT_H__
+#define __SCC_FORMAT_H__
+
+#include <stddef.h>
+#include <stdint.h>
+
+#define scf_byte_t uint8_t
+#define scf_enum_t uint32_t
+#define scf_size_t uint32_t
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/** SCF魔数 */
+#define SCF_MAGIC "SCF\0"
+
+/** SCF版本号 */
+#define SCF_VERSION 1
+
+/** 架构类型 */
+typedef enum {
+    SCF_ARCH_UNKNOWN = 0,
+    SCF_ARCH_RV32 = 1,
+    SCF_ARCH_RV64 = 2,
+    SCF_ARCH_X86 = 3,
+    SCF_ARCH_X64 = 4,
+} scf_arch_t;
+
+/** 文件标志位 */
+typedef enum {
+    SCF_FLAG_EXECUTABLE = 0x01,  // 可执行文件
+    SCF_FLAG_RELOCATABLE = 0x02, // 可重定位文件
+    SCF_FLAG_EXE_RELOC =
+        SCF_FLAG_EXECUTABLE | SCF_FLAG_RELOCATABLE, // 内部链接后的可执行文件
+} scf_flags_t;
+
+/** 符号类型 */
+typedef enum {
+    SCF_SYM_TYPE_UNDEF = 0,  // 未定义
+    SCF_SYM_TYPE_FUNC = 1,   // 函数
+    SCF_SYM_TYPE_DATA = 2,   // 数据
+    SCF_SYM_TYPE_OBJECT = 3, // 对象
+} scf_sym_type_t;
+
+/** 符号绑定类型 */
+typedef enum {
+    SCF_SYM_BIND_LOCAL = 0,  // 局部符号
+    SCF_SYM_BIND_GLOBAL = 1, // 全局符号
+    SCF_SYM_BIND_WEAK = 2,   // 弱引用
+} scf_sym_bind_t;
+
+/** 符号可见性 */
+typedef enum {
+    SCF_SYM_VIS_DEFAULT = 0,   // 默认可见性
+    SCF_SYM_VIS_HIDDEN = 1,    // 隐藏
+    SCF_SYM_VIS_PROTECTED = 2, // 受保护
+} scf_sym_vis_t;
+
+/** 段类型 */
+typedef enum {
+    SCF_SECT_NONE = 0,   // 无
+    SCF_SECT_CODE = 1,   // 代码段
+    SCF_SECT_DATA = 2,   // 数据段
+    SCF_SECT_BSS = 3,    // BSS段（未初始化数据）
+    SCF_SECT_RODATA = 4, // 只读数据
+} scf_sect_type_t;
+
+/** 重定位类型 */
+typedef enum {
+    SCF_RELOC_ABS = 1, // 绝对地址
+    SCF_RELOC_REL = 2, // 相对地址
+    SCF_RELOC_PC = 3,  // PC相对
+} scf_reloc_type_t;
+
+/**
+ * @brief SCF文件头
+ */
+typedef struct {
+    scf_byte_t magic[4];    // 魔数: "SCF\0"
+    scf_enum_t type;        // 类型
+    scf_enum_t version;     // 版本号
+    scf_enum_t arch;        // 架构
+    scf_enum_t flags;       // 标志位
+    scf_size_t entry_point; // 入口点地址
+    scf_size_t data_size;
+    scf_size_t code_size;
+    scf_size_t strtab_size;
+    scf_size_t sym_count;
+    scf_size_t reloc_count;
+} scf_header_t;
+
+/**
+ * @brief SCF段
+ */
+typedef struct {
+    scf_size_t size;
+    scf_enum_t scf_sect_type;
+    scf_byte_t data[1];
+} scf_sect_t;
+
+/**
+ * @brief SCF符号表
+ */
+typedef struct {
+    scf_size_t name_offset;
+    scf_enum_t scf_sym_type;
+    scf_enum_t scf_sym_bind;
+    scf_enum_t scf_sym_vis;
+    scf_enum_t scf_sect_type;
+    scf_size_t scf_sect_offset;
+    scf_size_t scf_sym_size;
+} scf_sym_t;
+
+/**
+ * @brief SCF重定向条目
+ */
+typedef struct {
+    scf_size_t offset;    // 在段中的偏移量
+    scf_size_t sym_idx;   // 符号索引
+    scf_enum_t type;      // 重定位类型
+    scf_enum_t sect_type; // 段类型（代码段/数据段）
+    scf_size_t addend;    // 加数
+} scf_reloc_t;
+
+/*
+scf 文件结构
+
+scf_header_t header;
+scf_sect_data_t text;
+scf_sect_data_t data;
+scf_sect_data_t symtab;
+scf_sect_data_t reloc;
+scf_sect_data_t strtab;
+
+*/
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* __SCC_FORMAT_H__ */

libs/format/include/scf_impl.h

@@ -0,0 +1,42 @@
+#ifndef __SCC_FORMAT_IMPL_H__
+#define __SCC_FORMAT_IMPL_H__
+
+#include <libcore.h>
+#include <libutils.h>
+#include <scf.h>
+
+typedef SCC_VEC(u8) scf_sect_data_t;
+typedef SCC_VEC(scf_sym_t) scf_sym_vec_t;
+typedef SCC_VEC(scf_reloc_t) scf_reloc_vec_t;
+
+typedef struct {
+    scf_header_t header;
+    scf_sect_data_t text;
+    scf_sect_data_t data;
+    scf_sect_data_t symtab;
+    scf_sect_data_t reloc;
+    scf_sect_data_t strtab;
+
+    scc_strpool_t strpool;
+    scc_hashtable_t str2idx;
+    scf_sym_vec_t syms;
+    scf_reloc_vec_t relocs;
+} scf_t;
+
+void scf_init(scf_t *scf);
+cbool scf_parse(scf_t *scf, const char *buffer, usize size);
+cbool scf_write(scf_t *scf, char *buffer, usize size);
+
+cbool scf_exchange_section(scf_t *scf, scf_sect_type_t type,
+                           scf_sect_data_t **section);
+cbool scf_add_sym(scf_t *scf, scf_sym_t *sym);
+cbool scf_add_reloc(scf_t *scf, scf_reloc_t *reloc);
+cbool scf_apply_relocations(scf_t *scf);
+cbool scf_write_done(scf_t *scf); // 在写入前进行内部整理
+
+cbool scf_check_valid(scf_t *scf);
+
+typedef SCC_VEC(scf_t) scf_vec_t;
+cbool scf_link_all(scf_vec_t scfs, scf_t *outscf);
+
+#endif /* __SCC_FORMAT_IMPL_H__ */

libs/format/src/scf.c

@@ -0,0 +1,469 @@
+/**
+ * @file scf.c
+ * @brief SCF 格式实现
+ */
+
+#include <scf_impl.h>
+
+/**
+ * @brief 初始化 SCF 结构
+ * @param scf 指向 scf_t 结构的指针
+ */
+void scf_init(scf_t *scf) {
+    if (!scf) {
+        return;
+    }
+
+    scc_memset(&scf->header, 0, sizeof(scf->header));
+    scc_memcpy(scf->header.magic, SCF_MAGIC, 4);
+    scf->header.version = SCF_VERSION;
+    scf->header.arch = SCF_ARCH_UNKNOWN;
+    scf->header.flags = 0;
+    scf->header.entry_point = 0;
+    scf->header.data_size = 0;
+    scf->header.code_size = 0;
+    scf->header.strtab_size = 0;
+    scf->header.sym_count = 0;
+    scf->header.reloc_count = 0;
+
+    scc_vec_init(scf->text);
+    scc_vec_init(scf->data);
+    scc_vec_init(scf->symtab);
+    scc_vec_init(scf->reloc);
+    scc_vec_init(scf->strtab);
+
+    scc_strpool_init(&scf->strpool);
+    scf->str2idx.hash_func = (void *)scc_strhash32;
+    scf->str2idx.key_cmp = (void *)scc_strcmp;
+    scc_hashtable_init(&scf->str2idx);
+
+    scc_vec_init(scf->syms);
+    scc_vec_init(scf->relocs);
+}
+
+/**
+ * @brief 从缓冲区读取并解析 SCF 数据
+ * @param scf 指向 scf_t 结构的指针
+ * @param buffer 输入缓冲区
+ * @param size 缓冲区大小
+ * @return 成功返回 true，失败返回 false
+ */
+cbool scf_parse(scf_t *scf, const char *buffer, usize size) {
+    if (!scf || !buffer || size < sizeof(scf_header_t)) {
+        return false;
+    }
+
+    // 读取头部
+    const scf_header_t *header = (const scf_header_t *)buffer;
+    if (scc_memcmp(header->magic, SCF_MAGIC, 4) != 0) {
+        return false;
+    }
+
+    scf->header = *header;
+
+    // 计算各段偏移
+    usize offset = sizeof(scf_header_t);
+
+    // 读取 text 段
+    if (scf->header.code_size > 0) {
+        if (offset + scf->header.code_size > size) {
+            return false;
+        }
+        // 调整 text 向量大小
+        while (scf->text.size < scf->header.code_size) {
+            scc_vec_push(scf->text, 0);
+        }
+        scc_memcpy(scf->text.data, buffer + offset, scf->header.code_size);
+        offset += scf->header.code_size;
+    }
+
+    // 读取 data 段
+    if (scf->header.data_size > 0) {
+        if (offset + scf->header.data_size > size) {
+            return false;
+        }
+        // 调整 data 向量大小
+        while (scf->data.size < scf->header.data_size) {
+            scc_vec_push(scf->data, 0);
+        }
+        scc_memcpy(scf->data.data, buffer + offset, scf->header.data_size);
+        offset += scf->header.data_size;
+    }
+
+    // 读取符号表
+    if (scf->header.sym_count > 0) {
+        usize symtab_size = scf->header.sym_count * sizeof(scf_sym_t);
+        if (offset + symtab_size > size) {
+            return false;
+        }
+        // 调整 syms 向量大小
+        while (scf->syms.size < scf->header.sym_count) {
+            scf_sym_t sym = {0};
+            scc_vec_push(scf->syms, sym);
+        }
+        scc_memcpy(scf->syms.data, buffer + offset, symtab_size);
+        offset += symtab_size;
+    }
+
+    // 读取重定位表
+    if (scf->header.reloc_count > 0) {
+        usize reloc_size = scf->header.reloc_count * sizeof(scf_reloc_t);
+        if (offset + reloc_size > size) {
+            return false;
+        }
+        // 调整 relocs 向量大小
+        while (scf->relocs.size < scf->header.reloc_count) {
+            scf_reloc_t reloc = {0};
+            scc_vec_push(scf->relocs, reloc);
+        }
+        scc_memcpy(scf->relocs.data, buffer + offset, reloc_size);
+        offset += reloc_size;
+    }
+
+    // 读取字符串表
+    if (scf->header.strtab_size > 0) {
+        if (offset + scf->header.strtab_size > size) {
+            return false;
+        }
+        // 调整 strtab 向量大小
+        while (scf->strtab.size < scf->header.strtab_size) {
+            scc_vec_push(scf->strtab, 0);
+        }
+        scc_memcpy(scf->strtab.data, buffer + offset, scf->header.strtab_size);
+        offset += scf->header.strtab_size;
+    }
+    // 允许 offset <= size，因为缓冲区可能比实际数据大
+    if (offset > size) {
+        return false;
+    }
+
+    return true;
+}
+
+/**
+ * @brief 将 SCF 数据写入缓冲区
+ * @param scf 指向 scf_t 结构的指针
+ * @param buffer 输出缓冲区
+ * @param size 缓冲区大小
+ * @return 成功返回 true，失败返回 false
+ */
+cbool scf_write(scf_t *scf, char *buffer, usize size) {
+    if (!scf || !buffer) {
+        return false;
+    }
+
+    // 计算所需大小
+    usize needed = sizeof(scf_header_t);
+    needed += scf->header.code_size; // text 段
+    needed += scf->header.data_size; // data 段
+    needed += scf->header.sym_count * sizeof(scf_sym_t);
+    needed += scf->header.reloc_count * sizeof(scf_reloc_t);
+    needed += scf->header.strtab_size;
+
+    if (size < needed) {
+        return false;
+    }
+
+    // 写入头部
+    scf_header_t *header = (scf_header_t *)buffer;
+    *header = scf->header;
+    usize offset = sizeof(scf_header_t);
+
+    // 写入 text 段
+    if (scf->header.code_size > 0) {
+        scc_memcpy(buffer + offset, scf->text.data, scf->header.code_size);
+        offset += scf->header.code_size;
+    }
+
+    // 写入 data 段
+    if (scf->header.data_size > 0) {
+        scc_memcpy(buffer + offset, scf->data.data, scf->header.data_size);
+        offset += scf->header.data_size;
+    }
+
+    // 写入符号表
+    if (scf->header.sym_count > 0) {
+        usize symtab_size = scf->header.sym_count * sizeof(scf_sym_t);
+        scc_memcpy(buffer + offset, scf->syms.data, symtab_size);
+        offset += symtab_size;
+    }
+
+    // 写入重定位表
+    if (scf->header.reloc_count > 0) {
+        usize reloc_size = scf->header.reloc_count * sizeof(scf_reloc_t);
+        scc_memcpy(buffer + offset, scf->relocs.data, reloc_size);
+        offset += reloc_size;
+    }
+
+    // 写入字符串表
+    if (scf->header.strtab_size > 0) {
+        scc_memcpy(buffer + offset, scf->strtab.data, scf->header.strtab_size);
+        offset += scf->header.strtab_size;
+    }
+
+    Assert(offset <= size);
+
+    return true;
+}
+
+/**
+ * @brief 交换段数据
+ * @param scf 指向 scf_t 结构的指针
+ * @param type 段类型
+ * @param section 指向新段数据的指针（双向交换）
+ * @return 成功返回 true，失败返回 false
+ */
+cbool scf_exchange_section(scf_t *scf, scf_sect_type_t type,
+                           scf_sect_data_t **section) {
+    if (!scf || !section || !*section) {
+        return false;
+    }
+
+    scf_sect_data_t *target_section = NULL;
+    scf_size_t *size_field = NULL;
+
+    // 根据类型选择目标段和大小字段
+    switch (type) {
+    case SCF_SECT_CODE:
+        target_section = &scf->text;
+        size_field = &scf->header.code_size;
+        break;
+    case SCF_SECT_DATA:
+        target_section = &scf->data;
+        size_field = &scf->header.data_size;
+        break;
+    case SCF_SECT_RODATA:
+        // 当前实现中没有单独的rodata段，使用data段
+        target_section = &scf->data;
+        size_field = &scf->header.data_size;
+        break;
+    case SCF_SECT_BSS:
+        // BSS段不存储数据，只记录大小
+        // 这里暂时不支持交换BSS段
+        return false;
+    default:
+        return false;
+    }
+
+    // 交换段数据
+    scf_sect_data_t temp = *target_section;
+    *target_section = **section;
+    **section = temp;
+
+    // 更新大小字段
+    if (size_field) {
+        // 保存旧大小（如果需要的话）
+        // scf_size_t temp_size = *size_field; // 未使用，注释掉
+        *size_field = (scf_size_t)(*target_section).size;
+        // 如果调用者需要知道新的大小，可以在这里设置
+        // 但当前API没有提供这个功能
+    }
+
+    return true;
+}
+
+/**
+ * @brief 添加符号到符号表
+ * @param scf 指向 scf_t 结构的指针
+ * @param sym 指向要添加的符号的指针
+ * @return 成功返回 true，失败返回 false
+ */
+cbool scf_add_sym(scf_t *scf, scf_sym_t *sym) {
+    if (!scf || !sym) {
+        return false;
+    }
+
+    // 添加到符号向量
+    scc_vec_push(scf->syms, *sym);
+    scf->header.sym_count++;
+
+    return true;
+}
+
+/**
+ * @brief 添加重定位条目
+ * @param scf 指向 scf_t 结构的指针
+ * @param reloc 指向要添加的重定位条目的指针
+ * @return 成功返回 true，失败返回 false
+ */
+cbool scf_add_reloc(scf_t *scf, scf_reloc_t *reloc) {
+    if (!scf || !reloc) {
+        return false;
+    }
+
+    // 添加到重定位向量
+    scc_vec_push(scf->relocs, *reloc);
+    scf->header.reloc_count++;
+
+    return true;
+}
+
+/**
+ * @brief 检查 SCF 结构是否有效
+ * @param scf 指向 scf_t 结构的指针
+ * @return 有效返回 true，无效返回 false
+ */
+cbool scf_check_valid(scf_t *scf) {
+    if (!scf) {
+        return false;
+    }
+
+    // 检查魔数
+    if (scc_memcmp(scf->header.magic, SCF_MAGIC, 4) != 0) {
+        return false;
+    }
+
+    // 检查版本
+    if (scf->header.version != SCF_VERSION) {
+        return false;
+    }
+
+    // 检查架构
+    if (scf->header.arch > SCF_ARCH_X64) {
+        return false;
+    }
+
+    // 检查各段大小是否一致
+    if (scf->header.code_size != scf->text.size) {
+        return false;
+    }
+    if (scf->header.data_size != scf->data.size) {
+        return false;
+    }
+    if (scf->header.sym_count != scf->syms.size) {
+        return false;
+    }
+    if (scf->header.reloc_count != scf->relocs.size) {
+        return false;
+    }
+    if (scf->header.strtab_size != scf->strtab.size) {
+        return false;
+    }
+
+    return true;
+}
+
+/**
+ * @brief 应用重定位到段数据
+ * @param scf 指向 scf_t 结构的指针
+ * @return 成功返回 true，失败返回 false
+ */
+cbool scf_apply_relocations(scf_t *scf) {
+    if (!scf) {
+        return false;
+    }
+
+    // 遍历所有重定位条目
+    for (usize i = 0; i < scf->relocs.size; i++) {
+        scf_reloc_t *reloc = &scf->relocs.data[i];
+
+        // 根据段类型选择目标段
+        scf_sect_data_t *target_section = NULL;
+        switch (reloc->sect_type) {
+        case SCF_SECT_CODE:
+            target_section = &scf->text;
+            break;
+        case SCF_SECT_DATA:
+        case SCF_SECT_RODATA:
+            target_section = &scf->data;
+            break;
+        default:
+            // 不支持的段类型
+            continue;
+        }
+
+        // 检查偏移量是否有效
+        if (reloc->offset + sizeof(scf_size_t) > target_section->size) {
+            // 偏移量超出段范围
+            return false;
+        }
+
+        // 获取符号地址
+        scf_size_t symbol_address = 0;
+        if (reloc->sym_idx < scf->syms.size) {
+            scf_sym_t *sym = &scf->syms.data[reloc->sym_idx];
+            symbol_address = sym->scf_sect_offset;
+        }
+
+        // 计算重定位值
+        scf_size_t relocation_value = symbol_address + reloc->addend;
+
+        // 根据重定位类型应用
+        scf_size_t *target =
+            (scf_size_t *)(target_section->data + reloc->offset);
+        switch (reloc->type) {
+        case SCF_RELOC_ABS:
+            // 绝对地址：直接替换
+            *target = relocation_value;
+            break;
+        case SCF_RELOC_REL:
+            // 相对地址：计算相对于当前位置的偏移
+            // 使用 uintptr_t 进行安全的指针到整数转换
+            *target = relocation_value - (scf_size_t)(uintptr_t)target;
+            break;
+        case SCF_RELOC_PC:
+            // PC相对：计算相对于PC的偏移
+            // 使用 uintptr_t 进行安全的指针到整数转换
+            *target = relocation_value - (scf_size_t)(uintptr_t)(target + 1);
+            break;
+        default:
+            // 不支持的重定位类型
+            return false;
+        }
+    }
+
+    return true;
+}
+
+/**
+ * @brief 在写入前进行内部整理
+ * @param scf 指向 scf_t 结构的指针
+ * @return 成功返回 true，失败返回 false
+ */
+cbool scf_write_done(scf_t *scf) {
+    if (!scf) {
+        return false;
+    }
+
+    // 应用所有重定位
+    if (!scf_apply_relocations(scf)) {
+        return false;
+    }
+
+    // 更新头部中的大小字段
+    scf->header.code_size = (scf_size_t)scf->text.size;
+    scf->header.data_size = (scf_size_t)scf->data.size;
+    scf->header.sym_count = (scf_size_t)scf->syms.size;
+    scf->header.reloc_count = (scf_size_t)scf->relocs.size;
+    scf->header.strtab_size = (scf_size_t)scf->strtab.size;
+
+    // 设置标志位为内部链接后的可执行文件
+    scf->header.flags |= SCF_FLAG_EXE_RELOC;
+
+    return true;
+}
+
+/**
+ * @brief 链接多个 SCF 文件
+ * @param scfs 包含多个 SCF 文件的向量
+ * @param outscf 输出链接后的 SCF 文件
+ * @return 成功返回 true，失败返回 false
+ */
+cbool scf_link_all(scf_vec_t scfs, scf_t *outscf) {
+    if (!outscf || scfs.size == 0) {
+        return false;
+    }
+
+    // 初始化输出 SCF
+    scf_init(outscf);
+
+    // 简单实现：只链接第一个文件
+    // 实际实现应该合并所有文件的段、解析符号引用、应用重定位等
+    if (scfs.size > 0) {
+        scf_t *first = &scfs.data[0];
+        // 这里应该进行深拷贝，但为了简单起见，我们只复制头部
+        outscf->header = first->header;
+    }
+
+    return true;
+}

libs/format/tests/test_scf.c

@@ -0,0 +1,109 @@
+/**
+ * @file test_scf.c
+ * @brief SCF format tests
+ */
+
+#include <scf_impl.h>
+#include <stdio.h>
+#include <string.h>
+
+int main() {
+    printf("Testing SCF format implementation...\n");
+
+    // Test 1: Initialization
+    scf_t scf;
+    scf_init(&scf);
+
+    if (memcmp(scf.header.magic, SCF_MAGIC, 4) != 0) {
+        printf("FAIL: Magic number incorrect\n");
+        return 1;
+    }
+
+    if (scf.header.version != SCF_VERSION) {
+        printf("FAIL: Version incorrect\n");
+        return 1;
+    }
+
+    if (scf.header.arch != SCF_ARCH_UNKNOWN) {
+        printf("FAIL: Architecture incorrect\n");
+        return 1;
+    }
+
+    printf("Test 1 PASSED: Initialization successful\n");
+
+    // Test 2: Adding symbols
+    scf_sym_t sym = {0};
+    sym.name_offset = 0;
+    sym.scf_sym_type = SCF_SYM_TYPE_FUNC;
+    sym.scf_sym_bind = SCF_SYM_BIND_GLOBAL;
+    sym.scf_sym_vis = SCF_SYM_VIS_DEFAULT;
+    sym.scf_sect_type = SCF_SECT_CODE;
+    sym.scf_sect_offset = 0;
+    sym.scf_sym_size = 16;
+
+    if (!scf_add_sym(&scf, &sym)) {
+        printf("FAIL: Cannot add symbol\n");
+        return 1;
+    }
+
+    if (scf.header.sym_count != 1) {
+        printf("FAIL: Symbol count incorrect\n");
+        return 1;
+    }
+
+    printf("Test 2 PASSED: Symbol addition successful\n");
+
+    // Test 3: Adding relocations
+    scf_reloc_t reloc = {0};
+    reloc.offset = 0; // 偏移量
+    reloc.sym_idx = 0;
+    reloc.type = SCF_RELOC_ABS;
+    reloc.sect_type = SCF_SECT_CODE; // 代码段
+    reloc.addend = 0;
+
+    if (!scf_add_reloc(&scf, &reloc)) {
+        printf("FAIL: Cannot add relocation\n");
+        return 1;
+    }
+
+    if (scf.header.reloc_count != 1) {
+        printf("FAIL: Relocation count incorrect\n");
+        return 1;
+    }
+
+    printf("Test 3 PASSED: Relocation addition successful\n");
+
+    // Test 4: Checking validity
+    if (!scf_check_valid(&scf)) {
+        printf("FAIL: SCF structure invalid\n");
+        return 1;
+    }
+
+    printf("Test 4 PASSED: SCF structure valid\n");
+
+    // Test 5: Writing and reading
+    char buffer[1024];
+    if (!scf_write(&scf, buffer, sizeof(buffer))) {
+        printf("FAIL: Cannot write to buffer\n");
+        return 1;
+    }
+
+    scf_t scf2;
+    scf_init(&scf2);
+
+    if (!scf_parse(&scf2, buffer, sizeof(buffer))) {
+        printf("FAIL: Cannot read from buffer\n");
+        return 1;
+    }
+
+    // Compare the two structures
+    if (memcmp(&scf.header, &scf2.header, sizeof(scf_header_t)) != 0) {
+        printf("FAIL: Header mismatch\n");
+        return 1;
+    }
+
+    printf("Test 5 PASSED: Write/read successful\n");
+
+    printf("All tests passed!\n");
+    return 0;
+}

libs/format/tests/test_scf_x64.c

@@ -0,0 +1,184 @@
+/**
+ * @file test_scf_x64.c
+ * @brief SCF x64 architecture tests
+ *
+ * This test creates a simple x64 executable similar to the Rust example
+ * provided by the user.
+ */
+
+#include <scf_impl.h>
+#include <stdio.h>
+#include <string.h>
+
+int main() {
+    printf("Testing SCF x64 format implementation...\n");
+
+    // Test 1: Initialize SCF for x64 architecture
+    scf_t scf;
+    scf_init(&scf);
+
+    // Set architecture to x64
+    scf.header.arch = SCF_ARCH_X64;
+    scf.header.flags = SCF_FLAG_EXE_RELOC;
+
+    printf("Test 1 PASSED: x64 initialization successful\n");
+
+    // Test 2: Add .text section with x64 machine code
+    // x64 machine code for:
+    //   sub rsp, 0x28
+    //   lea rcx, [rip + data_offset]  ; will be relocated
+    //   call [rip + printf_iat]       ; will be relocated
+    //   add rsp, 0x28
+    //   xor eax, eax
+    //   ret
+    unsigned char x64_code[] = {
+        0x48, 0x83, 0xEC, 0x28, // sub rsp, 0x28
+        0x48, 0x8D, 0x0D, 0x00, 0x00, 0x00,
+        0x00, // lea rcx, [rip + 0] (to be relocated)
+        0xFF, 0x15, 0x00, 0x00, 0x00, 0x00, // call [rip + 0] (to be relocated)
+        0x48, 0x83, 0xC4, 0x28,             // add rsp, 0x28
+        0x33, 0xC0,                         // xor eax, eax
+        0xC3                                // ret
+    };
+
+    // Add code to .text section
+    for (usize i = 0; i < sizeof(x64_code); i++) {
+        scc_vec_push(scf.text, x64_code[i]);
+    }
+    scf.header.code_size = (scf_size_t)scf.text.size;
+
+    printf("Test 2 PASSED: x64 code added to .text section\n");
+
+    // Test 3: Add .data section with string
+    const char hello_world[] = "Hello, World from SCF x64 Test!\n\0";
+    for (usize i = 0; i < sizeof(hello_world); i++) {
+        scc_vec_push(scf.data, hello_world[i]);
+    }
+    scf.header.data_size = (scf_size_t)scf.data.size;
+
+    printf("Test 3 PASSED: Data string added to .data section\n");
+
+    // Test 4: Add symbols
+    scf_sym_t data_sym = {0};
+    data_sym.name_offset = 0; // Would need string table for actual names
+    data_sym.scf_sym_type = SCF_SYM_TYPE_DATA;
+    data_sym.scf_sym_bind = SCF_SYM_BIND_GLOBAL;
+    data_sym.scf_sym_vis = SCF_SYM_VIS_DEFAULT;
+    data_sym.scf_sect_type = SCF_SECT_DATA;
+    data_sym.scf_sect_offset = 0; // Start of data section
+    data_sym.scf_sym_size = sizeof(hello_world);
+
+    if (!scf_add_sym(&scf, &data_sym)) {
+        printf("FAIL: Cannot add data symbol\n");
+        return 1;
+    }
+
+    scf_sym_t code_sym = {0};
+    code_sym.name_offset = 0;
+    code_sym.scf_sym_type = SCF_SYM_TYPE_FUNC;
+    code_sym.scf_sym_bind = SCF_SYM_BIND_GLOBAL;
+    code_sym.scf_sym_vis = SCF_SYM_VIS_DEFAULT;
+    code_sym.scf_sect_type = SCF_SECT_CODE;
+    code_sym.scf_sect_offset = 0; // Start of code section
+    code_sym.scf_sym_size = sizeof(x64_code);
+
+    if (!scf_add_sym(&scf, &code_sym)) {
+        printf("FAIL: Cannot add code symbol\n");
+        return 1;
+    }
+
+    printf("Test 4 PASSED: Symbols added\n");
+
+    // Test 5: Add relocations
+    // First relocation: data reference at offset 7 in code (lea rcx, [rip +
+    // data_offset])
+    scf_reloc_t data_reloc = {0};
+    data_reloc.offset = 7;          // Offset in code section
+    data_reloc.sym_idx = 0;         // Index of data symbol (first symbol added)
+    data_reloc.type = SCF_RELOC_PC; // PC-relative relocation
+    data_reloc.sect_type = SCF_SECT_CODE;
+    data_reloc.addend = -4; // RIP-relative addressing adjustment
+
+    if (!scf_add_reloc(&scf, &data_reloc)) {
+        printf("FAIL: Cannot add data relocation\n");
+        return 1;
+    }
+
+    // Second relocation: external function reference at offset 13 in code (call
+    // [rip + printf_iat])
+    scf_reloc_t func_reloc = {0};
+    func_reloc.offset = 13; // Offset in code section
+    func_reloc.sym_idx =
+        1; // Index of code symbol (would be external in real case)
+    func_reloc.type = SCF_RELOC_PC; // PC-relative relocation
+    func_reloc.sect_type = SCF_SECT_CODE;
+    func_reloc.addend = -4; // RIP-relative addressing adjustment
+
+    if (!scf_add_reloc(&scf, &func_reloc)) {
+        printf("FAIL: Cannot add function relocation\n");
+        return 1;
+    }
+
+    printf("Test 5 PASSED: Relocations added\n");
+
+    // Test 6: Apply relocations
+    if (!scf_apply_relocations(&scf)) {
+        printf("FAIL: Cannot apply relocations\n");
+        return 1;
+    }
+
+    printf("Test 6 PASSED: Relocations applied\n");
+
+    // Test 7: Prepare for writing (internal整理)
+    if (!scf_write_done(&scf)) {
+        printf("FAIL: Cannot prepare for writing\n");
+        return 1;
+    }
+
+    printf("Test 7 PASSED: Prepared for writing\n");
+
+    // Test 8: Write to buffer
+    char buffer[4096];
+    if (!scf_write(&scf, buffer, sizeof(buffer))) {
+        printf("FAIL: Cannot write to buffer\n");
+        return 1;
+    }
+
+    printf("Test 8 PASSED: Written to buffer\n");
+
+    // Test 9: Parse from buffer
+    scf_t scf2;
+    scf_init(&scf2);
+
+    if (!scf_parse(&scf2, buffer, sizeof(buffer))) {
+        printf("FAIL: Cannot parse from buffer\n");
+        return 1;
+    }
+
+    // Verify architecture
+    if (scf2.header.arch != SCF_ARCH_X64) {
+        printf("FAIL: Architecture not preserved\n");
+        return 1;
+    }
+
+    printf("Test 9 PASSED: Parsed from buffer, architecture preserved\n");
+
+    // Test 10: Verify structure
+    if (!scf_check_valid(&scf2)) {
+        printf("FAIL: Parsed structure invalid\n");
+        return 1;
+    }
+
+    printf("Test 10 PASSED: Parsed structure valid\n");
+
+    printf("\nAll x64 tests passed!\n");
+    printf("Created SCF file with:\n");
+    printf("  Architecture: x64\n");
+    printf("  Code size: %u bytes\n", scf.header.code_size);
+    printf("  Data size: %u bytes\n", scf.header.data_size);
+    printf("  Symbols: %u\n", scf.header.sym_count);
+    printf("  Relocations: %u\n", scf.header.reloc_count);
+    printf("  Flags: 0x%x\n", scf.header.flags);
+
+    return 0;
+}

libs/lex_parser/cbuild.toml

@@ -1,5 +1,5 @@
 [package]
-name = "smcc_lex_parser"
+name = "scc_lex_parser"
 version = "0.1.0"
 
-dependencies = [{ name = "libcore", path = "../../runtime/libcore" }]
+dependencies = [{ name = "scc_core", path = "../../runtime/scc_core" }]

libs/lex_parser/include/lex_parser.h

@@ -1,26 +1,32 @@
-#ifndef __SMCC_LEX_PARSER_H__
-#define __SMCC_LEX_PARSER_H__
+#ifndef __SCC_LEX_PARSER_H__
+#define __SCC_LEX_PARSER_H__
 
-#include <libcore.h>
+#include <scc_core.h>
 
-static inline cbool lex_parse_is_endline(int ch) {
+static inline cbool scc_lex_parse_is_endline(int ch) {
     return ch == '\n' || ch == '\r';
 }
 
-static inline cbool lex_parse_is_whitespace(int ch) {
+static inline cbool scc_lex_parse_is_whitespace(int ch) {
     return ch == ' ' || ch == '\t';
 }
 
-int lex_parse_char(scc_probe_stream_t *input, scc_pos_t *pos);
-cbool lex_parse_string(scc_probe_stream_t *input, scc_pos_t *pos,
-                       scc_cstring_t *output);
-cbool lex_parse_number(scc_probe_stream_t *input, scc_pos_t *pos,
-                       usize *output);
-cbool lex_parse_identifier(scc_probe_stream_t *input, scc_pos_t *pos,
-                           scc_cstring_t *output);
-void lex_parse_skip_endline(scc_probe_stream_t *input, scc_pos_t *pos);
-void lex_parse_skip_block_comment(scc_probe_stream_t *input, scc_pos_t *pos);
-void lex_parse_skip_line(scc_probe_stream_t *input, scc_pos_t *pos);
-void lex_parse_skip_whitespace(scc_probe_stream_t *input, scc_pos_t *pos);
+// TODO identifier check is right?
+static inline cbool scc_lex_parse_is_identifier_prefix(int ch) {
+    return (ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z') || ch == '_';
+}
 
-#endif /* __SMCC_LEX_PARSER_H__ */
+int scc_lex_parse_char(scc_probe_stream_t *input, scc_pos_t *pos);
+cbool scc_lex_parse_string(scc_probe_stream_t *input, scc_pos_t *pos,
+                           scc_cstring_t *output);
+cbool scc_lex_parse_number(scc_probe_stream_t *input, scc_pos_t *pos,
+                           usize *output);
+cbool scc_lex_parse_identifier(scc_probe_stream_t *input, scc_pos_t *pos,
+                               scc_cstring_t *output);
+void scc_lex_parse_skip_endline(scc_probe_stream_t *input, scc_pos_t *pos);
+void scc_lex_parse_skip_block_comment(scc_probe_stream_t *input,
+                                      scc_pos_t *pos);
+void scc_lex_parse_skip_line(scc_probe_stream_t *input, scc_pos_t *pos);
+void scc_lex_parse_skip_whitespace(scc_probe_stream_t *input, scc_pos_t *pos);
+
+#endif /* __SCC_LEX_PARSER_H__ */

libs/lex_parser/src/lex_parser.c

@@ -1,19 +1,19 @@
 #include <lex_parser.h>
 
-void lex_parse_skip_endline(scc_probe_stream_t *input, scc_pos_t *pos) {
+void scc_lex_parse_skip_endline(scc_probe_stream_t *input, scc_pos_t *pos) {
     Assert(input != null && pos != null);
-    scc_probe_stream_reset(input);
+    // scc_probe_stream_reset(input);
     int ch = scc_probe_stream_peek(input);
     if (ch == '\r') {
-        scc_probe_stream_consume(input);
+        scc_probe_stream_next(input);
         ch = scc_probe_stream_peek(input);
         if (ch == '\n') {
-            scc_probe_stream_consume(input);
+            scc_probe_stream_next(input);
         }
-        core_pos_next_line(pos);
+        scc_pos_next_line(pos);
     } else if (ch == '\n') {
-        scc_probe_stream_consume(input);
-        core_pos_next_line(pos);
+        scc_probe_stream_next(input);
+        scc_pos_next_line(pos);
     } else {
         LOG_WARN("not a newline character");
     }
@@ -57,81 +57,82 @@ static inline int got_simple_escape(int ch) {
     /* clang-format on */
 }
 
-void lex_parse_skip_line(scc_probe_stream_t *input, scc_pos_t *pos) {
+void scc_lex_parse_skip_line(scc_probe_stream_t *input, scc_pos_t *pos) {
     scc_probe_stream_t *stream = input;
     Assert(stream != null && pos != null);
-    scc_probe_stream_reset(stream);
+    // scc_probe_stream_reset(stream);
     while (1) {
         int ch = scc_probe_stream_peek(stream);
 
-        if (ch == core_stream_eof) {
+        if (ch == scc_stream_eof) {
             return;
         }
 
         // TODO endline
-        if (lex_parse_is_endline(ch)) {
-            lex_parse_skip_endline(stream, pos);
+        if (scc_lex_parse_is_endline(ch)) {
+            scc_lex_parse_skip_endline(stream, pos);
             return;
         } else {
-            scc_probe_stream_consume(stream);
-            core_pos_next(pos);
+            scc_probe_stream_next(stream);
+            scc_pos_next(pos);
         }
     }
 }
 
-void lex_parse_skip_block_comment(scc_probe_stream_t *input, scc_pos_t *pos) {
+void scc_lex_parse_skip_block_comment(scc_probe_stream_t *input,
+                                      scc_pos_t *pos) {
     scc_probe_stream_t *stream = input;
     Assert(stream != null && pos != null);
     int ch;
-    scc_probe_stream_reset(stream);
-    ch = scc_probe_stream_consume(stream);
-    core_pos_next(pos);
+    // scc_probe_stream_reset(stream);
+    ch = scc_probe_stream_next(stream);
+    scc_pos_next(pos);
     // FIXME Assertion
     Assert(ch == '/');
-    ch = scc_probe_stream_consume(stream);
-    core_pos_next(pos);
+    ch = scc_probe_stream_next(stream);
+    scc_pos_next(pos);
     Assert(ch == '*');
 
     // all ready match `/*`
     while (1) {
-        scc_probe_stream_reset(stream);
+        // scc_probe_stream_reset(stream);
         ch = scc_probe_stream_peek(stream);
 
-        if (ch == core_stream_eof) {
+        if (ch == scc_stream_eof) {
             LOG_WARN("Unterminated block comment");
             return;
         }
 
-        if (lex_parse_is_endline(ch)) {
-            lex_parse_skip_endline(stream, pos);
+        if (scc_lex_parse_is_endline(ch)) {
+            scc_lex_parse_skip_endline(stream, pos);
             continue;
         }
-        scc_probe_stream_consume(stream);
-        core_pos_next(pos);
+        scc_probe_stream_next(stream);
+        scc_pos_next(pos);
         if (ch == '*') {
             ch = scc_probe_stream_peek(stream);
             if (ch == '/') {
-                scc_probe_stream_consume(stream);
-                core_pos_next(pos);
+                scc_probe_stream_next(stream);
+                scc_pos_next(pos);
                 return;
             }
         }
     }
 }
 
-void lex_parse_skip_whitespace(scc_probe_stream_t *input, scc_pos_t *pos) {
+void scc_lex_parse_skip_whitespace(scc_probe_stream_t *input, scc_pos_t *pos) {
     scc_probe_stream_t *stream = input;
     Assert(stream != null && pos != null);
-    scc_probe_stream_reset(stream);
+    // scc_probe_stream_reset(stream);
     while (1) {
         int ch = scc_probe_stream_peek(stream);
 
-        if (!lex_parse_is_whitespace(ch)) {
+        if (!scc_lex_parse_is_whitespace(ch)) {
             return;
         }
 
-        scc_probe_stream_consume(stream);
-        core_pos_next(pos);
+        scc_probe_stream_next(stream);
+        scc_pos_next(pos);
     }
 }
 
@@ -142,14 +143,14 @@ static inline cbool _lex_parse_uint(scc_probe_stream_t *input, scc_pos_t *pos,
         return false;
     }
     Assert(base == 2 || base == 8 || base == 10 || base == 16);
-    scc_probe_stream_reset(input);
+    // scc_probe_stream_reset(input);
     int ch, tmp;
     usize n = 0;
     usize offset = pos->offset;
     while (1) {
         ch = scc_probe_stream_peek(input);
 
-        if (ch == core_stream_eof) {
+        if (ch == scc_stream_eof) {
             break;
         } else if (ch >= 'a' && ch <= 'z') {
             tmp = ch - 'a' + 10;
@@ -166,8 +167,8 @@ static inline cbool _lex_parse_uint(scc_probe_stream_t *input, scc_pos_t *pos,
             return false;
         }
 
-        scc_probe_stream_consume(input);
-        core_pos_next(pos);
+        scc_probe_stream_next(input);
+        scc_pos_next(pos);
         n = n * base + tmp;
         // TODO number overflow
     }
@@ -187,32 +188,31 @@ static inline cbool _lex_parse_uint(scc_probe_stream_t *input, scc_pos_t *pos,
  * @return int
  * https://cppreference.cn/w/c/language/character_constant
  */
-int lex_parse_char(scc_probe_stream_t *input, scc_pos_t *pos) {
+int scc_lex_parse_char(scc_probe_stream_t *input, scc_pos_t *pos) {
     scc_probe_stream_t *stream = input;
     Assert(stream != null && pos != null);
-    scc_probe_stream_reset(stream);
-    int ch = scc_probe_stream_peek(stream);
-    int ret = core_stream_eof;
+    int ch = scc_probe_stream_next(stream);
+    scc_pos_next(pos);
+    int ret = scc_stream_eof;
 
-    if (ch == core_stream_eof) {
+    if (ch == scc_stream_eof) {
         LOG_WARN("Unexpected EOF at begin");
         goto ERR;
     } else if (ch != '\'') {
         LOG_WARN("Unexpected character '%c' at begin", ch);
         goto ERR;
     }
-    scc_probe_stream_consume(stream);
-    core_pos_next(pos);
+    // scc_probe_stream_next(stream);
 
-    ch = scc_probe_stream_consume(stream);
-    core_pos_next(pos);
+    ch = scc_probe_stream_next(stream);
+    scc_pos_next(pos);
 
-    if (ch == core_stream_eof) {
+    if (ch == scc_stream_eof) {
         LOG_WARN("Unexpected EOF at middle");
         goto ERR;
     } else if (ch == '\\') {
-        ch = scc_probe_stream_consume(stream);
-        core_pos_next(pos);
+        ch = scc_probe_stream_next(stream);
+        scc_pos_next(pos);
         if (ch == '0') {
             // 数字转义序列
             // \nnn	任意八进制值	码元 nnn
@@ -237,15 +237,15 @@ int lex_parse_char(scc_probe_stream_t *input, scc_pos_t *pos) {
     } else {
         ret = ch;
     }
-    if ((ch = scc_probe_stream_consume(stream)) != '\'') {
+    if ((ch = scc_probe_stream_next(stream)) != '\'') {
         LOG_ERROR("Unclosed character literal '%c' at end, expect `'`", ch);
-        core_pos_next(pos);
+        scc_pos_next(pos);
         goto ERR;
     }
 
     return ret;
 ERR:
-    return core_stream_eof;
+    return scc_stream_eof;
 }
 
 /**
@@ -257,38 +257,38 @@ ERR:
  * @return cbool
  * https://cppreference.cn/w/c/language/string_literal
  */
-cbool lex_parse_string(scc_probe_stream_t *input, scc_pos_t *pos,
-                       scc_cstring_t *output) {
+cbool scc_lex_parse_string(scc_probe_stream_t *input, scc_pos_t *pos,
+                           scc_cstring_t *output) {
     scc_probe_stream_t *stream = input;
     Assert(stream != null && pos != null && output != null);
-    scc_probe_stream_reset(stream);
+    // scc_probe_stream_reset(stream);
     int ch = scc_probe_stream_peek(stream);
 
     Assert(scc_cstring_is_empty(output));
-    if (ch == core_stream_eof) {
+    if (ch == scc_stream_eof) {
         LOG_WARN("Unexpected EOF at begin");
         goto ERR;
     } else if (ch != '"') {
         LOG_WARN("Unexpected character '%c' at begin", ch);
         goto ERR;
     }
-    scc_probe_stream_consume(stream);
-    core_pos_next(pos);
+    scc_probe_stream_next(stream);
+    scc_pos_next(pos);
 
     scc_cstring_t str = scc_cstring_from_cstr("");
     while (1) {
         ch = scc_probe_stream_peek(stream);
 
-        if (ch == core_stream_eof) {
+        if (ch == scc_stream_eof) {
             LOG_ERROR("Unexpected EOF at string literal");
             goto ERR;
-        } else if (lex_parse_is_endline(ch)) {
+        } else if (scc_lex_parse_is_endline(ch)) {
             LOG_ERROR("Unexpected newline at string literal");
             goto ERR;
         } else if (ch == '\\') {
             // TODO bad practice and maybe bugs here
-            scc_probe_stream_consume(stream);
-            ch = scc_probe_stream_consume(stream);
+            scc_probe_stream_next(stream);
+            ch = scc_probe_stream_next(stream);
             int val = got_simple_escape(ch);
             if (val == -1) {
                 LOG_ERROR("Invalid escape character it is \\%c [%d]", ch, ch);
@@ -297,13 +297,13 @@ cbool lex_parse_string(scc_probe_stream_t *input, scc_pos_t *pos,
                 continue;
             }
         } else if (ch == '"') {
-            scc_probe_stream_consume(stream);
-            core_pos_next(pos);
+            scc_probe_stream_next(stream);
+            scc_pos_next(pos);
             break;
         }
 
-        scc_probe_stream_consume(stream);
-        core_pos_next(pos);
+        scc_probe_stream_next(stream);
+        scc_pos_next(pos);
         scc_cstring_append_ch(&str, ch);
     }
 
@@ -323,36 +323,36 @@ ERR:
  * @return cbool
  * https://cppreference.cn/w/c/language/integer_constant
  */
-cbool lex_parse_number(scc_probe_stream_t *input, scc_pos_t *pos,
-                       usize *output) {
+cbool scc_lex_parse_number(scc_probe_stream_t *input, scc_pos_t *pos,
+                           usize *output) {
     scc_probe_stream_t *stream = input;
     Assert(stream != null && pos != null && output != null);
-    scc_probe_stream_reset(stream);
+    // scc_probe_stream_reset(stream);
     int ch = scc_probe_stream_peek(stream);
     int base = 10; // 默认十进制
 
-    if (ch == core_stream_eof) {
+    if (ch == scc_stream_eof) {
         LOG_WARN("Unexpected EOF at begin");
         goto ERR;
     }
 
     if (ch == '0') {
         // 消费 '0'
-        scc_probe_stream_consume(stream);
-        core_pos_next(pos);
+        scc_probe_stream_next(stream);
+        scc_pos_next(pos);
 
         // 查看下一个字符
         ch = scc_probe_stream_peek(stream);
         if (ch == 'x' || ch == 'X') {
             // 十六进制
             base = 16;
-            scc_probe_stream_consume(stream);
-            core_pos_next(pos);
+            scc_probe_stream_next(stream);
+            scc_pos_next(pos);
         } else if (ch == 'b' || ch == 'B') {
             // 二进制 (C23扩展)
             base = 2;
-            scc_probe_stream_consume(stream);
-            core_pos_next(pos);
+            scc_probe_stream_next(stream);
+            scc_pos_next(pos);
         } else if (ch >= '0' && ch <= '7') {
             // 八进制
             base = 8;
@@ -374,7 +374,7 @@ cbool lex_parse_number(scc_probe_stream_t *input, scc_pos_t *pos,
     }
 
     // 解析整数部分
-    scc_probe_stream_reset(stream);
+    // scc_probe_stream_reset(stream);
     usize n;
     if (_lex_parse_uint(stream, pos, base, &n) == false) {
         // 如果没有匹配任何数字，但输入是 '0'，已经处理过了
@@ -383,8 +383,8 @@ cbool lex_parse_number(scc_probe_stream_t *input, scc_pos_t *pos,
             // 单个数字的情况，例如 "1"
             // 我们需要消费这个数字并返回它的值
             if (ch >= '1' && ch <= '9') {
-                scc_probe_stream_consume(stream);
-                core_pos_next(pos);
+                scc_probe_stream_next(stream);
+                scc_pos_next(pos);
                 *output = ch - '0';
                 return true;
             }
@@ -406,22 +406,21 @@ ERR:
  * @return cbool
  * https://cppreference.cn/w/c/language/identifier
  */
-cbool lex_parse_identifier(scc_probe_stream_t *input, scc_pos_t *pos,
-                           scc_cstring_t *output) {
+cbool scc_lex_parse_identifier(scc_probe_stream_t *input, scc_pos_t *pos,
+                               scc_cstring_t *output) {
     Assert(input != null && pos != null && output != null);
     Assert(scc_cstring_is_empty(output));
     scc_probe_stream_t *stream = input;
-    scc_probe_stream_reset(stream);
+    // scc_probe_stream_reset(stream);
     int ch = scc_probe_stream_peek(stream);
 
-    if (ch == core_stream_eof) {
+    if (ch == scc_stream_eof) {
         LOG_WARN("Unexpected EOF at begin");
-    } else if (ch == '_' || (ch >= 'a' && ch <= 'z') ||
-               (ch >= 'A' && ch <= 'Z')) {
+    } else if (scc_lex_parse_is_identifier_prefix(ch)) {
         while (1) {
             scc_cstring_append_ch(output, ch);
-            scc_probe_stream_consume(stream);
-            core_pos_next(pos);
+            scc_probe_stream_next(stream);
+            scc_pos_next(pos);
             ch = scc_probe_stream_peek(stream);
             if ((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z') ||
                 (ch == '_') || (ch >= '0' && ch <= '9')) {

libs/lex_parser/tests/test_char.c

@@ -4,12 +4,16 @@
 
 cbool check_char(const char *str, int expect, int *output) {
     log_set_level(&__default_logger_root, 0);
-    scc_pos_t pos = scc_pos_init();
+    scc_pos_t pos = scc_pos_create();
     scc_mem_probe_stream_t mem_stream;
     scc_probe_stream_t *stream =
         scc_mem_probe_stream_init(&mem_stream, str, scc_strlen(str), false);
-    *output = lex_parse_char(stream, &pos);
-    return *output == expect;
+    *output = scc_lex_parse_char(stream, &pos);
+    cbool ret1 = *output == expect;
+    scc_probe_stream_reset(stream);
+    *output = scc_lex_parse_char(stream, &pos);
+    cbool ret2 = *output == expect;
+    return ret1 && ret2;
 }
 
 #define CHECK_CHAR_VALID(str, expect)                                          \
@@ -22,8 +26,8 @@ cbool check_char(const char *str, int expect, int *output) {
 #define CHECK_CHAR_INVALID(str)                                                \
     do {                                                                       \
         int _output;                                                           \
-        check_char(str, core_stream_eof, &_output);                            \
-        TEST_CHECK(_output == core_stream_eof);                                \
+        check_char(str, scc_stream_eof, &_output);                             \
+        TEST_CHECK(_output == scc_stream_eof);                                 \
     } while (0)
 
 void test_simple_char(void) {

libs/lex_parser/tests/test_identifier.c

@@ -5,12 +5,12 @@
 cbool check_identifier(const char *str, const char *expect,
                        scc_cstring_t *output) {
     log_set_level(&__default_logger_root, 0);
-    scc_pos_t pos = scc_pos_init();
+    scc_pos_t pos = scc_pos_create();
     scc_mem_probe_stream_t mem_stream;
     scc_probe_stream_t *stream =
         scc_mem_probe_stream_init(&mem_stream, str, scc_strlen(str), false);
 
-    cbool ret = lex_parse_identifier(stream, &pos, output);
+    cbool ret = scc_lex_parse_identifier(stream, &pos, output);
     if (ret && expect) {
         return strcmp(output->data, expect) == 0;
     }
@@ -19,7 +19,7 @@ cbool check_identifier(const char *str, const char *expect,
 
 #define CHECK_IDENTIFIER_VALID(str, expect)                                    \
     do {                                                                       \
-        scc_cstring_t _output = scc_cstring_new();                             \
+        scc_cstring_t _output = scc_cstring_create();                          \
         cbool ret = check_identifier(str, expect, &_output);                   \
         TEST_CHECK(ret == true);                                               \
         TEST_CHECK(strcmp(_output.data, expect) == 0);                         \
@@ -28,7 +28,7 @@ cbool check_identifier(const char *str, const char *expect,
 
 #define CHECK_IDENTIFIER_INVALID(str)                                          \
     do {                                                                       \
-        scc_cstring_t _output = scc_cstring_new();                             \
+        scc_cstring_t _output = scc_cstring_create();                          \
         cbool ret = check_identifier(str, NULL, &_output);                     \
         TEST_CHECK(ret == false);                                              \
         scc_cstring_free(&_output);                                            \

libs/lex_parser/tests/test_number.c

@@ -5,11 +5,11 @@ cbool check(const char *str, usize expect, usize *output) {
     // TODO maybe have other logger
     (void)(expect);
     log_set_level(&__default_logger_root, 0);
-    scc_pos_t pos = scc_pos_init();
+    scc_pos_t pos = scc_pos_create();
     scc_mem_probe_stream_t mem_stream;
     scc_probe_stream_t *stream =
         scc_mem_probe_stream_init(&mem_stream, str, scc_strlen(str), false);
-    return lex_parse_number(stream, &pos, output);
+    return scc_lex_parse_number(stream, &pos, output);
 }
 
 #define CHECK_VALID(str, expect)                                               \

libs/lex_parser/tests/test_skip_block_comment.c

@@ -4,18 +4,19 @@
 
 void check_skip_block_comment(const char *str, const char *expect_remaining) {
     log_set_level(&__default_logger_root, 0);
-    scc_pos_t pos = scc_pos_init();
+    scc_pos_t pos = scc_pos_create();
     scc_mem_probe_stream_t mem_stream;
     scc_probe_stream_t *stream =
         scc_mem_probe_stream_init(&mem_stream, str, scc_strlen(str), false);
 
-    lex_parse_skip_block_comment(stream, &pos);
+    scc_lex_parse_skip_block_comment(stream, &pos);
+    scc_probe_stream_sync(stream);
 
     // Check remaining content
     char buffer[256] = {0};
     int i = 0;
     int ch;
-    while ((ch = scc_probe_stream_consume(stream)) != core_stream_eof &&
+    while ((ch = scc_probe_stream_consume(stream)) != scc_stream_eof &&
            i < 255) {
         buffer[i++] = (char)ch;
     }

libs/lex_parser/tests/test_skip_line.c

@@ -4,18 +4,19 @@
 
 void check_skip_line(const char *str, const char *expect_remaining) {
     log_set_level(&__default_logger_root, 0);
-    scc_pos_t pos = scc_pos_init();
+    scc_pos_t pos = scc_pos_create();
     scc_mem_probe_stream_t mem_stream;
     scc_probe_stream_t *stream =
         scc_mem_probe_stream_init(&mem_stream, str, scc_strlen(str), false);
 
-    lex_parse_skip_line(stream, &pos);
+    scc_lex_parse_skip_line(stream, &pos);
+    scc_probe_stream_sync(stream);
 
     // Check remaining content
     char buffer[256] = {0};
     int i = 0;
     int ch;
-    while ((ch = scc_probe_stream_consume(stream)) != core_stream_eof &&
+    while ((ch = scc_probe_stream_consume(stream)) != scc_stream_eof &&
            i < 255) {
         buffer[i++] = (char)ch;
     }

libs/lex_parser/tests/test_string.c

@@ -4,12 +4,12 @@
 
 cbool check_string(const char *str, const char *expect, scc_cstring_t *output) {
     log_set_level(&__default_logger_root, 0);
-    scc_pos_t pos = scc_pos_init();
+    scc_pos_t pos = scc_pos_create();
     scc_mem_probe_stream_t mem_stream;
     scc_probe_stream_t *stream =
         scc_mem_probe_stream_init(&mem_stream, str, scc_strlen(str), false);
 
-    cbool ret = lex_parse_string(stream, &pos, output);
+    cbool ret = scc_lex_parse_string(stream, &pos, output);
     if (ret && expect) {
         return strcmp(output->data, expect) == 0;
     }
@@ -18,7 +18,7 @@ cbool check_string(const char *str, const char *expect, scc_cstring_t *output) {
 
 #define CHECK_STRING_VALID(str, expect)                                        \
     do {                                                                       \
-        scc_cstring_t _output = scc_cstring_new();                             \
+        scc_cstring_t _output = scc_cstring_create();                          \
         cbool ret = check_string(str, expect, &_output);                       \
         TEST_CHECK(ret == true);                                               \
         TEST_CHECK(strcmp(_output.data, expect) == 0);                         \
@@ -27,7 +27,7 @@ cbool check_string(const char *str, const char *expect, scc_cstring_t *output) {
 
 #define CHECK_STRING_INVALID(str)                                              \
     do {                                                                       \
-        scc_cstring_t _output = scc_cstring_new();                             \
+        scc_cstring_t _output = scc_cstring_create();                          \
         cbool ret = check_string(str, NULL, &_output);                         \
         TEST_CHECK(ret == false);                                              \
         scc_cstring_free(&_output);                                            \

libs/lexer/cbuild.toml

@@ -1,8 +1,8 @@
 [package]
-name = "smcc_lex"
+name = "scc_lex"
 version = "0.1.0"
 
 dependencies = [
-    { name = "libcore", path = "../../runtime/libcore" },
-    { name = "smcc_lex_parser", path = "../lex_parser" },
+    { name = "scc_core", path = "../../runtime/scc_core" },
+    { name = "lex_parser", path = "../lex_parser" },
 ]

libs/lexer/include/lexer.h

@@ -7,20 +7,14 @@
 #define __SCC_LEXER_H__
 
 #include "lexer_token.h"
-#include <libcore.h>
-
-typedef struct lexer_token {
-    scc_tok_type_t type;
-    scc_cvalue_t value;
-    scc_pos_t loc;
-} lexer_tok_t;
+#include <scc_core.h>
 
 /**
  * @brief 词法分析器核心结构体
  *
  * 封装词法分析所需的状态信息和缓冲区管理
  */
-typedef struct cc_lexer {
+typedef struct scc_lexer {
     scc_probe_stream_t *stream;
     scc_pos_t pos;
 } scc_lexer_t;
@@ -39,7 +33,7 @@ void scc_lexer_init(scc_lexer_t *lexer, scc_probe_stream_t *stream);
  *
  * 此函数会返回所有类型的token，包括空白符等无效token
  */
-void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token);
+void scc_lexer_get_token(scc_lexer_t *lexer, scc_lexer_tok_t *token);
 
 /**
  * @brief 获取有效token
@@ -48,6 +42,63 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token);
  *
  * 此函数会自动跳过空白符等无效token，返回对语法分析有意义的token
  */
-void scc_lexer_get_valid_token(scc_lexer_t *lexer, lexer_tok_t *token);
+void scc_lexer_get_valid_token(scc_lexer_t *lexer, scc_lexer_tok_t *token);
+
+typedef SCC_VEC(scc_lexer_tok_t) scc_lexer_tok_vec_t;
+
+typedef struct scc_lexer_stream scc_lexer_stream_t;
+struct scc_lexer_stream {
+    scc_lexer_t *lexer;
+    scc_lexer_tok_vec_t toks; // 循环缓冲区
+    usize curr_pos;           // 当前读取位置（逻辑位置）
+    usize probe_pos;          // 已填充位置（逻辑位置）
+    cbool need_comment;
+
+    /// @brief 向前读取n个token
+    const scc_lexer_tok_t *(*peek)(scc_lexer_stream_t *stream, usize n);
+
+    /// @brief 指针推进到offset
+    void (*advance)(scc_lexer_stream_t *stream, usize offset);
+
+    /// @brief 销毁并释放资源
+    void (*drop)(scc_lexer_stream_t *stream);
+};
+
+/**
+ * @brief 将词法分析器转换成流式输出(自带缓冲区)
+ * @param[in] lexer 已经词法分析器实例
+ * @param[out] stream 输出流对象指针
+ * @param[in] need_comment 输出时是否需要注释
+ */
+void scc_lexer_to_stream(scc_lexer_t *lexer, scc_lexer_stream_t *stream,
+                         cbool need_comment);
+
+static inline const scc_lexer_tok_t *
+scc_lexer_stream_current(scc_lexer_stream_t *stream) {
+    Assert(stream != null);
+    return stream->peek(stream, 0);
+}
+
+static inline const scc_lexer_tok_t *
+scc_lexer_stream_peek(scc_lexer_stream_t *stream, usize n) {
+    Assert(stream != null);
+    return stream->peek(stream, n);
+}
+
+static inline void scc_lexer_stream_consume(scc_lexer_stream_t *stream) {
+    Assert(stream != null);
+    return stream->advance(stream, 1);
+}
+
+static inline void scc_lexer_stream_advance(scc_lexer_stream_t *stream,
+                                            usize n) {
+    Assert(stream != null);
+    return stream->advance(stream, n);
+}
+
+static inline void scc_lexer_stream_drop(scc_lexer_stream_t *stream) {
+    Assert(stream != null);
+    return stream->drop(stream);
+}
 
 #endif /* __SCC_LEXER_H__ */

libs/lexer/include/lexer_log.h

@@ -1,48 +1,48 @@
-#ifndef __SMCC_LEXER_LOG_H__
-#define __SMCC_LEXER_LOG_H__
+#ifndef __SCC_LEXER_LOG_H__
+#define __SCC_LEXER_LOG_H__
 
-#include <libcore.h>
+#include <scc_core.h>
 
 #ifndef LEX_LOG_LEVEL
 #define LEX_LOG_LEVEL 4
 #endif
 
 #if LEX_LOG_LEVEL <= 1
-#define LEX_NOTSET(fmt, ...) MLOG_NOTSET(&__smcc_lexer_log, fmt, ##__VA_ARGS__)
+#define LEX_NOTSET(fmt, ...) MLOG_NOTSET(&__scc_lexer_log, fmt, ##__VA_ARGS__)
 #else
 #define LEX_NOTSET(fmt, ...)
 #endif
 
 #if LEX_LOG_LEVEL <= 2
-#define LEX_DEBUG(fmt, ...) MLOG_DEBUG(&__smcc_lexer_log, fmt, ##__VA_ARGS__)
+#define LEX_DEBUG(fmt, ...) MLOG_DEBUG(&__scc_lexer_log, fmt, ##__VA_ARGS__)
 #else
 #define LEX_DEBUG(fmt, ...)
 #endif
 
 #if LEX_LOG_LEVEL <= 3
-#define LEX_INFO(fmt, ...) MLOG_INFO(&__smcc_lexer_log, fmt, ##__VA_ARGS__)
+#define LEX_INFO(fmt, ...) MLOG_INFO(&__scc_lexer_log, fmt, ##__VA_ARGS__)
 #else
 #define LEX_INFO(fmt, ...)
 #endif
 
 #if LEX_LOG_LEVEL <= 4
-#define LEX_WARN(fmt, ...) MLOG_WARN(&__smcc_lexer_log, fmt, ##__VA_ARGS__)
+#define LEX_WARN(fmt, ...) MLOG_WARN(&__scc_lexer_log, fmt, ##__VA_ARGS__)
 #else
 #define LEX_WARN(fmt, ...)
 #endif
 
 #if LEX_LOG_LEVEL <= 5
-#define LEX_ERROR(fmt, ...) MLOG_ERROR(&__smcc_lexer_log, fmt, ##__VA_ARGS__)
+#define LEX_ERROR(fmt, ...) MLOG_ERROR(&__scc_lexer_log, fmt, ##__VA_ARGS__)
 #else
 #define LEX_ERROR(fmt, ...)
 #endif
 
 #if LEX_LOG_LEVEL <= 6
-#define LEX_FATAL(fmt, ...) MLOG_FATAL(&__smcc_lexer_log, fmt, ##__VA_ARGS__)
+#define LEX_FATAL(fmt, ...) MLOG_FATAL(&__scc_lexer_log, fmt, ##__VA_ARGS__)
 #else
 #define LEX_FATAL(fmt, ...)
 #endif
 
-extern logger_t __smcc_lexer_log;
+extern logger_t __scc_lexer_log;
 
-#endif // __SMCC_LEXER_LOG_H__
+#endif /* __SCC_LEXER_LOG_H__ */

libs/lexer/include/lexer_token.h

@@ -1,7 +1,7 @@
-#ifndef __SMCC_CC_TOKEN_H__
-#define __SMCC_CC_TOKEN_H__
+#ifndef __SCC_LEXER_TOKEN_H__
+#define __SCC_LEXER_TOKEN_H__
 
-#include <libcore.h>
+#include <scc_core.h>
 
 typedef enum scc_cstd {
     SCC_CSTD_C89,
@@ -137,4 +137,24 @@ typedef enum scc_tok_subtype {
 scc_tok_subtype_t scc_get_tok_subtype(scc_tok_type_t type);
 const char *scc_get_tok_name(scc_tok_type_t type);
 
-#endif
+typedef struct scc_lexer_token {
+    scc_tok_type_t type;
+    scc_cvalue_t value;
+    scc_pos_t loc;
+} scc_lexer_tok_t;
+
+static inline cbool scc_lexer_tok_match(const scc_lexer_tok_t *tok,
+                                        scc_tok_type_t type) {
+    return tok->type == type;
+}
+
+static inline cbool scc_lexer_tok_expect(const scc_lexer_tok_t *tok,
+                                         scc_tok_type_t type) {
+    if (!scc_lexer_tok_match(tok, type)) {
+        LOG_ERROR("expected token %d, got %d\n", type, tok->type);
+        return false;
+    }
+    return true;
+}
+
+#endif /* __SCC_LEXER_TOKEN_H__ */

libs/lexer/src/lexer.c

@@ -1,31 +1,3 @@
-/**
- * 仿照LCCompiler的词法分析部分
- *
- * 如下为LCC的README in 2025.2
-This hierarchy is the distribution for lcc version 4.2.
-
-lcc version 3.x is described in the book "A Retargetable C Compiler:
-Design and Implementation" (Addison-Wesley, 1995, ISBN 0-8053-1670-1).
-There are significant differences between 3.x and 4.x, most notably in
-the intermediate code. For details, see
-https://drh.github.io/lcc/documents/interface4.pdf.
-
-VERSION 4.2 IS INCOMPATIBLE WITH EARLIER VERSIONS OF LCC. DO NOT
-UNLOAD THIS DISTRIBUTION ON TOP OF A 3.X DISTRIBUTION.
-
-LCC is a C89 ("ANSI C") compiler designed to be highly retargetable.
-
-LOG describes the changes since the last release.
-
-CPYRIGHT describes the conditions under you can use, copy, modify, and
-distribute lcc or works derived from lcc.
-
-doc/install.html is an HTML file that gives a complete description of
-the distribution and installation instructions.
-
-Chris Fraser / cwf@aya.yale.edu
-David Hanson / drh@drhanson.net
- */
 #include <lex_parser.h>
 #include <lexer.h>
 #include <lexer_log.h>
@@ -77,23 +49,23 @@ static inline int keyword_cmp(const char *name, int len) {
 
 void scc_lexer_init(scc_lexer_t *lexer, scc_probe_stream_t *stream) {
     lexer->stream = stream;
-    lexer->pos = scc_pos_init();
+    lexer->pos = scc_pos_create();
     // FIXME
-    lexer->pos.name = scc_cstring_from_cstr(scc_cstring_as_cstr(&stream->name));
+    lexer->pos.name = scc_cstring_copy(&stream->name);
 }
 
 #define set_err_token(token) ((token)->type = SCC_TOK_UNKNOWN)
 
-static void parse_line(scc_lexer_t *lexer, lexer_tok_t *token) {
+static void parse_line(scc_lexer_t *lexer, scc_lexer_tok_t *token) {
     token->loc = lexer->pos;
     scc_probe_stream_t *stream = lexer->stream;
     scc_probe_stream_reset(stream);
     int ch = scc_probe_stream_next(stream);
 
     usize n;
-    scc_cstring_t str = scc_cstring_new();
+    scc_cstring_t str = scc_cstring_create();
 
-    if (ch == core_stream_eof) {
+    if (ch == scc_stream_eof) {
         LEX_WARN("Unexpected EOF at begin");
         goto ERR;
     } else if (ch != '#') {
@@ -105,7 +77,7 @@ static void parse_line(scc_lexer_t *lexer, lexer_tok_t *token) {
 
     for (int i = 0; i < (int)sizeof(line); i++) {
         ch = scc_probe_stream_consume(stream);
-        core_pos_next(&lexer->pos);
+        scc_pos_next(&lexer->pos);
         if (ch != line[i]) {
             LEX_WARN("Maroc does not support in lexer rather in preprocessor, "
                      "it will be ignored");
@@ -113,13 +85,13 @@ static void parse_line(scc_lexer_t *lexer, lexer_tok_t *token) {
         }
     }
 
-    if (lex_parse_number(lexer->stream, &lexer->pos, &n) == false) {
+    if (scc_lex_parse_number(stream, &lexer->pos, &n) == false) {
         LEX_ERROR("Invalid line number");
         goto SKIP_LINE;
     }
 
     if (scc_probe_stream_consume(stream) != ' ') {
-        lex_parse_skip_line(lexer->stream, &lexer->pos);
+        scc_lex_parse_skip_line(stream, &lexer->pos);
         token->loc.line = token->value.n;
     }
 
@@ -127,26 +99,28 @@ static void parse_line(scc_lexer_t *lexer, lexer_tok_t *token) {
         LEX_ERROR("Invalid `#` line");
         goto SKIP_LINE;
     }
-    if (lex_parse_string(lexer->stream, &lexer->pos, &str) == false) {
+    if (scc_lex_parse_string(stream, &lexer->pos, &str) == false) {
         LEX_ERROR("Invalid filename");
         goto SKIP_LINE;
     }
 
-    lex_parse_skip_line(lexer->stream, &lexer->pos);
+    scc_lex_parse_skip_line(stream, &lexer->pos);
+    scc_probe_stream_sync(stream);
     token->loc.line = n;
     // FIXME memory leak
-    token->loc.name = scc_cstring_from_cstr(scc_cstring_as_cstr(&str));
+    token->loc.name = scc_cstring_copy(&str);
     scc_cstring_free(&str);
     return;
 SKIP_LINE:
-    lex_parse_skip_line(lexer->stream, &lexer->pos);
+    scc_lex_parse_skip_line(stream, &lexer->pos);
+    scc_probe_stream_sync(stream);
 ERR:
     set_err_token(token);
     scc_cstring_free(&str);
 }
 
 // /zh/c/language/operator_arithmetic.html
-void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
+void scc_lexer_get_token(scc_lexer_t *lexer, scc_lexer_tok_t *token) {
     token->loc = lexer->pos;
     token->type = SCC_TOK_UNKNOWN;
     scc_probe_stream_t *stream = lexer->stream;
@@ -212,11 +186,15 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
             type = SCC_TOK_ASSIGN_DIV;
             goto double_char;
         case '/':
-            lex_parse_skip_line(lexer->stream, &lexer->pos);
+            scc_probe_stream_reset(stream);
+            scc_lex_parse_skip_line(stream, &lexer->pos);
+            scc_probe_stream_sync(stream);
             token->type = SCC_TOK_LINE_COMMENT;
             goto END;
         case '*':
-            lex_parse_skip_block_comment(lexer->stream, &lexer->pos);
+            scc_probe_stream_reset(stream);
+            scc_lex_parse_skip_block_comment(stream, &lexer->pos);
+            scc_probe_stream_sync(stream);
             token->type = SCC_TOK_BLOCK_COMMENT;
             goto END;
         default:
@@ -323,33 +301,17 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
             break;
         }
         break;
-    case '[':
-        type = SCC_TOK_L_BRACKET;
-        break;
-    case ']':
-        type = SCC_TOK_R_BRACKET;
-        break;
-    case '(':
-        type = SCC_TOK_L_PAREN;
-        break;
-    case ')':
-        type = SCC_TOK_R_PAREN;
-        break;
-    case '{':
-        type = SCC_TOK_L_BRACE;
-        break;
-    case '}':
-        type = SCC_TOK_R_BRACE;
-        break;
-    case ';':
-        type = SCC_TOK_SEMICOLON;
-        break;
-    case ',':
-        type = SCC_TOK_COMMA;
-        break;
-    case ':':
-        type = SCC_TOK_COLON;
-        break;
+        /* clang-format off */
+    case '[': type = SCC_TOK_L_BRACKET; break;
+    case ']': type = SCC_TOK_R_BRACKET; break;
+    case '(': type = SCC_TOK_L_PAREN; break;
+    case ')': type = SCC_TOK_R_PAREN; break;
+    case '{': type = SCC_TOK_L_BRACE; break;
+    case '}': type = SCC_TOK_R_BRACE; break;
+    case ';': type = SCC_TOK_SEMICOLON; break;
+    case ',': type = SCC_TOK_COMMA; break;
+    case ':': type = SCC_TOK_COLON; break;
+        /* clang-format on */
     case '.':
         if (scc_probe_stream_next(stream) == '.' &&
             scc_probe_stream_next(stream) == '.') {
@@ -369,7 +331,8 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
         break;
     case '\r':
     case '\n':
-        lex_parse_skip_endline(lexer->stream, &lexer->pos);
+        scc_lex_parse_skip_endline(stream, &lexer->pos);
+        scc_probe_stream_sync(stream);
         token->type = SCC_TOK_BLANK;
         goto END;
     case '#':
@@ -377,15 +340,17 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
         token->type = SCC_TOK_BLANK;
         goto END;
     case '\0':
-    case core_stream_eof:
+    case scc_stream_eof:
         // EOF
         type = SCC_TOK_EOF;
         break;
     case '\'': {
         token->loc = lexer->pos;
         token->type = SCC_TOK_CHAR_LITERAL;
-        int ch = lex_parse_char(lexer->stream, &lexer->pos);
-        if (ch == core_stream_eof) {
+        scc_probe_stream_reset(stream);
+        int ch = scc_lex_parse_char(stream, &lexer->pos);
+        scc_probe_stream_sync(stream);
+        if (ch == scc_stream_eof) {
             LEX_ERROR("Unexpected character literal");
             token->type = SCC_TOK_UNKNOWN;
         } else {
@@ -396,8 +361,10 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
     case '"': {
         token->loc = lexer->pos;
         token->type = SCC_TOK_STRING_LITERAL;
-        scc_cstring_t output = scc_cstring_new();
-        if (lex_parse_string(lexer->stream, &lexer->pos, &output) == true) {
+        scc_cstring_t output = scc_cstring_create();
+        scc_probe_stream_reset(stream);
+        if (scc_lex_parse_string(stream, &lexer->pos, &output) == true) {
+            scc_probe_stream_sync(stream);
             token->value.cstr.data = scc_cstring_as_cstr(&output);
             token->value.cstr.len = scc_cstring_len(&output);
         } else {
@@ -414,7 +381,9 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
         token->loc = lexer->pos;
         token->type = SCC_TOK_INT_LITERAL;
         usize output;
-        if (lex_parse_number(lexer->stream, &lexer->pos, &output) == true) {
+        scc_probe_stream_reset(stream);
+        if (scc_lex_parse_number(stream, &lexer->pos, &output) == true) {
+            scc_probe_stream_sync(stream);
             token->value.n = output;
         } else {
             LEX_ERROR("Unexpected number literal");
@@ -431,8 +400,10 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
     case 'O': case 'P': case 'Q': case 'R': case 'S': case 'T': case 'U':
     case 'V': case 'W': case 'X': case 'Y': case 'Z': case '_':
         /* clang-format on */
-        scc_cstring_t str = scc_cstring_new();
-        cbool ret = lex_parse_identifier(lexer->stream, &lexer->pos, &str);
+        scc_cstring_t str = scc_cstring_create();
+        scc_probe_stream_reset(stream);
+        cbool ret = scc_lex_parse_identifier(stream, &lexer->pos, &str);
+        scc_probe_stream_sync(stream);
         Assert(ret == true);
 
         int res = keyword_cmp(scc_cstring_as_cstr(&str), scc_cstring_len(&str));
@@ -453,13 +424,13 @@ void scc_lexer_get_token(scc_lexer_t *lexer, lexer_tok_t *token) {
     goto once_char;
 triple_char:
     scc_probe_stream_consume(stream);
-    core_pos_next(&lexer->pos);
+    scc_pos_next(&lexer->pos);
 double_char:
     scc_probe_stream_consume(stream);
-    core_pos_next(&lexer->pos);
+    scc_pos_next(&lexer->pos);
 once_char:
     scc_probe_stream_consume(stream);
-    core_pos_next(&lexer->pos);
+    scc_pos_next(&lexer->pos);
     token->type = type;
 END:
     LEX_DEBUG("get token `%s` in %s:%d:%d", scc_get_tok_name(token->type),
@@ -467,7 +438,7 @@ END:
 }
 
 // scc_lexer_get_token maybe got invalid (with parser)
-void scc_lexer_get_valid_token(scc_lexer_t *lexer, lexer_tok_t *token) {
+void scc_lexer_get_valid_token(scc_lexer_t *lexer, scc_lexer_tok_t *token) {
     scc_tok_subtype_t type;
     do {
         scc_lexer_get_token(lexer, token);

libs/lexer/src/lexer_log.c

@@ -1,6 +1,6 @@
 #include <lexer_log.h>
 
-logger_t __smcc_lexer_log = {
+logger_t __scc_lexer_log = {
     .name = "lexer",
     .level = LOG_LEVEL_ALL,
     .handler = log_default_handler,

libs/lexer/src/lexer_stream.c

@@ -0,0 +1,138 @@
+#include <lexer.h>
+
+static void lexer_stream_extend(scc_lexer_stream_t *stream, usize n) {
+    Assert(stream != null);
+    // 检查是否需要扩容
+    if ((stream->probe_pos - stream->curr_pos + n) >= stream->toks.cap) {
+        // 需要扩容 - 创建新缓冲区
+        usize new_cap = stream->toks.cap * 2;
+        if (new_cap < stream->probe_pos - stream->curr_pos + n + 1) {
+            new_cap = stream->probe_pos - stream->curr_pos + n + 1;
+        }
+
+        scc_lexer_tok_t *new_data =
+            scc_realloc(null, new_cap * sizeof(scc_lexer_tok_t));
+        if (!new_data) {
+            LOG_FATAL("lexer_stream_extend: realloc failed\n");
+        }
+
+        // 将旧缓冲区中的数据拷贝到新缓冲区，保持顺序
+        usize data_count = stream->probe_pos - stream->curr_pos;
+        for (usize i = 0; i < data_count; ++i) {
+            usize old_idx = (stream->curr_pos + i) % stream->toks.cap;
+            new_data[i] = stream->toks.data[old_idx];
+        }
+
+        // 释放旧缓冲区
+        if (stream->toks.data) {
+            scc_free(stream->toks.data);
+        }
+
+        // 更新结构体
+        stream->toks.data = new_data;
+        stream->toks.cap = new_cap;
+        stream->curr_pos = 0;
+        stream->probe_pos = data_count;
+    }
+
+    // 填充新token
+    for (usize i = 0; i < n; ++i) {
+        usize idx = (stream->probe_pos + i) % stream->toks.cap;
+        if (stream->need_comment)
+            scc_lexer_get_token(stream->lexer, &stream->toks.data[idx]);
+        else
+            scc_lexer_get_valid_token(stream->lexer, &stream->toks.data[idx]);
+    }
+
+    stream->probe_pos += n;
+}
+
+static const scc_lexer_tok_t *lexer_stream_peek(scc_lexer_stream_t *stream,
+                                                usize n) {
+    Assert(stream != null);
+
+    // 计算需要的前看token数量
+    usize available = stream->probe_pos - stream->curr_pos;
+    if (n >= available) {
+        // 需要扩展缓冲区
+        usize need = n - available + 1;
+        lexer_stream_extend(stream, need);
+    }
+
+    // 计算实际缓冲区中的位置
+    usize idx = (stream->curr_pos + n) % stream->toks.cap;
+    return &stream->toks.data[idx];
+}
+
+static void lexer_stream_advance(scc_lexer_stream_t *stream, usize offset) {
+    Assert(stream != null);
+
+    if (stream->curr_pos + offset > stream->probe_pos) {
+        // 尝试填充更多token
+        usize need = stream->curr_pos + offset - stream->probe_pos;
+        lexer_stream_extend(stream, need);
+    }
+
+    stream->curr_pos += offset;
+
+    // 可选：当已消费的token过多时，压缩缓冲区
+    if (stream->curr_pos > stream->toks.cap * 3 / 4) {
+        // 压缩缓冲区：将有效数据移动到前面
+        usize data_count = stream->probe_pos - stream->curr_pos;
+        scc_lexer_tok_t *temp =
+            scc_realloc(null, data_count * sizeof(scc_lexer_tok_t));
+        if (!temp)
+            return; // 压缩失败也没关系
+
+        for (usize i = 0; i < data_count; ++i) {
+            usize old_idx = (stream->curr_pos + i) % stream->toks.cap;
+            temp[i] = stream->toks.data[old_idx];
+        }
+
+        scc_free(stream->toks.data);
+        stream->toks.data = temp;
+        stream->toks.cap = data_count;
+        stream->curr_pos = 0;
+        stream->probe_pos = data_count;
+    }
+}
+
+static void lexer_stream_drop(scc_lexer_stream_t *stream) {
+    Assert(stream != null);
+
+    // 清理所有token（如果有需要清理的内部资源）
+    for (usize i = 0; i < stream->toks.cap; ++i) {
+        // 这里假设scc_lexer_tok_t可能包含需要释放的资源
+        // 如果有，需要调用相应的清理函数
+        // 例如: if (stream->toks.data[i].needs_free)
+        // scc_free(stream->toks.data[i].ptr);
+    }
+
+    scc_vec_free(stream->toks);
+    stream->lexer = null;
+    stream->curr_pos = 0;
+    stream->probe_pos = 0;
+    stream->need_comment = false;
+
+    stream->peek = null;
+    stream->advance = null;
+    stream->drop = null;
+}
+
+void scc_lexer_to_stream(scc_lexer_t *lexer, scc_lexer_stream_t *stream,
+                         cbool need_comment) {
+    Assert(lexer != null && stream != null);
+
+    stream->lexer = lexer;
+    stream->curr_pos = 0;
+    stream->probe_pos = 0;
+    stream->need_comment = need_comment;
+
+    // 初始化循环缓冲区
+    scc_vec_init(stream->toks);
+    scc_vec_realloc(stream->toks, 8); // 初始容量为8
+
+    stream->peek = lexer_stream_peek;
+    stream->advance = lexer_stream_advance;
+    stream->drop = lexer_stream_drop;
+}

libs/lexer/tests/test_parse.c

@@ -7,7 +7,7 @@
 static inline void test_lexer_string(const char *input,
                                      scc_tok_type_t expected_type) {
     scc_lexer_t lexer;
-    lexer_tok_t token;
+    scc_lexer_tok_t token;
     scc_mem_probe_stream_t stream;
 
     scc_lexer_init(&lexer, scc_mem_probe_stream_init(&stream, input,

libs/lexer/tests/test_run.c

@@ -27,8 +27,9 @@ int main(int argc, char *argv[]) {
         log_set_level(NULL, LOG_LEVEL_ALL);
     } else {
         // FIXME it is a hack lexer_logger
-        log_set_level(&__smcc_lexer_log, LOG_LEVEL_NOTSET);
-        log_set_level(NULL, LOG_LEVEL_INFO | LOG_LEVEL_WARN | LOG_LEVEL_ERROR);
+        log_set_level(&__scc_lexer_log, LOG_LEVEL_NOTSET);
+        log_set_level(NULL, LOG_LEVEL_INFO | LOG_LEVEL_WARN | LOG_LEVEL_ERROR |
+                                LOG_LEVEL_FATAL);
     }
 
     const char *file_name = __FILE__;
@@ -70,11 +71,11 @@ int main(int argc, char *argv[]) {
     scc_cstring_clear(&stream->name);
     scc_cstring_append_cstr(&stream->name, file_name, strlen(file_name));
     scc_lexer_init(&lexer, stream);
-    lexer_tok_t tok;
+    scc_lexer_tok_t tok;
 
     while (1) {
         scc_lexer_get_valid_token(&lexer, &tok);
-        if (tok.type ==  SCC_TOK_EOF) {
+        if (tok.type == SCC_TOK_EOF) {
             break;
         }
         LOG_DEBUG("token `%s` at %s:%u:%u", scc_get_tok_name(tok.type),

libs/pprocessor/cbuild.toml

@@ -1,8 +1,8 @@
 [package]
-name = "smcc_pprocesser"
+name = "scc_pprocesser"
 
 dependencies = [
-    { name = "libcore", path = "../../runtime/libcore" },
-    { name = "libutils", path = "../../runtime/libutils" },
-    { name = "smcc_lex_parser", path = "../lex_parser" },
+    { name = "scc_core", path = "../../runtime/scc_core" },
+    { name = "scc_utils", path = "../../runtime/scc_utils" },
+    { name = "lex_parser", path = "../lex_parser" },
 ]

libs/pprocessor/include/pp_macro.h

@@ -0,0 +1,96 @@
+#ifndef __SCC_PP_MACRO_H__
+#define __SCC_PP_MACRO_H__
+
+#include <scc_core.h>
+#include <scc_utils.h>
+
+// 宏定义类型
+typedef enum {
+    SCC_PP_MACRO_OBJECT,   // 对象宏
+    SCC_PP_MACRO_FUNCTION, // 函数宏
+    SCC_PP_MACRO_NONE,     // 不是宏
+} scc_pp_macro_type_t;
+
+typedef SCC_VEC(scc_cstring_t) scc_pp_macro_list_t;
+
+// 宏定义结构
+typedef struct scc_macro {
+    scc_cstring_t name;           // 宏名称
+    scc_pp_macro_type_t type;     // 宏类型
+    scc_pp_macro_list_t replaces; // 替换列表
+    scc_pp_macro_list_t params;   // 参数列表（仅函数宏）
+} scc_pp_macro_t;
+
+typedef struct scc_macro_table {
+    scc_hashtable_t table; // 宏定义表
+} scc_pp_macro_table_t;
+
+/**
+ * @brief 创建宏对象
+ * @param name 宏名称
+ * @param type 宏类型
+ * @return 创建的宏对象指针，失败返回NULL
+ */
+scc_pp_macro_t *scc_pp_macro_new(const scc_cstring_t *name,
+                                 scc_pp_macro_type_t type);
+
+/**
+ * @brief 销毁宏对象
+ * @param macro 要销毁的宏对象
+ */
+void scc_pp_macro_drop(scc_pp_macro_t *macro);
+
+/**
+ * @brief 添加对象宏
+ * @param pp 预处理器实例
+ * @param name 宏名称
+ * @param replacement 替换文本列表
+ * @return 成功返回true，失败返回false
+ */
+cbool scc_pp_add_object_macro(scc_pp_macro_table_t *pp,
+                              const scc_cstring_t *name,
+                              const scc_pp_macro_list_t *replacement);
+
+/**
+ * @brief 添加函数宏
+ * @param pp 预处理器实例
+ * @param name 宏名称
+ * @param params 参数列表
+ * @param replacement 替换文本列表
+ * @return 成功返回true，失败返回false
+ */
+cbool scc_pp_add_function_macro(scc_pp_macro_table_t *pp,
+                                const scc_cstring_t *name,
+                                const scc_pp_macro_list_t *params,
+                                const scc_pp_macro_list_t *replacement);
+/**
+ * @brief
+ *
+ * @param pp
+ * @param macro
+ * @return scc_pp_macro_t*
+ */
+scc_pp_macro_t *scc_pp_macro_table_set(scc_pp_macro_table_t *pp,
+                                       scc_pp_macro_t *macro);
+
+/**
+ * @brief 查找宏定义
+ * @param pp 预处理器实例
+ * @param name 宏名称
+ * @return 找到的宏对象指针，未找到返回NULL
+ */
+scc_pp_macro_t *scc_pp_macro_table_get(scc_pp_macro_table_t *pp,
+                                       const scc_cstring_t *name);
+
+/**
+ * @brief 从预处理器中删除宏
+ * @param pp 预处理器实例
+ * @param name 宏名称
+ * @return 成功删除返回true，未找到返回false
+ */
+cbool scc_pp_macro_table_remove(scc_pp_macro_table_t *pp,
+                                const scc_cstring_t *name);
+
+void scc_pp_marco_table_init(scc_pp_macro_table_t *macros);
+void scc_pp_macro_table_drop(scc_pp_macro_table_t *macros);
+#endif /* __SCC_PP_MACRO_H__ */

libs/pprocessor/include/pp_parse.h

@@ -0,0 +1,18 @@
+#ifndef __SCC_PP_PARSE_H__
+#define __SCC_PP_PARSE_H__
+
+#include <pp_macro.h>
+#include <scc_core.h>
+void scc_pp_parse_directive(scc_probe_stream_t *stream, scc_pos_t *pos,
+                            scc_pp_macro_table_t *macros);
+cbool scc_pp_parse_macro_replace_list(scc_probe_stream_t *stream,
+                                      scc_pp_macro_list_t *list);
+cbool scc_pp_parse_macro_arguments(scc_probe_stream_t *stream,
+                                   scc_pp_macro_list_t *args);
+
+// expand
+cbool scc_pp_expand_macro(scc_probe_stream_t *stream,
+                          scc_pp_macro_table_t *macros,
+                          scc_probe_stream_t **out_stream, int depth);
+
+#endif /* __SCC_PP_PARSE_H__ */

libs/pprocessor/include/pp_token.h

@@ -1,30 +1,30 @@
-#ifndef __SMCC_PP_TOKEN_H__
-#define __SMCC_PP_TOKEN_H__
+#ifndef __SCC_PP_TOKEN_H__
+#define __SCC_PP_TOKEN_H__
 
 /* clang-format off */
 /// https://cppreference.cn/w/c/preprocessor
-#define PP_INST_TOKEN \
-    X(define    , PP_STD, PP_TOK_DEFINE    ) \
-    X(undef     , PP_STD, PP_TOK_UNDEF     ) \
-    X(include   , PP_STD, PP_TOK_INCLUDE   ) \
-    X(if        , PP_STD, PP_TOK_IF        ) \
-    X(ifdef     , PP_STD, PP_TOK_IFDEF     ) \
-    X(ifndef    , PP_STD, PP_TOK_IFNDEF    ) \
-    X(else      , PP_STD, PP_TOK_ELSE      ) \
-    X(elif      , PP_STD, PP_TOK_ELIF      ) \
-    X(elifdef   , PP_STD, PP_TOK_ELIFDEF   ) \
-    X(elifndef  , PP_C23, PP_TOK_ELIFNDEF  ) \
-    X(endif     , PP_STD, PP_TOK_ENDIF     ) \
-    X(line      , PP_STD, PP_TOK_LINE      ) \
-    X(embed     , PP_C23, PP_TOK_EMBED     ) \
-    X(error     , PP_STD, PP_TOK_ERROR     ) \
-    X(warning   , PP_C23, PP_TOK_WARNING   ) \
-    X(pragma    , PP_STD, PP_TOK_PRAMA     ) \
+#define SCC_PP_INST_TOKEN \
+    X(define    , SCC_PP_STD, SCC_PP_TOK_DEFINE     ) \
+    X(elif      , SCC_PP_STD, SCC_PP_TOK_ELIF       ) \
+    X(elifdef   , SCC_PP_STD, SCC_PP_TOK_ELIFDEF    ) \
+    X(elifndef  , SCC_PP_STD, SCC_PP_TOK_ELIFNDEF   ) \
+    X(else      , SCC_PP_STD, SCC_PP_TOK_ELSE       ) \
+    X(embed     , SCC_PP_STD, SCC_PP_TOK_EMBED      ) \
+    X(endif     , SCC_PP_STD, SCC_PP_TOK_ENDIF      ) \
+    X(error     , SCC_PP_STD, SCC_PP_TOK_ERROR      ) \
+    X(if        , SCC_PP_STD, SCC_PP_TOK_IF         ) \
+    X(ifdef     , SCC_PP_C23, SCC_PP_TOK_IFDEF      ) \
+    X(ifndef    , SCC_PP_STD, SCC_PP_TOK_IFNDEF     ) \
+    X(include   , SCC_PP_STD, SCC_PP_TOK_INCLUDE    ) \
+    X(line      , SCC_PP_C23, SCC_PP_TOK_LINE       ) \
+    X(pragma    , SCC_PP_STD, SCC_PP_TOK_PRAGMA     ) \
+    X(undef     , SCC_PP_C23, SCC_PP_TOK_UNDEF      ) \
+    X(warning   , SCC_PP_STD, SCC_PP_TOK_WARNING    ) \
     // END
 /* clang-format on */
 
 #define X(name, type, tok) tok,
-typedef enum pp_token { PP_INST_TOKEN } pp_token_t;
+typedef enum scc_pp_token { SCC_PP_INST_TOKEN } scc_pp_token_t;
 #undef X
 
-#endif /* __SMCC_PP_TOKEN_H__ */
+#endif /* __SCC_PP_TOKEN_H__ */

libs/pprocessor/include/pprocessor.h

@@ -1,30 +1,14 @@
-// pprocessor.h - 更新后的头文件
 /**
  * @file pprocessor.h
  * @brief C语言预处理器核心数据结构与接口
  */
 
-#ifndef __SMCC_PP_H__
-#define __SMCC_PP_H__
+#ifndef __SCC_PP_H__
+#define __SCC_PP_H__
 
-#include <libcore.h>
-#include <libutils.h>
-
-// 宏定义类型
-typedef enum {
-    MACRO_OBJECT,   // 对象宏
-    MACRO_FUNCTION, // 函数宏
-} macro_type_t;
-
-typedef VEC(cstring_t) macro_list_t;
-
-// 宏定义结构
-typedef struct smcc_macro {
-    cstring_t name;        // 宏名称
-    macro_type_t type;     // 宏类型
-    macro_list_t replaces; // 替换列表
-    macro_list_t params;   // 参数列表（仅函数宏）
-} smcc_macro_t;
+#include <pp_macro.h>
+#include <scc_core.h>
+#include <scc_utils.h>
 
 // 条件编译状态
 typedef enum {
@@ -41,12 +25,12 @@ typedef struct if_stack_item {
 } if_stack_item_t;
 
 // 预处理器状态结构
-typedef struct smcc_preprocessor {
-    core_stream_t *stream;         // 输出流
-    strpool_t strpool;             // 字符串池
-    hashmap_t macros;              // 宏定义表
-    VEC(if_stack_item_t) if_stack; // 条件编译栈
-} smcc_pp_t;
+typedef struct scc_pproc {
+    scc_probe_stream_t *stream; // 输出流
+    scc_strpool_t strpool;      // 字符串池
+    scc_pp_macro_table_t macro_table;
+    SCC_VEC(if_stack_item_t) if_stack; // 条件编译栈
+} scc_pproc_t;
 
 /**
  * @brief 初始化预处理器
@@ -54,19 +38,25 @@ typedef struct smcc_preprocessor {
  * @param[in] input 输入流对象指针
  * @return output 输出流对象指针
  */
-core_stream_t *pp_init(smcc_pp_t *pp, core_stream_t *input);
-
-/**
- * @brief 执行预处理
- * @param[in] pp 预处理器实例
- * @return 处理结果
- */
-int pp_process(smcc_pp_t *pp);
+// TODO 内存释放问题
+scc_probe_stream_t *scc_pproc_init(scc_pproc_t *pp, scc_probe_stream_t *input);
 
 /**
  * @brief 销毁预处理器
  * @param[in] pp 预处理器实例
  */
-void pp_drop(smcc_pp_t *pp);
+void scc_pproc_drop(scc_pproc_t *pp);
 
-#endif /* __SMCC_PP_H__ */
+/// inner private struct
+
+typedef SCC_VEC(u8) scc_pp_buffer_t;
+typedef struct pp_stream {
+    scc_probe_stream_t stream;
+    scc_probe_stream_t *input;
+    scc_pproc_t *self;
+
+    scc_pos_t pos;
+    scc_probe_stream_t *tmp_stream;
+} scc_pp_stream_t;
+
+#endif /* __SCC_PP_H__ */

libs/pprocessor/src/expand_macro.c

@@ -0,0 +1,412 @@
+#include <lex_parser.h>
+#include <pp_macro.h>
+#include <pp_parse.h>
+static inline void scc_generate_cstr(scc_cstring_t *buff) {
+    scc_cstring_t out_buff = scc_cstring_create();
+    scc_cstring_append_ch(&out_buff, '\"');
+    // TODO it is too simple
+    scc_cstring_append(&out_buff, buff);
+    scc_cstring_append_ch(&out_buff, '\"');
+
+    // FIXME 可能有着更好的解决方案
+    scc_cstring_clear(buff);
+    scc_cstring_append(buff, &out_buff);
+    scc_cstring_free(&out_buff);
+}
+
+#define SCC_PP_IS_LIST_BLANK(i)                                                \
+    ((i) < list->size && scc_vec_at(*list, (i)).data[0] == ' ' &&              \
+     scc_vec_at(*list, (i)).data[1] == '\0')
+#define SCC_PP_IS_LIST_TO_STRING(i)                                            \
+    ((i) < list->size && scc_vec_at(*list, (i)).data[0] == '#' &&              \
+     scc_vec_at(*list, (i)).data[1] == '\0')
+#define SCC_PP_IS_LIST_CONNECT(i)                                              \
+    ((i) < list->size && scc_vec_at(*list, (i)).data[0] == '#' &&              \
+     scc_vec_at(*list, (i)).data[1] == '#' &&                                  \
+     scc_vec_at(*list, (i)).data[2] == '\0')
+#define SCC_PP_USE_CONNECT(font, rear)                                         \
+    if (rear < list->size) {                                                   \
+        scc_cstring_append(out_buff, &scc_vec_at(*list, font));                \
+        scc_cstring_append(out_buff, &scc_vec_at(*list, rear));                \
+    } else {                                                                   \
+        scc_cstring_append(out_buff, &scc_vec_at(*list, font));                \
+    }
+// for # ## to generator string
+static inline cbool scc_pp_expand_string_unsafe(const scc_pp_macro_list_t *list,
+                                                scc_cstring_t *out_buff) {
+    for (usize i = 0; i < list->size; ++i) {
+        if (SCC_PP_IS_LIST_BLANK(i + 1)) {
+            if (SCC_PP_IS_LIST_CONNECT(i + 2)) {
+                SCC_PP_USE_CONNECT(i, i + 3);
+                i += 3;
+                continue;
+            }
+        } else if (SCC_PP_IS_LIST_CONNECT(i + 1)) {
+            SCC_PP_USE_CONNECT(i, i + 2);
+            i += 2;
+            continue;
+        } else if (SCC_PP_IS_LIST_TO_STRING(i)) {
+            i += 1;
+            if (i < list->size) {
+                scc_generate_cstr(&scc_vec_at(*list, i));
+            } else {
+                LOG_ERROR("# need a valid literator");
+                break;
+            }
+        }
+        scc_cstring_append(out_buff, &scc_vec_at(*list, i));
+    }
+    return true;
+}
+
+// 展开对象宏
+cbool scc_pp_expand_object_macro(const scc_pp_macro_t *macro,
+                                 scc_cstring_t *out_buff) {
+    Assert(macro->type == SCC_PP_MACRO_OBJECT && macro->params.size == 0);
+    Assert(scc_cstring_is_empty(out_buff) == true);
+    // 对象宏输出替换文本并进行递归展开
+    scc_pp_expand_string_unsafe(&macro->replaces, out_buff);
+    return true;
+}
+
+// 展开函数宏
+cbool scc_pp_expand_function_macro(const scc_pp_macro_t *macro,
+                                   const scc_pp_macro_list_t *origin_params,
+                                   scc_pp_macro_list_t *params,
+                                   scc_cstring_t *out_buff) {
+    Assert(macro->type == SCC_PP_MACRO_FUNCTION);
+    Assert(out_buff != null);
+    Assert(scc_cstring_is_empty(out_buff) == true);
+    Assert(params->size == macro->params.size || params->size == 0);
+
+    scc_pp_macro_list_t list;
+    scc_vec_init(list);
+    for (usize i = 0; i < macro->replaces.size; ++i) {
+        // TODO ... __VA_ARGS__
+        for (usize j = 0; j < macro->params.size; ++j) {
+            if (scc_strcmp(
+                    scc_cstring_as_cstr(&scc_vec_at(macro->replaces, i)),
+                    scc_cstring_as_cstr(&scc_vec_at(macro->params, j))) != 0) {
+                continue;
+            }
+            if (params->size == 0) {
+                scc_vec_push(list, scc_cstring_from_cstr(""));
+                goto END;
+            }
+            Assert(&scc_vec_at(*params, j) != null);
+            scc_vec_push(list, scc_cstring_copy(&scc_vec_at(*params, j)));
+            goto END;
+        }
+        scc_vec_push(list, scc_cstring_copy(&scc_vec_at(macro->replaces, i)));
+    END:;
+    }
+    scc_pp_expand_string_unsafe(&list, out_buff);
+    for (usize i = 0; i < list.size; ++i) {
+        scc_cstring_free(&scc_vec_at(list, i));
+    }
+    scc_vec_free(list);
+    return true;
+}
+
+// 状态管理结构
+typedef struct {
+    scc_pp_macro_table_t *macros;
+    scc_pp_macro_table_t painted_blue; // 正在展开的宏
+    int depth;
+} scc_expansion_ctx_t;
+
+// 进入宏展开
+static void enter_macro_expansion(scc_expansion_ctx_t *state,
+                                  scc_pp_macro_t *macro) {
+    // 添加到活动宏集合
+    scc_pp_macro_table_set(&state->painted_blue,
+                           scc_pp_macro_new(&macro->name, macro->type));
+}
+
+// 离开宏展开（开始重新扫描）
+static void leave_macro_expansion(scc_expansion_ctx_t *state,
+                                  scc_pp_macro_t *macro) {
+    // 从活动宏移除，添加到禁用宏
+    scc_pp_macro_table_remove(&state->painted_blue, &macro->name);
+}
+
+// 检查是否可以展开
+static cbool can_expand_macro(scc_expansion_ctx_t *state,
+                              scc_pp_macro_t *macro) {
+    return scc_pp_macro_table_get(&state->painted_blue, &macro->name) == null;
+}
+
+typedef struct {
+    scc_cstring_t identifier;
+    scc_pp_macro_list_t args;
+} scc_expand_unit_t;
+
+void scc_pp_parse_expand_macro(scc_probe_stream_t *stream, scc_pos_t *pos,
+                               scc_expand_unit_t *unit) {
+    Assert(stream != null && pos != null && unit != null);
+    // TODO Assert empty unit
+    scc_cstring_init(&unit->identifier);
+    scc_vec_init(unit->args);
+    cbool ret = scc_lex_parse_identifier(stream, pos, &unit->identifier);
+    Assert(ret == true);
+    scc_probe_stream_sync(stream);
+    do {
+        if (scc_lex_parse_is_whitespace(scc_probe_stream_peek(stream))) {
+            scc_lex_parse_skip_whitespace(stream, pos);
+        }
+        if (scc_probe_stream_peek(stream) != '(') {
+            // TODO maybe error
+            // 因为没能正确表述空值
+            break;
+        }
+        // TODO maybe like f()() => maybe twice expand
+        ret = scc_pp_parse_macro_arguments(stream, &unit->args);
+        scc_probe_stream_sync(stream);
+        Assert(ret == true);
+    } while (0);
+}
+
+static cbool expand_buffer(const scc_cstring_t *in, scc_cstring_t *out,
+                           scc_expansion_ctx_t *state);
+
+static inline void args2str_append(const scc_pp_macro_list_t *args,
+                                   scc_cstring_t *out) {
+    Assert(args != null);
+    if (args->size != 0) {
+        scc_cstring_append_ch(out, '(');
+        for (usize i = 0; i < args->size; ++i) {
+            scc_cstring_append(out, &scc_vec_at(*args, i));
+            if (i != args->size - 1) {
+                scc_cstring_append_ch(out, ',');
+                scc_cstring_append_ch(out, ' ');
+            }
+        }
+        scc_cstring_append_ch(out, ')');
+    }
+}
+
+static cbool expand_macro(const scc_expand_unit_t *unit, scc_cstring_t *out,
+                          scc_expansion_ctx_t *state) {
+    scc_pp_macro_t *macro =
+        scc_pp_macro_table_get(state->macros, &unit->identifier);
+    if (macro == null) {
+        scc_cstring_append(out, &unit->identifier);
+        args2str_append(&unit->args, out);
+        return true;
+    }
+    cbool ret;
+    if (macro->type == SCC_PP_MACRO_OBJECT) {
+        if (can_expand_macro(state, macro)) {
+            ret = scc_pp_expand_object_macro(macro, out);
+            Assert(ret == true);
+            args2str_append(&unit->args, out);
+        } else {
+            scc_cstring_append(out, &macro->name);
+            args2str_append(&unit->args, out);
+            return true;
+        }
+    } else if (macro->type == SCC_PP_MACRO_FUNCTION) {
+        /**
+         * 1. 参数先展开
+         * 2. 替换后重扫描
+         * 3. 蓝色集合中不展开
+         * 4. #, ## 不展开
+         * 5. 最后的括号要检查
+         */
+
+        scc_pp_macro_list_t expanded_params;
+        scc_vec_init(expanded_params);
+        // expand params fisrt and recursive
+        for (usize i = 0; i < unit->args.size; ++i) {
+            scc_cstring_t param = scc_vec_at(unit->args, i);
+            scc_cstring_t out = scc_cstring_create();
+            expand_buffer(&param, &out, state);
+            scc_vec_push(expanded_params, out);
+        }
+
+        if (can_expand_macro(state, macro)) {
+            ret = scc_pp_expand_function_macro(macro, &unit->args,
+                                               &expanded_params, out);
+            Assert(ret == true);
+        } else {
+            scc_cstring_append(out, &macro->name);
+            args2str_append(&expanded_params, out);
+            return true;
+        }
+        // TODO memory leak
+    } else {
+        UNREACHABLE();
+    }
+
+    // 重新扫描展开结果
+    scc_cstring_t rescanned = scc_cstring_create();
+    enter_macro_expansion(state, macro);
+    expand_buffer(out, &rescanned, state);
+    leave_macro_expansion(state, macro);
+
+    scc_cstring_free(out);
+    *out = rescanned;
+    return true;
+}
+
+static cbool expand_buffer(const scc_cstring_t *in, scc_cstring_t *out,
+                           scc_expansion_ctx_t *state) {
+    scc_probe_stream_t *in_stream = scc_mem_probe_stream_alloc(
+        scc_cstring_as_cstr(in), scc_cstring_len(in), false);
+
+    int ch;
+    scc_pos_t pos;
+    scc_expand_unit_t unit;
+    while ((ch = scc_probe_stream_peek(in_stream)) != scc_stream_eof) {
+        if (scc_lex_parse_is_identifier_prefix(ch)) {
+            // 递归检查
+            scc_cstring_t expanded_buffer = scc_cstring_create();
+            scc_pp_parse_expand_macro(in_stream, &pos, &unit);
+            scc_probe_stream_reset(in_stream);
+            if (expand_macro(&unit, &expanded_buffer, state) == false) {
+                scc_cstring_free(&expanded_buffer);
+                return false;
+            } else {
+                scc_cstring_append(out, &expanded_buffer);
+                scc_cstring_free(&expanded_buffer);
+            }
+        } else {
+            scc_cstring_append_ch(out, scc_probe_stream_consume(in_stream));
+        }
+    }
+
+    scc_probe_stream_drop(in_stream);
+    return true;
+}
+
+// static cbool scc_pp_expand_macro_impl(scc_probe_stream_t *stream,
+//                                       scc_expansion_ctx_t *ctx, scc_pos_t
+//                                       *pos, scc_cstring_t *output) {
+//     // TODO self position and it maybe is a stack on #include ?
+//     // 递归扫描
+//     if (ctx->depth-- <= 0) {
+//         scc_cstring_free(output);
+//         return false;
+//     }
+//     cbool ret;
+//     scc_pp_macro_table_t *macros = ctx->macros;
+
+//     scc_cstring_t identifier = scc_cstring_create();
+//     ret = scc_lex_parse_identifier(stream, pos, &identifier);
+//     Assert(ret == true);
+
+//     scc_pp_macro_t *macro = scc_pp_macro_table_get(macros, &identifier);
+
+//     // 1. 不是宏，直接输出标识符
+//     if (macro == null) {
+//         // 不是宏，直接输出
+//         usize length = scc_cstring_len(&identifier);
+//         *out_stream = scc_mem_probe_stream_alloc(
+//             scc_cstring_move_cstr(&identifier), length, true);
+//         return true;
+//     } else {
+//         scc_cstring_free(&identifier);
+//     }
+
+//     // 收集参数（如果是函数宏）
+//     scc_pp_macro_list_t params;
+//     scc_pp_macro_list_t expanded_params;
+//     scc_vec_init(params);
+//     scc_vec_init(expanded_params);
+//     if (macro->type == SCC_PP_MACRO_FUNCTION) {
+//         // TODO when expand need check another func with () at the end
+//         scc_lex_parse_skip_whitespace(stream, &pos);
+//         if (scc_probe_stream_peek(stream) != '(') {
+//             goto ORIGIN;
+//         }
+//         ret = scc_pp_parse_macro_arguments(stream, &params);
+//         Assert(ret == true);
+
+//         // expand params fisrt and recursive
+//         for (usize i = 0; i < params.size; ++i) {
+//             scc_cstring_t param = scc_vec_at(params, i);
+//             scc_cstring_t out = scc_cstring_create();
+//             expand_buffer(&param, &out, ctx);
+//             scc_vec_push(expanded_params, out);
+//         }
+//     }
+
+//     // 2. 检查到重复展开跳过
+//     // 检查是否可以展开
+//     if (!can_expand_macro(ctx, macro)) {
+//     ORIGIN:
+//         // 输出原始调用
+//         scc_cstring_t original = scc_cstring_create();
+//         scc_cstring_append(&original, &macro->name);
+//         if (macro->type == SCC_PP_MACRO_FUNCTION && expanded_params.size !=
+//         0) {
+//             scc_cstring_append_ch(&original, '(');
+//             for (usize i = 0; i < expanded_params.size; ++i) {
+//                 scc_cstring_append(&original, &scc_vec_at(expanded_params,
+//                 i)); if (i != expanded_params.size - 1) {
+//                     scc_cstring_append_ch(&original, ',');
+//                     scc_cstring_append_ch(&original, ' ');
+//                 }
+//             }
+//             scc_cstring_append_ch(&original, ')');
+//         }
+//         *out_stream = scc_mem_probe_stream_alloc(
+//             scc_cstring_as_cstr(&original), scc_cstring_len(&original),
+//             true);
+//         scc_vec_free(params);
+//         scc_vec_free(expanded_params);
+//         return true;
+//     }
+
+//     // 开始展开
+//     scc_cstring_t expanded = scc_cstring_create();
+//     if (macro->type == SCC_PP_MACRO_OBJECT) {
+//         ret = scc_pp_expand_object_macro(macro, &expanded);
+//         Assert(ret == true);
+//     } else if (macro->type == SCC_PP_MACRO_FUNCTION) {
+//         ret = scc_pp_expand_function_macro(macro, &params, &expanded_params,
+//                                            &expanded);
+//         Assert(ret == true);
+//     } else {
+//         UNREACHABLE();
+//     }
+
+//     // 重新扫描展开结果
+//     // 将展开内容变换成stream并递归展开
+//     scc_cstring_t rescanned = scc_cstring_create();
+//     enter_macro_expansion(ctx, macro);
+//     expand_buffer(&expanded, &rescanned, ctx);
+//     leave_macro_expansion(ctx, macro);
+
+//     scc_cstring_free(&expanded);
+//     // TODO memory leak
+
+//     *out_stream = scc_mem_probe_stream_alloc(scc_cstring_as_cstr(&rescanned),
+//                                              scc_cstring_len(&rescanned),
+//                                              true);
+//     return true;
+// }
+
+cbool scc_pp_expand_macro(scc_probe_stream_t *stream,
+                          scc_pp_macro_table_t *macros,
+                          scc_probe_stream_t **out_stream, int depth) {
+    Assert(depth > 0 && stream != null && macros != null && out_stream != null);
+    scc_expansion_ctx_t state;
+    state.depth = depth;
+    scc_pp_marco_table_init(&state.painted_blue);
+    state.macros = macros;
+    scc_cstring_t buffer = scc_cstring_create();
+    scc_pos_t pos;
+    // cbool ret = scc_pp_expand_macro_impl(stream, &state, &pos, &buffer);
+
+    // expand_buffer(stream, &buffer, &state);
+    scc_expand_unit_t unit;
+    scc_pp_parse_expand_macro(stream, &pos, &unit);
+    scc_probe_stream_reset(stream);
+    expand_macro(&unit, &buffer, &state);
+
+    *out_stream = scc_mem_probe_stream_alloc(scc_cstring_as_cstr(&buffer),
+                                             scc_cstring_len(&buffer), true);
+    scc_pp_macro_table_drop(&state.painted_blue);
+    return true;
+}

libs/pprocessor/src/macro.c

@@ -0,0 +1,155 @@
+#include <pp_macro.h>
+
+// 创建宏对象
+scc_pp_macro_t *scc_pp_macro_new(const scc_cstring_t *name,
+                                 scc_pp_macro_type_t type) {
+    scc_pp_macro_t *macro = scc_malloc(sizeof(scc_pp_macro_t));
+    if (!macro) {
+        LOG_ERROR("Failed to allocate memory for macro");
+        return null;
+    }
+
+    macro->name = scc_cstring_copy(name);
+    macro->type = type;
+    scc_vec_init(macro->params);
+    scc_vec_init(macro->replaces);
+
+    return macro;
+}
+
+// 销毁宏对象
+void scc_pp_macro_drop(scc_pp_macro_t *macro) {
+    if (!macro)
+        return;
+
+    scc_cstring_free(&macro->name);
+
+    // 释放参数列表
+    for (usize i = 0; i < macro->params.size; ++i) {
+        scc_cstring_free(&scc_vec_at(macro->params, i));
+    }
+    scc_vec_free(macro->params);
+
+    // 释放替换列表
+    for (usize i = 0; i < macro->replaces.size; ++i) {
+        scc_cstring_free(&scc_vec_at(macro->replaces, i));
+    }
+    scc_vec_free(macro->replaces);
+
+    scc_free(macro);
+}
+
+// 添加对象宏
+cbool scc_pp_add_object_macro(scc_pp_macro_table_t *macros,
+                              const scc_cstring_t *name,
+                              const scc_pp_macro_list_t *replacement) {
+    if (!macros || !name || !replacement)
+        return false;
+
+    scc_pp_macro_t *macro = scc_pp_macro_new(name, SCC_PP_MACRO_OBJECT);
+    if (!macro)
+        return false;
+
+    macro->replaces = *replacement;
+
+    // 检查是否已存在同名宏
+    scc_pp_macro_t *existing = scc_hashtable_get(&macros->table, &macro->name);
+    if (existing) {
+        LOG_WARN("Redefining macro: %s", scc_cstring_as_cstr(&macro->name));
+        scc_pp_macro_drop(existing);
+    }
+
+    scc_hashtable_set(&macros->table, &macro->name, macro);
+    return true;
+}
+
+// 添加函数宏
+cbool scc_pp_add_function_macro(scc_pp_macro_table_t *macros,
+                                const scc_cstring_t *name,
+                                const scc_pp_macro_list_t *params,
+                                const scc_pp_macro_list_t *replacement) {
+    if (!macros || !name || !params || !replacement)
+        return false;
+
+    scc_pp_macro_t *macro = scc_pp_macro_new(name, SCC_PP_MACRO_FUNCTION);
+    if (!macro)
+        return false;
+
+    // 复制参数列表
+    macro->params = *params;
+    macro->replaces = *replacement;
+
+    // 检查是否已存在同名宏
+    scc_pp_macro_t *existing = scc_hashtable_get(&macros->table, &macro->name);
+    if (existing) {
+        LOG_WARN("Redefining macro: %s", scc_cstring_as_cstr(&macro->name));
+        scc_pp_macro_drop(existing);
+    }
+
+    scc_hashtable_set(&macros->table, &macro->name, macro);
+    return true;
+}
+
+/// marco_table
+
+scc_pp_macro_t *scc_pp_macro_table_set(scc_pp_macro_table_t *pp,
+                                       scc_pp_macro_t *macro) {
+    Assert(pp != null && macro != null);
+    return scc_hashtable_set(&pp->table, &macro->name, macro);
+}
+
+// 查找宏定义
+scc_pp_macro_t *scc_pp_macro_table_get(scc_pp_macro_table_t *pp,
+                                       const scc_cstring_t *name) {
+    return scc_hashtable_get(&pp->table, name);
+}
+
+// 从预处理器中删除宏
+cbool scc_pp_macro_table_remove(scc_pp_macro_table_t *pp,
+                                const scc_cstring_t *name) {
+    if (!pp || !name)
+        return false;
+
+    scc_pp_macro_t *macro = scc_hashtable_get(&pp->table, name);
+    if (!macro)
+        return false;
+
+    scc_hashtable_del(&pp->table, name);
+    scc_pp_macro_drop(macro);
+    return true;
+}
+
+static u32 hash_func(const void *key) {
+    const scc_cstring_t *string = (const scc_cstring_t *)key;
+    return scc_strhash32(scc_cstring_as_cstr(string));
+}
+
+static int hash_cmp(const void *key1, const void *key2) {
+    const scc_cstring_t *str1 = (const scc_cstring_t *)key1;
+    const scc_cstring_t *str2 = (const scc_cstring_t *)key2;
+
+    if (str1->size != str2->size) {
+        return str1->size - str2->size;
+    }
+    return scc_strcmp(scc_cstring_as_cstr(str1), scc_cstring_as_cstr(str2));
+}
+
+void scc_pp_marco_table_init(scc_pp_macro_table_t *macros) {
+    Assert(macros != null);
+    macros->table.hash_func = hash_func;
+    macros->table.key_cmp = hash_cmp;
+    scc_hashtable_init(&macros->table);
+}
+
+static int macro_free(const void *key, void *value, void *context) {
+    (void)key;
+    (void)context;
+    scc_pp_macro_drop(value);
+    return 0;
+}
+
+void scc_pp_macro_table_drop(scc_pp_macro_table_t *macros) {
+    Assert(macros != null);
+    scc_hashtable_foreach(&macros->table, macro_free, null);
+    scc_hashtable_drop(&macros->table);
+}

libs/pprocessor/src/parse.c

@@ -0,0 +1,295 @@
+#include <lex_parser.h>
+#include <pp_macro.h>
+#include <pp_parse.h>
+#include <pp_token.h>
+
+static const struct {
+    const char *name;
+    scc_pp_token_t tok;
+} keywords[] = {
+#define X(name, type, tok) {#name, tok},
+    SCC_PP_INST_TOKEN
+#undef X
+};
+
+// 使用二分查找查找关键字
+static inline int keyword_cmp(const char *name, int len) {
+    int low = 0;
+    int high = sizeof(keywords) / sizeof(keywords[0]) - 1;
+    while (low <= high) {
+        int mid = (low + high) / 2;
+        const char *key = keywords[mid].name;
+        int cmp = 0;
+
+        // 自定义字符串比较逻辑
+        for (int i = 0; i < len; i++) {
+            if (name[i] != key[i]) {
+                cmp = (unsigned char)name[i] - (unsigned char)key[i];
+                break;
+            }
+            if (name[i] == '\0')
+                break; // 遇到终止符提前结束
+        }
+
+        if (cmp == 0) {
+            // 完全匹配检查（长度相同）
+            if (key[len] == '\0')
+                return mid;
+            cmp = -1; // 当前关键词比输入长
+        }
+
+        if (cmp < 0) {
+            high = mid - 1;
+        } else {
+            low = mid + 1;
+        }
+    }
+    return -1; // Not a keyword.
+}
+
+static inline void try_to_cut_list(scc_pp_macro_list_t *list,
+                                   scc_cstring_t *buff) {
+    if (scc_cstring_len(buff) != 0) {
+        scc_vec_push(*list, *buff);
+        *buff = scc_cstring_create();
+    }
+}
+
+cbool scc_pp_parse_macro_replace_list(scc_probe_stream_t *stream,
+                                      scc_pp_macro_list_t *list) {
+    Assert(stream != null && list != null);
+    // scc_probe_stream_reset(stream);
+
+    scc_vec_init(*list);
+    scc_cstring_t replacement = scc_cstring_create();
+    int ch;
+    scc_pos_t pos = scc_pos_create();
+
+    while ((ch = scc_probe_stream_peek(stream)) != scc_stream_eof) {
+        if (scc_lex_parse_is_endline(ch)) {
+            break;
+        }
+        if (scc_lex_parse_is_identifier_prefix(ch)) {
+            try_to_cut_list(list, &replacement);
+            cbool ret = scc_lex_parse_identifier(stream, &pos, &replacement);
+            Assert(ret == true);
+            try_to_cut_list(list, &replacement);
+        } else if (ch == '#') {
+            // 处理 # 和 ## 操作符
+            scc_probe_stream_next(stream);
+            try_to_cut_list(list, &replacement);
+
+            scc_cstring_append_ch(&replacement, '#');
+            if (scc_probe_stream_peek(stream) == '#') {
+                // ## 连接操作符
+                scc_probe_stream_next(stream);
+                scc_cstring_append_ch(&replacement, '#');
+            }
+            // 我需要尽可能防止空白字符干扰解析
+            scc_lex_parse_skip_whitespace(stream, &pos);
+            try_to_cut_list(list, &replacement);
+        } else if (scc_lex_parse_is_whitespace(ch)) {
+            try_to_cut_list(list, &replacement);
+            scc_lex_parse_skip_whitespace(stream, &pos);
+            scc_cstring_append_ch(&replacement, ' ');
+            try_to_cut_list(list, &replacement);
+        } else {
+            scc_probe_stream_next(stream);
+            scc_cstring_append_ch(&replacement, (char)ch);
+        }
+    }
+
+    if (scc_cstring_len(&replacement) != 0) {
+        scc_vec_push(*list, replacement);
+        replacement = scc_cstring_create();
+    }
+
+    // for (usize i = 0; i < list->size; ++i) {
+    //     LOG_DEBUG("list %d: %s", (int)i,
+    //               scc_cstring_as_cstr(&scc_vec_at(*list, i)));
+    // }
+    return true;
+}
+
+// 解析宏参数列表
+cbool scc_pp_parse_macro_arguments(scc_probe_stream_t *stream,
+                                   scc_pp_macro_list_t *args) {
+    Assert(stream != null && args != null);
+
+    scc_vec_init(*args);
+    int ch;
+    // scc_probe_stream_reset(stream);
+
+    // 跳过 '('
+    ch = scc_probe_stream_peek(stream);
+    if (ch != '(') {
+        return false;
+    }
+    scc_probe_stream_next(stream); // 消费 '('
+
+    int paren_depth = 1;
+    scc_cstring_t current_arg = scc_cstring_create();
+    scc_pos_t pos = scc_pos_create();
+
+    while (paren_depth > 0) {
+        ch = scc_probe_stream_peek(stream);
+        if (ch == scc_stream_eof) {
+            scc_cstring_free(&current_arg);
+            scc_cstring_free(&pos.name);
+            return false;
+        }
+
+        if (ch == '(') {
+            paren_depth++;
+            scc_cstring_append_ch(&current_arg, (char)ch);
+            scc_probe_stream_next(stream);
+        } else if (ch == ')') {
+            paren_depth--;
+            if (paren_depth > 0) {
+                scc_cstring_append_ch(&current_arg, (char)ch);
+            }
+            scc_probe_stream_next(stream);
+        } else if (ch == ',' && paren_depth == 1) {
+            // 参数分隔符
+            scc_vec_push(*args, current_arg);
+            current_arg = scc_cstring_create();
+            scc_probe_stream_next(stream);
+            // 跳过参数后的空白
+            scc_lex_parse_skip_whitespace(stream, &pos);
+        } else {
+            scc_cstring_append_ch(&current_arg, (char)ch);
+            scc_probe_stream_next(stream);
+        }
+    }
+
+    // 添加最后一个参数
+    if (!scc_cstring_is_empty(&current_arg)) {
+        scc_vec_push(*args, current_arg);
+    } else {
+        scc_cstring_free(&current_arg);
+    }
+
+    scc_cstring_free(&pos.name);
+    return true;
+}
+
+static cbool safe_skip_backspace_if_endline(scc_probe_stream_t *stream,
+                                            scc_pos_t *pos) {
+    // scc_probe_stream_reset(stream);
+    int ch = scc_probe_stream_peek(stream);
+    // FIXME maybe it not correct
+    while (ch == '\r' || ch == '\n' || ch == ' ' || ch == '\t') {
+        if (scc_lex_parse_is_endline(ch)) {
+            scc_lex_parse_skip_endline(stream, pos);
+            return true;
+        }
+        scc_probe_stream_next(stream);
+        ch = scc_probe_stream_peek(stream);
+    }
+    // scc_probe_stream_reset(stream);
+    return false;
+}
+
+void scc_pp_parse_directive(scc_probe_stream_t *stream, scc_pos_t *pos,
+                            scc_pp_macro_table_t *macros) {
+    Assert(stream != null);
+
+    // scc_probe_stream_reset(stream);
+    // 跳过 '#' 和后续空白
+    if (scc_probe_stream_peek(stream) != '#') {
+        LOG_WARN("Invalid directive");
+        return;
+    }
+    scc_pos_next(pos);
+    scc_probe_stream_next(stream);
+    if (safe_skip_backspace_if_endline(stream, pos))
+        return;
+
+    // 解析指令名称
+    scc_cstring_t directive = scc_cstring_create();
+    if (!scc_lex_parse_identifier(stream, pos, &directive)) {
+        goto ERR;
+    }
+    if (safe_skip_backspace_if_endline(stream, pos))
+        goto FREE;
+
+    scc_pp_token_t token = keyword_cmp(scc_cstring_as_cstr(&directive),
+                                       scc_cstring_len(&directive));
+
+    scc_cstring_t name = scc_cstring_create();
+    switch (token) {
+    case SCC_PP_TOK_DEFINE: {
+        if (!scc_lex_parse_identifier(stream, pos, &name)) {
+            goto ERR;
+        }
+
+        // 检查是否是函数宏：宏名后是否直接跟着 '('（没有空白字符）
+        // scc_probe_stream_reset(stream);
+        int ch = scc_probe_stream_peek(stream);
+        cbool has_whitespace = scc_lex_parse_is_whitespace(ch);
+        if (has_whitespace && safe_skip_backspace_if_endline(stream, pos)) {
+            goto FREE;
+        }
+
+        if (!has_whitespace && ch == '(') {
+            // 函数宏
+            scc_pp_macro_list_t params;
+            if (!scc_pp_parse_macro_arguments(stream, &params)) {
+                goto ERR;
+            }
+
+            ch = scc_probe_stream_peek(stream);
+            if (ch == ')') {
+                scc_probe_stream_next(stream); // 消费 ')'
+            }
+            if (safe_skip_backspace_if_endline(stream, pos)) {
+                goto FREE;
+            }
+
+            scc_pp_macro_list_t replacement;
+            scc_pp_parse_macro_replace_list(stream, &replacement);
+            scc_pp_add_function_macro(macros, &name, &params, &replacement);
+        } else {
+            // 对象宏
+            scc_pp_macro_list_t replacement;
+            scc_pp_parse_macro_replace_list(stream, &replacement);
+            scc_pp_add_object_macro(macros, &name, &replacement);
+        }
+        break;
+    }
+    case SCC_PP_TOK_UNDEF: {
+        if (scc_lex_parse_identifier(stream, pos, &name)) {
+            cbool ret = scc_pp_macro_table_remove(macros, &name);
+            Assert(ret == true);
+        }
+        break;
+    }
+    case SCC_PP_TOK_INCLUDE:
+    case SCC_PP_TOK_IF:
+    case SCC_PP_TOK_IFDEF:
+    case SCC_PP_TOK_IFNDEF:
+    case SCC_PP_TOK_ELSE:
+    case SCC_PP_TOK_ELIF:
+    case SCC_PP_TOK_ELIFDEF:
+    case SCC_PP_TOK_ELIFNDEF:
+    case SCC_PP_TOK_ENDIF:
+    case SCC_PP_TOK_LINE:
+    case SCC_PP_TOK_EMBED:
+    case SCC_PP_TOK_ERROR:
+    case SCC_PP_TOK_WARNING:
+    case SCC_PP_TOK_PRAGMA:
+        // 暂时跳过这一行
+        scc_lex_parse_skip_line(stream, pos);
+        TODO();
+        break;
+    default:
+        LOG_WARN("Unknown preprocessor directive: %s",
+                 scc_cstring_as_cstr(&directive));
+        scc_lex_parse_skip_line(stream, pos);
+    }
+ERR:
+    scc_lex_parse_skip_line(stream, pos);
+FREE:
+    scc_cstring_free(&directive);
+    scc_cstring_free(&name);
+}

libs/pprocessor/src/pprocessor.c

@@ -4,424 +4,129 @@
  */
 
 #include <lex_parser.h>
+#include <pp_macro.h>
+#include <pp_parse.h>
 #include <pp_token.h>
 #include <pprocessor.h>
-#define PPROCESSER_BUFFER_SIZE (1024)
 
-static u32 hash_func(cstring_t *string) {
-    return smcc_strhash32(cstring_as_cstr(string));
-}
+#ifdef TEST_MODE
+#define MAX_MACRO_EXPANSION_DEPTH 16
+#else
+#define MAX_MACRO_EXPANSION_DEPTH 64 // 防止无限递归的最大展开深度
+#endif
 
-static int hash_cmp(const cstring_t *str1, const cstring_t *str2) {
-    if (str1->size != str2->size) {
-        return str1->size - str2->size;
-    }
-
-    return smcc_strcmp(cstring_as_cstr(str1), cstring_as_cstr(str2));
-}
-
-// 添加宏定义
-static void add_macro(smcc_pp_t *pp, const cstring_t *name,
-                      const macro_list_t *replaces, const macro_list_t *params,
-                      macro_type_t type) {
-    smcc_macro_t *macro = smcc_malloc(sizeof(smcc_macro_t));
-    macro->name = *name;
-    macro->type = type;
-
-    if (replaces) {
-        macro->replaces = *replaces;
-    } else {
-        vec_init(macro->replaces);
-    }
-
-    if (params) {
-        macro->params = *params;
-    } else {
-        vec_init(macro->params);
-    }
-
-    hashmap_set(&pp->macros, &macro->name, macro);
-}
-
-// 查找宏定义
-static smcc_macro_t *find_macro(smcc_pp_t *pp, cstring_t *name) {
-    return hashmap_get(&pp->macros, name);
-}
-
-// 条件编译处理框架
-static void handle_if(smcc_pp_t *pp, const char *condition) {
-    if_stack_item_t item;
-    int cond_value;
-    // cond_value = evaluate_condition(pp, condition);
-
-    item.state = cond_value ? IFState_TRUE : IFState_FALSE;
-    item.skip = !cond_value;
-    vec_push(pp->if_stack, item);
-}
-
-static void handle_else(smcc_pp_t *pp) {
-    if (pp->if_stack.size == 0) {
-        // 错误：没有匹配的#if
-        return;
-    }
-
-    if_stack_item_t *top = &vec_at(pp->if_stack, pp->if_stack.size - 1);
-    if (top->state == IFState_ELSE) {
-        // 错误：#else重复出现
-        return;
-    }
-
-    top->skip = !top->skip;
-    top->state = IFState_ELSE;
-}
-
-static void handle_include(smcc_pp_t *pp, const char *filename,
-                           int system_header) {
-    // 查找文件路径逻辑
-    // 创建新的输入流
-    // 递归处理包含文件
-}
-
-// 解析标识符
-static cstring_t parse_identifier(core_stream_t *stream) {
-    cstring_t identifier = cstring_new();
-    core_stream_reset_char(stream);
-    int ch = core_stream_peek_char(stream);
-
-    // 标识符以字母或下划线开头
-    if (!((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z') || ch == '_')) {
-        LOG_WARN("Invalid identifier");
-        return identifier;
-    }
-    do {
-        cstring_push(&identifier, (char)ch);
-        core_stream_next_char(stream); // 消费字符
-        ch = core_stream_peek_char(stream);
-    } while ((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z') ||
-             (ch >= '0' && ch <= '9') || ch == '_');
-
-    return identifier;
-}
-
-// 跳过空白字符 ' ' and '\t'
-static void skip_whitespace(core_stream_t *stream) {
-    int ch;
-    core_stream_reset_char(stream);
-    while ((ch = core_stream_peek_char(stream)) != core_stream_eof) {
-        if (ch == ' ' || ch == '\t') {
-            core_stream_next_char(stream);
-        } else {
-            break;
-        }
-    }
-}
-
-#define X(name, type, tok) SMCC_STR(name),
-static const char *token_strings[] = {PP_INST_TOKEN};
-#undef X
-
-static const struct {
-    const char *name;
-    pp_token_t tok;
-} keywords[] = {
-#define X(name, type, tok) {#name, tok},
-    PP_INST_TOKEN
-#undef X
-};
-
-// by using binary search to find the keyword
-static inline int keyword_cmp(const char *name, int len) {
-    int low = 0;
-    int high = sizeof(keywords) / sizeof(keywords[0]) - 1;
-    while (low <= high) {
-        int mid = (low + high) / 2;
-        const char *key = keywords[mid].name;
-        int cmp = 0;
-
-        // 自定义字符串比较逻辑
-        for (int i = 0; i < len; i++) {
-            if (name[i] != key[i]) {
-                cmp = (unsigned char)name[i] - (unsigned char)key[i];
-                break;
-            }
-            if (name[i] == '\0')
-                break; // 遇到终止符提前结束
-        }
-
-        if (cmp == 0) {
-            // 完全匹配检查（长度相同）
-            if (key[len] == '\0')
-                return mid;
-            cmp = -1; // 当前关键词比输入长
-        }
-
-        if (cmp < 0) {
-            high = mid - 1;
-        } else {
-            low = mid + 1;
-        }
-    }
-    return -1; // Not a keyword.
-}
-
-typedef struct pp_stream {
-    core_stream_t stream;
-    core_stream_t *input;
-    smcc_pp_t *self;
-
-    usize size;
-    usize pos;
-    char buffer[PPROCESSER_BUFFER_SIZE];
-} pp_stream_t;
-
-static cbool parse_list(pp_stream_t *_stream, macro_list_t *list,
-                        cbool is_param) {
-    Assert(_stream != null);
-    core_stream_t *stream = _stream->input;
+static int pp_stream_read_char(scc_probe_stream_t *_stream) {
+    scc_pp_stream_t *stream = (scc_pp_stream_t *)_stream;
     Assert(stream != null);
-    core_stream_reset_char(stream);
-
-    vec_init(*list);
     int ch;
-    cstring_t str = cstring_new();
-    core_pos_t pos;
 
-    while ((ch = core_stream_peek_char(stream)) != core_stream_eof) {
-        if (is_param) {
-            // ( 参数 ) ( 参数, ... ) ( ... )
-            if (lex_parse_is_whitespace(ch)) {
-                // TODO #define ( A A , B ) need ERROR
-                lex_parse_skip_whitespace(stream, &pos);
-                core_stream_reset_char(stream);
-            } else if (ch == ',') {
-                vec_push(*list, str);
-                str = cstring_new();
-                core_stream_next_char(stream);
-                continue;
-            } else if (ch == ')') {
-                break;
-            } else if (ch == core_stream_eof || lex_parse_is_endline(ch)) {
-                LOG_ERROR("Invalid parameter list");
-                return false;
-            }
-        } else {
-            // 替换列表
-            if (lex_parse_is_whitespace(ch)) {
-                lex_parse_skip_whitespace(stream, &pos);
-                vec_push(*list, str);
-                str = cstring_new();
-                core_stream_reset_char(stream);
-                continue;
-            } else if (lex_parse_is_endline(ch)) {
-                break;
-            }
-        }
-        core_stream_next_char(stream);
-        cstring_push(&str, (char)ch);
-    }
-    vec_push(*list, str);
-    str = cstring_new();
-    return true;
-}
-
-// 解析预处理指令
-static void parse_directive(pp_stream_t *_stream) {
-    Assert(_stream != null);
-    core_stream_t *stream = _stream->input;
-    Assert(stream != null);
-
-    int ch;
-    core_pos_t pos;
-    core_stream_reset_char(stream);
-    // 跳过 '#' 和后续空白
-    if (core_stream_peek_char(stream) != '#') {
-        LOG_WARN("Invalid directive");
-        return;
-    }
-    core_stream_next_char(stream);
-
-    // TODO 允许空指令（# 后跟换行符），且无任何效果。
-    skip_whitespace(stream);
-    // 解析指令名称
-    cstring_t directive = parse_identifier(stream);
-    if (cstring_is_empty(&directive)) {
-        LOG_ERROR("expected indentifier");
-        goto ERR;
-    }
-    skip_whitespace(stream);
-    core_stream_reset_char(stream);
-
-    pp_token_t token =
-        keyword_cmp(cstring_as_cstr(&directive), cstring_len(&directive));
-    switch (token) {
-    case PP_TOK_DEFINE: {
-        cstring_t name = parse_identifier(stream);
-        if (cstring_is_empty(&name)) {
-            LOG_ERROR("expected indentifier");
-            goto ERR;
-        }
-        skip_whitespace(stream);
-        core_stream_reset_char(stream);
-
-        int ch = core_stream_peek_char(stream);
-        if (ch == '(') {
-            macro_list_t params;
-            parse_list(_stream, &params, true);
-            ch = core_stream_next_char(stream);
-            if (ch != ')') {
-            }
-            goto ERR;
-        }
-        macro_list_t replacement;
-        parse_list(_stream, &replacement, false);
-        add_macro(_stream->self, &name, &replacement, NULL, MACRO_OBJECT);
-        break;
-    }
-    case PP_TOK_UNDEF:
-    case PP_TOK_INCLUDE:
-    case PP_TOK_IF:
-    case PP_TOK_IFDEF:
-    case PP_TOK_IFNDEF:
-    case PP_TOK_ELSE:
-    case PP_TOK_ELIF:
-    case PP_TOK_ELIFDEF:
-    case PP_TOK_ELIFNDEF:
-    case PP_TOK_ENDIF:
-    case PP_TOK_LINE:
-    case PP_TOK_EMBED:
-    case PP_TOK_ERROR:
-    case PP_TOK_WARNING:
-    case PP_TOK_PRAMA:
-        TODO();
-        break;
-    default:
-        LOG_WARN("Unknown preprocessor directive: %s",
-                 cstring_as_cstr(&directive));
-    }
-
-    // TODO: win \r\n linux \n mac \r => all need transport to \n
-    core_stream_reset_char(stream);
-    lex_parse_skip_line(stream, &pos);
-
-    cstring_free(&directive);
-    return;
-ERR:
-    // TODO skip line
-    LOG_FATAL("Unhandled preprocessor directive");
-}
-
-static inline void stream_push_string(pp_stream_t *stream, cstring_t *str) {
-    stream->size += cstring_len(str);
-    Assert(stream->size <= PPROCESSER_BUFFER_SIZE);
-    smcc_memcpy(stream->buffer, cstring_as_cstr(str), stream->size);
-}
-
-static inline void stream_push_char(pp_stream_t *stream, int ch) {
-    stream->buffer[stream->size++] = ch;
-    Assert(stream->size <= PPROCESSER_BUFFER_SIZE);
-}
-
-static int next_char(core_stream_t *_stream) {
-    pp_stream_t *stream = (pp_stream_t *)_stream;
-    Assert(stream != null);
 READ_BUF:
-    if (stream->size != 0) {
-        if (stream->pos < stream->size) {
-            return stream->buffer[stream->pos++];
-        } else {
-            stream->size = 0;
-            stream->pos = 0;
-        }
+    if (stream->tmp_stream != null &&
+        (ch = scc_probe_stream_consume(stream->tmp_stream)) != scc_stream_eof) {
+        return ch;
     }
 RETRY:
-    core_stream_reset_char(stream->input);
-    int ch = core_stream_peek_char(stream->input);
+    scc_probe_stream_reset(stream->input);
+    ch = scc_probe_stream_peek(stream->input);
+
     if (ch == '#') {
-        parse_directive(stream);
+        scc_pp_parse_directive(stream->input, &stream->pos,
+                               &stream->self->macro_table);
+        scc_probe_stream_sync(stream->input);
         goto RETRY;
-    } else if ((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z') ||
-               ch == '_') {
-        cstring_t identifier = parse_identifier(stream->input);
-        smcc_macro_t *macro = find_macro(stream->self, &identifier);
-        if (macro == null) {
-            stream_push_string(stream, &identifier);
-            cstring_free(&identifier);
-            goto READ_BUF;
-        } else {
-            cstring_free(&identifier);
+    } else if (scc_lex_parse_is_identifier_prefix(ch)) {
+        cbool ret =
+            scc_pp_expand_macro(stream->input, &stream->self->macro_table,
+                                &stream->tmp_stream, MAX_MACRO_EXPANSION_DEPTH);
+        if (ret == false) {
+            LOG_ERROR("macro_expand_error");
         }
-        if (macro->type == MACRO_OBJECT) {
-            for (usize i = 0; i < macro->replaces.size; ++i) {
-                stream_push_string(stream, &vec_at(macro->replaces, i));
-                // usize never using `-`
-                if (i + 1 < macro->replaces.size)
-                    stream_push_char(stream, ' ');
-            }
-            goto READ_BUF;
-        } else if (macro->type == MACRO_FUNCTION) {
-            TODO();
+        goto READ_BUF;
+    } else if (scc_probe_stream_next(stream->input) == '/') {
+        ch = scc_probe_stream_peek(stream->input);
+        if (ch == '/') {
+            scc_probe_stream_reset(stream->input);
+            scc_lex_parse_skip_line(stream->input, &stream->pos);
+            scc_probe_stream_sync(stream->input);
+        } else if (ch == '*') {
+            scc_probe_stream_reset(stream->input);
+            scc_lex_parse_skip_block_comment(stream->input, &stream->pos);
+            scc_probe_stream_sync(stream->input);
         }
-        UNREACHABLE();
     }
-    return core_stream_next_char(stream->input);
+
+    // 非标识符字符，直接消费并返回
+    return scc_probe_stream_consume(stream->input);
 }
 
-static core_stream_t *pp_stream_init(smcc_pp_t *self, core_stream_t *input) {
-    pp_stream_t *stream = smcc_malloc(sizeof(pp_stream_t));
+static void pp_stream_drop(scc_probe_stream_t *_stream) {
+    scc_pp_stream_t *stream = (scc_pp_stream_t *)_stream;
+    Assert(stream != null);
+    scc_cstring_free(&stream->stream.name);
+    if (stream->tmp_stream) {
+        scc_probe_stream_drop(stream->tmp_stream);
+    }
+    scc_free(_stream);
+}
+
+static scc_probe_stream_t *pp_stream_init(scc_pproc_t *self,
+                                          scc_probe_stream_t *input) {
+    scc_pp_stream_t *stream = scc_malloc(sizeof(scc_pp_stream_t));
     if (stream == null) {
         LOG_FATAL("Failed to allocate memory for output stream");
-    }
-    if (stream == null || self == null) {
         return null;
     }
+
+    if (self == null) {
+        scc_free(stream);
+        return null;
+    }
+
     stream->self = self;
     stream->input = input;
-    stream->size = 0;
-    stream->pos = 0;
 
-    stream->stream.name = cstring_from_cstr("pipe_stream");
-    stream->stream.free_stream = null;
-    stream->stream.next_char = next_char;
-    stream->stream.peek_char = null;
-    stream->stream.reset_char = null;
+    stream->tmp_stream = null;
+    stream->pos = scc_pos_create();
+
+    stream->stream.name = scc_cstring_from_cstr("pp_stream");
+    stream->stream.consume = pp_stream_read_char;
+    stream->stream.peek = null;
+    stream->stream.next = null;
+    stream->stream.sync = null;
+    stream->stream.reset = null;
+    stream->stream.back = null;
     stream->stream.read_buf = null;
-    return (core_stream_t *)stream;
+    stream->stream.is_at_end = null;
+    stream->stream.drop = pp_stream_drop;
+
+    return (scc_probe_stream_t *)stream;
 }
 
-core_stream_t *pp_init(smcc_pp_t *pp, core_stream_t *input) {
+scc_probe_stream_t *scc_pproc_init(scc_pproc_t *pp, scc_probe_stream_t *input) {
     if (pp == null || input == null) {
         return null;
     }
-    core_mem_stream_t *stream = smcc_malloc(sizeof(core_mem_stream_t));
 
-    if (stream == null) {
-        LOG_FATAL("Failed to allocate memory for output stream");
-    }
     pp->stream = pp_stream_init(pp, input);
-    Assert(pp->stream != null);
+    if (pp->stream == null) {
+        return null;
+    }
+
+    scc_pp_marco_table_init(&pp->macro_table);
 
-    hashmap_init(&pp->macros);
-    pp->macros.hash_func = (u32 (*)(const void *))hash_func;
-    pp->macros.key_cmp = (int (*)(const void *, const void *))hash_cmp;
     return pp->stream;
 }
 
 // 销毁预处理器
-void pp_drop(smcc_pp_t *pp) {
-    if (pp == NULL)
+void scc_pproc_drop(scc_pproc_t *pp) {
+    if (pp == null)
         return;
 
-    // 清理所有宏定义
-    // 注意：需要实现 hashmap 的迭代和清理函数
-    hashmap_drop(&pp->macros);
+    scc_pp_macro_table_drop(&pp->macro_table);
 
-    // 清理字符串池
-    // strpool_destroy(&pp->strpool);
-
-    // 清理条件编译栈
-    // 需要释放栈中每个元素的资源（如果有的话）
-    // vec_free(pp->if_stack);
-
-    // 清理文件名
-    cstring_free(&pp->stream->name);
+    // 清理流
+    if (pp->stream) {
+        scc_probe_stream_drop(pp->stream);
+        pp->stream = null;
+    }
 }

libs/pprocessor/tests/test_pp.c

@@ -4,7 +4,7 @@
 #include <stdlib.h>
 #include <utest/acutest.h>
 
-static core_stream_t *from_file_stream(FILE *fp) {
+static scc_probe_stream_t *from_file_stream(FILE *fp) {
     if (fseek(fp, 0, SEEK_END) != 0) {
         perror("fseek failed");
         return NULL;
@@ -20,9 +20,9 @@ static core_stream_t *from_file_stream(FILE *fp) {
     usize read_ret = fread(buffer, 1, fsize, fp);
     fclose(fp);
 
-    core_mem_stream_t *mem_stream = malloc(sizeof(core_mem_stream_t));
-    core_stream_t *stream =
-        core_mem_stream_init(mem_stream, buffer, fsize, true);
+    scc_mem_probe_stream_t *mem_stream = malloc(sizeof(scc_mem_probe_stream_t));
+    scc_probe_stream_t *stream =
+        scc_mem_probe_stream_init(mem_stream, buffer, read_ret, true);
     return stream;
 }
 
@@ -37,28 +37,52 @@ static void test_file(const char *name) {
     FILE *fexpect = fopen(expected_fname, "r");
     assert(fexpect != NULL);
 
-    smcc_pp_t pp;
-    core_mem_stream_t stream;
-    core_stream_t *output_stream = pp_init(&pp, from_file_stream(fsrc));
-    core_stream_t *expect_stream = from_file_stream(fexpect);
-    while (1) {
-        int output_ch = core_stream_next_char(output_stream);
-        int expect_ch = core_stream_next_char(expect_stream);
-        TEST_CHECK(output_ch == expect_ch);
-        TEST_MSG("output: %c, expect: %c", output_ch, expect_ch);
-        if (output_ch == core_stream_eof) {
+    scc_pproc_t pp;
+    scc_probe_stream_t *output_stream =
+        scc_pproc_init(&pp, from_file_stream(fsrc));
+    scc_probe_stream_t *expect_stream = from_file_stream(fexpect);
+
+    fclose(fsrc);
+    fclose(fexpect);
+
+    TEST_CASE(src_fname);
+#define BUFFER_LEN 4096
+    int ch;
+    char expect_buffer[BUFFER_LEN];
+    char output_buffer[BUFFER_LEN];
+    usize size_produced = 0, size_expected = 0;
+    for (usize i = 0; i < BUFFER_LEN; ++i) {
+        ch = scc_probe_stream_consume(expect_stream);
+        if (ch != scc_stream_eof) {
+            expect_buffer[i] = (char)ch;
+        } else {
+            size_expected = i;
             break;
         }
     }
-    pp_drop(&pp);
+    for (usize i = 0; i < BUFFER_LEN; ++i) {
+        ch = scc_probe_stream_consume(output_stream);
+        if (ch != scc_stream_eof) {
+            output_buffer[i] = (char)ch;
+        } else {
+            size_produced = i;
+            break;
+        }
+    }
+    TEST_CHECK(size_produced == size_expected &&
+               memcmp(output_buffer, expect_buffer, size_produced) == 0);
+    TEST_DUMP("Expected:", expect_buffer, size_expected);
+    TEST_DUMP("Produced:", output_buffer, size_produced);
+
+    scc_pproc_drop(&pp);
 }
 
 static void test_basic(void) {
     char name[32];
-    // for (int i = 1; i <= 22; ++i) {
-    //     snprintf(name, sizeof(name), "%02d", i);
-    //     test_file(name);
-    // }
+    for (int i = 1; i <= 22; ++i) {
+        snprintf(name, sizeof(name), "%02d", i);
+        test_file(name);
+    }
 }
 
 TEST_LIST = {

libs/pprocessor/tests/test_run.c

@@ -2,19 +2,19 @@
 #include <stdio.h>
 
 int main(void) {
-    smcc_pp_t pp;
-    core_mem_stream_t input;
-    core_stream_t *output;
+    scc_pproc_t pp;
+    scc_mem_probe_stream_t input;
+    scc_probe_stream_t *output;
 
     const char buf[] = "#define A 123 \"asd\"\nA A A\n";
-    output =
-        pp_init(&pp, core_mem_stream_init(&input, buf, sizeof(buf) - 1, false));
+    output = scc_pproc_init(
+        &pp, scc_mem_probe_stream_init(&input, buf, sizeof(buf) - 1, false));
 
     int ch = 0;
 
     while (1) {
-        ch = core_stream_next_char(output);
-        if (ch == core_stream_eof) {
+        ch = scc_probe_stream_consume(output);
+        if (ch == scc_stream_eof) {
             break;
         }
         putc(ch, stdout);

libs/pprocessor/tests/test_unit.c

@@ -3,45 +3,50 @@
 #include <string.h>
 #include <utest/acutest.h>
 
-static cbool process_input(const char *input, cstring_t *output) {
-    smcc_pp_t pp;
-    core_mem_stream_t mem_stream;
-    core_stream_t *output_stream;
+static cbool process_input(const char *input, scc_cstring_t *output) {
+    scc_pproc_t pp;
+    scc_mem_probe_stream_t mem_stream;
+    scc_probe_stream_t *output_stream;
 
     // 初始化预处理器
-    output_stream = pp_init(
-        &pp, core_mem_stream_init(&mem_stream, input, strlen(input), false));
+    output_stream =
+        scc_pproc_init(&pp, scc_mem_probe_stream_init(&mem_stream, input,
+                                                      strlen(input), false));
 
     // 获取输出结果
     int ch;
-    *output = cstring_new();
+    *output = scc_cstring_create();
     while (1) {
-        ch = core_stream_next_char(output_stream);
-        if (ch == core_stream_eof) {
+        ch = scc_probe_stream_consume(output_stream);
+        if (ch == scc_stream_eof) {
             break;
         }
-        cstring_push(output, (char)ch);
+        scc_cstring_append_ch(output, (char)ch);
     }
     // 清理资源
-    pp_drop(&pp);
+    scc_pproc_drop(&pp);
 
     return true;
 }
 
 #define CHECK_PP_OUTPUT_EXACT(input, expect)                                   \
     do {                                                                       \
-        cstring_t output;                                                      \
+        scc_cstring_t output;                                                  \
         process_input(input, &output);                                         \
         assert(output.data != NULL);                                           \
         TEST_CHECK(strcmp(output.data, expect) == 0);                          \
+        TEST_MSG("Expected: %s", expect);                                      \
+        TEST_MSG("Produced: %s", output.data);                                 \
     } while (0)
 
 #define CHECK_PP_OUTPUT_CONTAIN(input, expect)                                 \
     do {                                                                       \
-        cstring_t output;                                                      \
+        scc_cstring_t output;                                                  \
         process_input(input, &output);                                         \
         assert(output.data != NULL);                                           \
         TEST_CHECK(strstr(output.data, expect) != NULL);                       \
+        TEST_MSG("Expected: %s", expect);                                      \
+        TEST_MSG("Produced: %s", output.data);                                 \
     } while (0)
 
 static void test_define_simple_object_macro(void) {
@@ -56,6 +61,15 @@ static void test_define_complex_object_macro(void) {
     CHECK_PP_OUTPUT_EXACT("#define PI 3.14159\nPI\n", "3.14159\n");
 }
 
+static void test_define_object_macro_backspace(void) {
+    TEST_CASE("object-like macro check backspace");
+    CHECK_PP_OUTPUT_EXACT("#define MAX  100\nMAX\n", "100\n");
+    CHECK_PP_OUTPUT_EXACT("#define NAME \ttest\r\nNAME\n", "test\n");
+    CHECK_PP_OUTPUT_EXACT("#define  \tVALUE  (100 \t+  50)\nVALUE\n",
+                          "(100 + 50)\n");
+    CHECK_PP_OUTPUT_EXACT("#define \tPI \t 3.14159\nPI\n", "3.14159\n");
+}
+
 static void test_define_function_macro(void) {
     TEST_CASE("function-like macro");
     CHECK_PP_OUTPUT_EXACT("#define ADD(a,b) a + b\nADD(1, 2)\n", "1 + 2\n");
@@ -76,7 +90,7 @@ static void test_define_concat_operator(void) {
     TEST_CASE("concatenation operator (##)");
     CHECK_PP_OUTPUT_EXACT("#define CONCAT(a,b) a##b\nCONCAT(hello,world)\n",
                           "helloworld\n");
-    CHECK_PP_OUTPUT_EXACT("#define JOIN(pre,suf) pre##suf\nJOIN(var,123)\n",
+    CHECK_PP_OUTPUT_EXACT("#define JOIN(pre,suf) pre ## suf\nJOIN(var, 123)\n",
                           "var123\n");
 }
 
@@ -90,12 +104,112 @@ static void test_define_nested_macros(void) {
         "((1 + 1) + 1)\n");
 }
 
+static void test_undef_macros(void) {
+    TEST_CASE("test_undef_macros");
+    CHECK_PP_OUTPUT_EXACT("#define x 1\n"
+                          "x\n"
+                          "#undef x\n"
+                          "x\n"
+                          "#define x 2\n"
+                          "x\n",
+                          "1\nx\n2\n");
+}
+
+static void hard_test_define_func_macros(void) {
+    TEST_CASE("func_macros_hard with pp_01");
+    CHECK_PP_OUTPUT_EXACT("#define hash_hash # ## #\n"
+                          "#define mkstr(a) # a\n"
+                          "#define in_between(a) mkstr(a)\n"
+                          "#define join(c, d) in_between(c hash_hash d)\n"
+                          "char p[] = join(x, y);\n",
+                          "char p[] = \"x ## y\";\n");
+
+    TEST_CASE("func_macros_hard with recursive define");
+    CHECK_PP_OUTPUT_EXACT("#define M1(x) M2(x + 1)\n"
+                          "#define M2(x) M1(x * 2)\n"
+                          "M1(5)\n",
+                          "M1(5 + 1 * 2)\n");
+    CHECK_PP_OUTPUT_EXACT("#define A B\n"
+                          "#define B C\n"
+                          "#define C 1\n"
+                          "A\n",
+                          "1\n");
+
+    TEST_CASE("func_macros_hard with self recursive call");
+    CHECK_PP_OUTPUT_EXACT("#define M(x) x\n"
+                          "M(M(10))\n",
+                          "10\n");
+    CHECK_PP_OUTPUT_EXACT("#define M(x) M(x)\n"
+                          "#define N(x) x\n"
+                          "N(M(1))\n",
+                          "M(1)\n");
+
+    TEST_CASE("func_macros_hard with define by macro");
+    CHECK_PP_OUTPUT_EXACT("#define M1(x) M1(x + 1)\n"
+                          "#define M2 M1\n"
+                          "#define M3(x) x\n"
+                          "M3(M3(M2)(0))\n",
+                          "M1(0 + 1)\n");
+
+    TEST_CASE("TODO");
+    CHECK_PP_OUTPUT_EXACT("#define str(x) # x\n"
+                          "str()\n",
+                          "\"\"\n");
+
+    TEST_CASE("TODO");
+    CHECK_PP_OUTPUT_EXACT("#define x 1\n"
+                          "#define f(a) f(x * (a))\n"
+                          "f(0)\n"
+                          "f(x)",
+                          "f(1 * (0))\n"
+                          "f(1 * (1))");
+    CHECK_PP_OUTPUT_EXACT("#define x x(0)\n"
+                          "#define f(a) f(x * (a))\n"
+                          "f(f(0))\n"
+                          "f(f(x))\n"
+                          "f(f(a))\n",
+                          "f(x(0) * (f(x(0) * (0))))\n"
+                          "f(x(0) * (f(x(0) * (x(0)))))\n"
+                          "f(x(0) * (f(x(0) * (a))))\n");
+}
+
+static void test_conditional_compilation(void) {
+    TEST_CASE("conditional compilation");
+    CHECK_PP_OUTPUT_EXACT("#if 1\ntrue\n#endif\n", "true\n");
+    CHECK_PP_OUTPUT_EXACT("#if 0\nfalse\n#endif\n", "");
+    CHECK_PP_OUTPUT_EXACT("#define FLAG 1\n#if FLAG\ntrue\n#endif\n", "true\n");
+}
+
+static void test_error_cases(void) {
+    TEST_CASE("macro redefinition");
+    // 应检测到警告或错误
+    // CHECK_PP_OUTPUT_CONTAIN("#define A 1\n#define A 2\n", "warning");
+
+    TEST_CASE("undefined macro");
+    CHECK_PP_OUTPUT_EXACT("UNDEFINED_MACRO\n", "UNDEFINED_MACRO\n");
+}
+
+static void test_edge_cases(void) {
+    TEST_CASE("empty macro");
+    CHECK_PP_OUTPUT_EXACT("#define EMPTY\nEMPTY\n", "\n");
+
+    TEST_CASE("macro with only spaces");
+    CHECK_PP_OUTPUT_EXACT("#define SPACE   \nSPACE\n", "\n");
+
+    TEST_CASE("deep nesting");
+    CHECK_PP_OUTPUT_EXACT("#define A B\n#define B C\n#define C 1\nA\n", "1\n");
+}
+
+#define TEST_LIST_CASE(func_name) {#func_name, func_name}
 TEST_LIST = {
-    {"test_define_simple_object_macro", test_define_simple_object_macro},
-    {"test_define_complex_object_macro", test_define_complex_object_macro},
-    {"test_define_function_macro", test_define_function_macro},
-    {"test_define_stringify_operator", test_define_stringify_operator},
-    {"test_define_concat_operator", test_define_concat_operator},
-    {"test_define_nested_macros", test_define_nested_macros},
+    TEST_LIST_CASE(test_define_simple_object_macro),
+    TEST_LIST_CASE(test_define_complex_object_macro),
+    TEST_LIST_CASE(test_define_object_macro_backspace),
+    TEST_LIST_CASE(test_define_function_macro),
+    TEST_LIST_CASE(test_define_stringify_operator),
+    TEST_LIST_CASE(test_define_concat_operator),
+    TEST_LIST_CASE(test_define_nested_macros),
+    TEST_LIST_CASE(test_undef_macros),
+    TEST_LIST_CASE(hard_test_define_func_macros),
     {NULL, NULL},
 };

runtime/libcore/include/libcore.h

@@ -1,14 +0,0 @@
-#ifndef __SCC_CORE_H__
-#define __SCC_CORE_H__
-
-#include <core_log.h>
-
-#include <core_impl.h>
-#include <core_macro.h>
-#include <core_mem.h>
-#include <core_pos.h>
-#include <core_str.h>
-#include <core_stream.h>
-#include <core_vec.h>
-
-#endif // __SCC_CORE_H__

runtime/libutils/cbuild.toml

@@ -1,7 +0,0 @@
-[package]
-name = "libutils"
-version = "0.1.0"
-
-dependencies = [
-    { name = "core", path = "../libcore" }
-]

runtime/libutils/include/libutils.h

@@ -1,9 +0,0 @@
-#ifndef __SMCC_UTILS_H__
-#define __SMCC_UTILS_H__
-
-#include "hashmap.h"
-#include "kllist.h"
-#include "strpool.h"
-#include <libcore.h>
-
-#endif // __SMCC_UTILS_H__

runtime/libutils/src/strpool.c

@@ -1,27 +0,0 @@
-#include "strpool.h"
-
-void init_strpool(strpool_t *pool) {
-    pool->ht.hash_func = (u32 (*)(const void *))scc_strhash32;
-    pool->ht.key_cmp = (int (*)(const void *, const void *))scc_strcmp;
-    hashmap_init(&pool->ht);
-}
-
-const char *strpool_intern(strpool_t *pool, const char *str) {
-    void *existing = hashmap_get(&pool->ht, str);
-    if (existing) {
-        return existing;
-    }
-
-    usize len = scc_strlen(str) + 1;
-    char *new_str = scc_malloc(len);
-    if (!new_str) {
-        LOG_ERROR("strpool: Failed to allocate memory for string");
-        return NULL;
-    }
-    scc_memcpy(new_str, str, len);
-
-    hashmap_set(&pool->ht, new_str, new_str);
-    return new_str;
-}
-
-void strpool_destroy(strpool_t *pool) { hashmap_drop(&pool->ht); }

runtime/libutils/src/vector-gdb.py

@@ -1,112 +0,0 @@
-# vector_gdb.py
-import gdb # type: ignore
-from gdb.printing import PrettyPrinter # type: ignore
-
-class VectorPrinter:
-    """兼容新旧注册方式的最终方案"""
-    
-    def __init__(self, val: gdb.Value):
-        self.val:gdb.Value = val
-        
-    def check_type(self) -> bool:
-        """类型检查（兼容匿名结构体）"""
-        try:
-            if self.val.type.code != gdb.TYPE_CODE_STRUCT:
-                return False
-            fields = self.val.type.fields()
-            if not fields:
-                return False
-            exp = ['size', 'cap', 'data']
-            for t in fields:
-                if t.name in exp:
-                    exp.remove(t.name)
-                else:
-                    return False
-            return True
-        except gdb.error:
-            return False
-    
-    def to_string(self):
-        if not self.check_type():
-            return "Not a vector"
-            
-        return "vector({} size={}, cap={})".format(
-            self.val.address,
-            self.val['size'],
-            self.val['cap'],
-        )
-    
-    def display_hint(self):
-        return 'array'
-    
-    def children(self):
-        """生成数组元素（关键改进点）"""
-        if not self.check_type():
-            return []
-            
-        size = int(self.val['size'])
-        cap = int(self.val['cap'])
-        data_ptr = self.val['data']
-        
-        if cap == 0 or data_ptr == 0:
-            return []
-        
-        # 使用 GDB 内置数组转换
-        array = data_ptr.dereference()
-        array = array.cast(data_ptr.type.target().array(cap - 1))
-        
-        for i in range(size):
-            # state = "<used>" if i < size else "<unused>"
-            try:
-                value = array[i]
-                yield (f"[{i}] {value.type} {value.address}", value)
-            except gdb.MemoryError:
-                yield (f"[{i}]", "<invalid>")
-
-# 注册方式一：传统append方法（您之前有效的方式）self
-def append_printer():
-    gdb.pretty_printers.append(
-        lambda val: VectorPrinter(val) if VectorPrinter(val).check_type() else None
-    )
-
-# 注册方式二：新版注册方法（备用方案）  
-def register_new_printer():
-    class VectorPrinterLocator(PrettyPrinter):
-        def __init__(self):
-            super().__init__("vector_printer")
-            
-        def __call__(self, val):
-            ret = VectorPrinter(val).check_type()
-            print(f"ret {ret}, type {val.type}, {[(i.name, i.type) for i in  val.type.fields()]}")
-            return None
-    
-    gdb.printing.register_pretty_printer(
-        gdb.current_objfile(),
-        VectorPrinterLocator()
-    )
-
-# 双重注册保证兼容性
-append_printer()  # 保留您原来有效的方式
-# register_new_printer()  # 添加新版注册
-
-class VectorInfoCommand(gdb.Command):
-    """保持原有命令不变"""
-    def __init__(self):
-        super().__init__("vector_info", gdb.COMMAND_USER)
-        
-    def invoke(self, argument, from_tty):
-        val = gdb.parse_and_eval(argument)
-        printer = VectorPrinter(val)
-        
-        if not printer.check_type():
-            print("Invalid vector")
-            return
-            
-        print("=== Vector Details ===")
-        print("Size:", val['size'])
-        print("Capacity:", val['cap'])
-        print("Elements:")
-        for name, value in printer.children():
-            print(f"  {name}: {value}")
-
-VectorInfoCommand()

runtime/log/include/log.c

@@ -79,12 +79,6 @@ void init_logger(logger_t *logger, const char *name) {
     log_set_level(logger, LOG_LEVEL_ALL);
 }
 
-logger_t *log_get(const char *name) {
-    // TODO for -Wunused-parameter
-    (void)name;
-    return &__default_logger_root;
-}
-
 void log_set_level(logger_t *logger, int level) {
     if (logger)
         logger->level = level;
@@ -98,9 +92,3 @@ void log_set_handler(logger_t *logger, log_handler handler) {
     else
         __default_logger_root.handler = handler;
 }
-
-void logger_destroy(logger_t *logger) {
-    // TODO for -Wunused-parameter
-    (void)logger;
-    return;
-}

runtime/log/include/log.h

@@ -96,16 +96,6 @@ extern logger_t __default_logger_root;
  */
 void init_logger(logger_t *logger, const char *name);
 
-// TODO log_set(); 暂未实现 日志注册
-
-/**
- * @brief 获取或创建日志器实例
- * @param[in] name 日志器名称（NULL表示获取默认日志器）
- * @return 日志器实例指针
- * @warning 若没有找到相应日志器则会返回根日志器
- */
-logger_t *log_get(const char *name);
-
 /**
  * @brief 设置日志级别
  * @param[in] logger 目标日志器实例
@@ -120,11 +110,6 @@ void log_set_level(logger_t *logger, int level);
  */
 void log_set_handler(logger_t *logger, log_handler handler);
 
-/**
- * @todo TODO impliment
- */
-void logger_destroy(logger_t *logger);
-
 #ifndef LOG_MAX_MAROC_BUF_SIZE
 #define LOG_MAX_MAROC_BUF_SIZE LOGGER_MAX_BUF_SIZE ///< 宏展开缓冲区尺寸
 #endif

runtime/runtime_gdb.py

@@ -0,0 +1,222 @@
+"https://sourceware.org/gdb/current/onlinedocs/gdb.html/Python-API.html#Python-API"
+
+import gdb  # type: ignore
+
+
+class VectorPrinter(gdb.ValuePrinter):
+    """兼容新旧注册方式的最终方案"""
+
+    def __init__(self, val: gdb.Value):
+        self.val: gdb.Value = val
+
+    @staticmethod
+    def check_type(val: gdb.Value) -> bool:
+        """类型检查（兼容匿名结构体）"""
+        try:
+            if val.type.code not in (gdb.TYPE_CODE_STRUCT, gdb.TYPE_CODE_TYPEDEF):
+                return False
+            if val.type.name == "scc_cstring_t":
+                return False
+            fields = val.type.fields()
+            if not fields:
+                return False
+            exp = ["size", "cap", "data"]
+            for t in fields:
+                if t.name in exp:
+                    exp.remove(t.name)
+                else:
+                    return False
+            return True
+        except gdb.error:
+            return False
+        except ValueError:
+            return False
+        except TypeError:
+            return False
+        except Exception as e:
+            print(f"[DEBUG] Unknown exception type: {type(e).__name__}")
+            print(f"[DEBUG] Exception details: {e}")
+            print(
+                f"[DEBUG] val type: {val.type if hasattr(val, 'type') else 'no type attr'}"
+            )
+            return False
+
+    def to_string(self):
+        """
+        GDB will call this method to display the string representation
+        of the value passed to the object's constructor.
+
+        This is a basic method, and is optional.
+
+        When printing from the CLI, if the to_string method exists,
+        then GDB will prepend its result to the values returned by children.
+        Exactly how this formatting is done is dependent on the display hint,
+        and may change as more hints are added. Also, depending on the print settings
+        (see Print Settings), the CLI may print just the result of to_string
+        in a stack trace, omitting the result of children.
+
+        If this method returns a string, it is printed verbatim.
+
+        Otherwise, if this method returns an instance of gdb.Value,
+        then GDB prints this value. This may result in a call to another pretty-printer.
+
+        If instead the method returns a Python value which is convertible to a gdb.Value,
+        then GDB performs the conversion and prints the resulting value. Again,
+        this may result in a call to another pretty-printer. Python scalars
+        (integers, floats, and booleans) and strings are convertible to gdb.Value;
+        other types are not.
+
+        Finally, if this method returns None then no further operations are performed
+        in this method and nothing is printed.
+
+        If the result is not one of these types, an exception is raised.
+
+        https://sourceware.org/gdb/current/onlinedocs/gdb.html/Pretty-Printing-API.html#Pretty-Printing-API
+        """
+        return (
+            f"vector({self.val.address} size={self.val['size']}, cap={self.val['cap']})"
+        )
+
+    def display_hint(self):
+        """
+        The CLI may call this method and use its result to change the formatting of a value.
+        The result will also be supplied to an MI consumer as a 'displayhint'
+        attribute of the variable being printed.
+
+        This is a basic method, and is optional. If it does exist,
+        this method must return a string or the special value None.
+
+        Some display hints are predefined by GDB:
+
+        'array'
+        Indicate that the object being printed is “array-like”.
+        The CLI uses this to respect parameters such as set print elements and set print array.
+
+        'map'
+        Indicate that the object being printed is “map-like”,
+        and that the children of this value can be assumed to alternate between keys and values.
+
+        'string'
+        Indicate that the object being printed is “string-like”.
+        If the printer's to_string method returns a Python string of some kind,
+        then GDB will call its internal language-specific string-printing function
+        to format the string. For the CLI this means adding quotation marks, possibly
+        escaping some characters, respecting set print elements, and the like.
+
+        The special value None causes GDB to apply the default display rules.
+
+        https://sourceware.org/gdb/current/onlinedocs/gdb.html/Pretty-Printing-API.html#Pretty-Printing-API
+        """
+        return "array"
+
+    def num_children(self):
+        """
+        This is not a basic method, so GDB will only ever call it for objects
+        derived from gdb.ValuePrinter.
+
+        If available, this method should return the number of children.
+
+        None may be returned if the number can't readily be computed.
+
+        https://sourceware.org/gdb/current/onlinedocs/gdb.html/Pretty-Printing-API.html#Pretty-Printing-API
+        """
+        return int(self.val["size"])
+
+    def children(self):
+        """
+        This is not a basic method, so GDB will only ever call it for objects
+        derived from gdb.ValuePrinter.
+
+        If available, this method should return the child item (that is,
+        a tuple holding the name and value of this child) indicated by n.
+        Indices start at zero.
+
+        GDB provides a function which can be used to look up the default
+        pretty-printer for a gdb.Value:
+
+        https://sourceware.org/gdb/current/onlinedocs/gdb.html/Pretty-Printing-API.html#Pretty-Printing-API
+        """
+        size = int(self.val["size"])
+        cap = int(self.val["cap"])
+        data_ptr = self.val["data"]
+
+        if cap == 0 or data_ptr == 0:
+            return []
+
+        # 使用 GDB 内置数组转换
+        array = data_ptr.dereference()
+        array = array.cast(data_ptr.type.target().array(cap - 1))
+
+        for i in range(size):
+            # state = "<used>" if i < size else "<unused>"
+            try:
+                value = array[i]
+                yield (f"[{i}] {value.type} {value.address}", value)
+            except gdb.MemoryError:
+                yield (f"[{i}]", "<invalid>")
+
+
+class HashTablePrinter(gdb.ValuePrinter):
+    def __init__(self, val: gdb.Value):
+        self.val: gdb.Value = val
+
+    @staticmethod
+    def check_type(val: gdb.Value) -> bool:
+        if val.type.name in ["scc_hashtable_t", "scc_hashtable"]:
+            return True
+        return False
+
+
+def append_printer():
+    "注册方式一：传统append方法（您之前有效的方式）self"
+    gdb.pretty_printers.append(
+        lambda val: VectorPrinter(val) if VectorPrinter.check_type(val) else None
+    )
+
+
+def register_new_printer():
+    "注册方式二：新版注册方法（备用方案）"
+
+    def str_lookup_function(val):
+        if VectorPrinter.check_type(val) is False:
+            return None
+        ret = VectorPrinter(val)
+        # print(
+        #     f"ret {ret}, type {val.type.name}, {[(i.name, i.type) for i in val.type.fields()]}"
+        # )
+        return ret
+
+    gdb.printing.register_pretty_printer(gdb.current_objfile(), str_lookup_function)
+    # if gdb.current_progspace() is not None:
+    #     pts = gdb.current_progspace().pretty_printers
+    #     print(pts, len(pts))
+    #     pts.append(str_lookup_function)
+
+
+class VectorInfoCommand(gdb.Command):
+    """保持原有命令不变"""
+
+    def __init__(self):
+        super().__init__("vector_info", gdb.COMMAND_USER)
+
+    def invoke(self, argument, from_tty):
+        val = gdb.parse_and_eval(argument)
+
+        if not VectorPrinter.check_type(val):
+            print("Invalid vector")
+            return
+
+        printer = VectorPrinter(val)
+        print("=== Vector Details ===")
+        print("Size:", val["size"])
+        print("Capacity:", val["cap"])
+        print("Elements:")
+        for name, value in printer.children():
+            print(f"  {name}: {value}")
+
+
+if __name__ == "__main__":
+    # 双重注册保证兼容性
+    # append_printer()  # 保留您原来有效的方式
+    register_new_printer()  # 添加新版注册
+    VectorInfoCommand()

runtime/scc_core/cbuild.toml

@@ -1,5 +1,5 @@
 [package]
-name = "libcore"
+name = "scc_core"
 version = "0.1.0"
 
 default_features = ["std_impl"]

runtime/scc_core/include/scc_core.h

@@ -0,0 +1,14 @@
+#ifndef __SCC_CORE_H__
+#define __SCC_CORE_H__
+
+#include <scc_core_log.h>
+
+#include <scc_core_impl.h>
+#include <scc_core_macro.h>
+#include <scc_core_mem.h>
+#include <scc_core_pos.h>
+#include <scc_core_str.h>
+#include <scc_core_stream.h>
+#include <scc_core_vec.h>
+
+#endif // __SCC_CORE_H__

runtime/scc_core/include/scc_core_impl.h

@@ -1,7 +1,7 @@
 #ifndef __SCC_CORE_IMPL_H__
 #define __SCC_CORE_IMPL_H__
 
-#include "core_type.h"
+#include "scc_core_type.h"
 
 /* ====== 内存管理核心接口 ====== */
 

runtime/scc_core/include/scc_core_mem.h

@@ -1,7 +1,7 @@
 #ifndef __SCC_CORE_MEM_H__
 #define __SCC_CORE_MEM_H__
 
-#include "core_type.h"
+#include "scc_core_type.h"
 
 void *scc_memcpy(void *dest, const void *src, usize n);
 void *scc_memmove(void *dest, const void *src, usize n);

runtime/scc_core/include/scc_core_pos.h

@@ -1,8 +1,8 @@
 #ifndef __SCC_CORE_POS_H__
 #define __SCC_CORE_POS_H__
 
-#include "core_str.h"
-#include "core_type.h"
+#include "scc_core_str.h"
+#include "scc_core_type.h"
 typedef struct scc_pos {
     scc_cstring_t name;
     usize line;
@@ -10,16 +10,16 @@ typedef struct scc_pos {
     usize offset;
 } scc_pos_t;
 
-static inline scc_pos_t scc_pos_init() {
-    return (scc_pos_t){scc_cstring_new(), 1, 1, 0};
+static inline scc_pos_t scc_pos_create() {
+    return (scc_pos_t){scc_cstring_create(), 1, 1, 0};
 }
 
-static inline void core_pos_next(scc_pos_t *pos) {
+static inline void scc_pos_next(scc_pos_t *pos) {
     pos->offset++;
     pos->col++;
 }
 
-static inline void core_pos_next_line(scc_pos_t *pos) {
+static inline void scc_pos_next_line(scc_pos_t *pos) {
     pos->offset++;
     pos->line++;
     pos->col = 1;

runtime/scc_core/include/scc_core_str.h

@@ -1,9 +1,10 @@
 #ifndef __SCC_CORE_STR_H__
 #define __SCC_CORE_STR_H__
 
-#include "core_impl.h"
-#include "core_log.h"
-#include "core_type.h"
+#include "scc_core_impl.h"
+#include "scc_core_log.h"
+#include "scc_core_mem.h"
+#include "scc_core_type.h"
 
 /**
  * @brief 动态字符串结构体
@@ -20,10 +21,17 @@ typedef struct scc_cstring {
  *
  * @return cstring_t 初始化后的对象
  */
-static inline scc_cstring_t scc_cstring_new(void) {
+static inline scc_cstring_t scc_cstring_create(void) {
     return (scc_cstring_t){.data = null, .size = 0, .cap = 0};
 }
 
+static inline void scc_cstring_init(scc_cstring_t *string) {
+    Assert(string != null);
+    string->data = null;
+    string->size = 0;
+    string->cap = 0;
+}
+
 /**
  * @brief 从 C 风格字符串创建一个动态字符串副本
  *
@@ -32,7 +40,7 @@ static inline scc_cstring_t scc_cstring_new(void) {
  */
 static inline scc_cstring_t scc_cstring_from_cstr(const char *s) {
     if (s == null) {
-        return scc_cstring_new();
+        return scc_cstring_create();
     }
 
     usize len = 0;
@@ -48,6 +56,10 @@ static inline scc_cstring_t scc_cstring_from_cstr(const char *s) {
     return (scc_cstring_t){.size = len + 1, .cap = len + 1, .data = data};
 }
 
+static inline scc_cstring_t scc_cstring_copy(const scc_cstring_t *s) {
+    return scc_cstring_from_cstr(s->data);
+}
+
 /**
  * @brief 释放动态字符串占用的内存资源
  *
@@ -57,7 +69,7 @@ static inline void scc_cstring_free(scc_cstring_t *str) {
     if (str == null) {
         return;
     }
-    if (str->cap != 0 && str->data != null) {
+    if (str->data != null) {
         scc_free(str->data);
         str->data = null;
     }
@@ -136,7 +148,13 @@ static inline void scc_cstring_append_ch(scc_cstring_t *str, char ch) {
  * @return usize 字符串实际长度
  */
 static inline usize scc_cstring_len(const scc_cstring_t *str) {
-    return str ? str->size - 1 : 0;
+    if (str == null) {
+        return 0;
+    }
+    if (str->size == 0) {
+        return 0;
+    }
+    return str->size - 1;
 }
 
 /**
@@ -171,9 +189,20 @@ static inline void scc_cstring_clear(scc_cstring_t *str) {
  */
 static inline char *scc_cstring_as_cstr(const scc_cstring_t *str) {
     if (str == null || str->data == null) {
-        return "";
+        return null;
     }
     return str->data;
 }
 
+static inline char *scc_cstring_move_cstr(scc_cstring_t *str) {
+    if (str == null || str->data == null) {
+        return null;
+    }
+    char *ret = str->data;
+    str->data = null;
+    str->cap = 0;
+    str->size = 0;
+    return ret;
+}
+
 #endif /* __SCC_CORE_STR_H__ */

runtime/scc_core/include/scc_core_stream.h

@@ -1,15 +1,15 @@
 #ifndef __SMCC_CORE_PROBE_STREAM_H__
 #define __SMCC_CORE_PROBE_STREAM_H__
 
-#include "core_impl.h"
-#include "core_macro.h"
-#include "core_mem.h"
-#include "core_str.h"
+#include "scc_core_impl.h"
+#include "scc_core_macro.h"
+#include "scc_core_mem.h"
+#include "scc_core_str.h"
 
 struct scc_probe_stream;
 typedef struct scc_probe_stream scc_probe_stream_t;
 
-#define core_stream_eof (-1)
+#define scc_stream_eof (-1)
 
 /**
  * @brief 带探针的流接口
@@ -100,21 +100,31 @@ typedef struct scc_mem_probe_stream {
     usize data_length;
     usize curr_pos;  // 当前读取位置
     usize probe_pos; // 探针位置（用于peek）
-    cbool owned;     // 是否拥有数据（需要释放）
+    cbool owned;     // 是否拥有数据（如果拥有将会自动释放）
 } scc_mem_probe_stream_t;
 
 /**
- * @brief 初始化内存探针流
+ * @brief 初始化内存探针流(由你负责scc_mem_probe_stream_t的释放)
  *
  * @param stream 流结构指针
  * @param data 数据指针
  * @param length 数据长度
- * @param need_copy 是否需要复制数据
+ * @param owned 是否拥有数据（如果拥有将会自动释放）
  * @return core_probe_stream_t* 成功返回流指针，失败返回NULL
  */
 scc_probe_stream_t *scc_mem_probe_stream_init(scc_mem_probe_stream_t *stream,
-                                             const char *data, usize length,
-                                             cbool need_copy);
+                                              const char *data, usize length,
+                                              cbool owned);
+/**
+ * @brief 构造内存探针流(其中drop会自动释放内存)
+ *
+ * @param data
+ * @param length
+ * @param owned 是否拥有数据（如果拥有将会自动释放）
+ * @return scc_probe_stream_t*
+ */
+scc_probe_stream_t *scc_mem_probe_stream_alloc(const char *data, usize length,
+                                               cbool owned);
 #endif
 
 #endif /* __SMCC_CORE_PROBE_STREAM_H__ */

runtime/scc_core/include/scc_core_vec.h

@@ -8,11 +8,34 @@
 #ifndef __SCC_CORE_VEC_H__
 #define __SCC_CORE_VEC_H__
 
-#include "core_impl.h"
-#include "core_type.h"
+#ifndef __SCC_CORE_VEC_USE_STD__
+#include "scc_core_impl.h"
+#include "scc_core_type.h"
 
 #define __scc_vec_realloc scc_realloc
 #define __scc_vec_free scc_free
+#else
+#include <stddef.h>
+#include <stdlib.h>
+
+typedef size_t usize;
+#define __scc_vec_realloc realloc
+#define __scc_vec_free free
+
+#ifndef LOG_FATAL
+#include <stdio.h>
+#define LOG_FATAL(...)                                                         \
+    do {                                                                       \
+        printf(__VA_ARGS__);                                                   \
+        exit(1);                                                               \
+    } while (0)
+#endif
+
+#ifndef Assert
+#include <assert.h>
+#define Assert(cond) assert(cond)
+#endif
+#endif
 
 /** @defgroup vec_struct 数据结构定义 */
 
@@ -51,6 +74,22 @@
         (vec).size = 0, (vec).cap = 0, (vec).data = 0;                         \
     } while (0)
 
+#define scc_vec_realloc(vec, new_cap)                                          \
+    do {                                                                       \
+        void *data =                                                           \
+            __scc_vec_realloc((vec).data, new_cap * sizeof(*(vec).data));      \
+        if (!data) {                                                           \
+            LOG_FATAL("vector_push: realloc failed\n");                        \
+        }                                                                      \
+        (vec).cap = new_cap;                                                   \
+        (vec).data = data;                                                     \
+    } while (0)
+
+#define scc_vec_size(vec) ((vec).size)
+#define scc_vec_cap(vec) ((vec).cap)
+#define scc_vec_foreach(vec, idx)                                              \
+    for (usize idx = 0; idx < scc_vec_size(vec); ++idx)
+
 /**
  * @def scc_vec_push(vec, value)
  * @brief 添加元素到向量末尾
@@ -64,13 +103,7 @@
     do {                                                                       \
         if ((vec).size >= (vec).cap) {                                         \
             int cap = (vec).cap ? (vec).cap * 2 : 4;                           \
-            void *data =                                                       \
-                __scc_vec_realloc((vec).data, cap * sizeof(*(vec).data));      \
-            if (!data) {                                                       \
-                LOG_FATAL("vector_push: realloc failed\n");                    \
-            }                                                                  \
-            (vec).cap = cap;                                                   \
-            (vec).data = data;                                                 \
+            scc_vec_realloc(vec, cap);                                         \
         }                                                                      \
         Assert((vec).data != null);                                            \
         (vec).data[(vec).size++] = value;                                      \

runtime/scc_core/src/cfg.std_impl.c

@@ -2,7 +2,7 @@
 #define _CRT_SECURE_NO_WARNINGS
 #endif
 
-#include <core_impl.h>
+#include <scc_core_impl.h>
 #define __SCC_LOG_IMPORT_SRC__
 #define log_snprintf scc_snprintf
 #define log_printf scc_printf

runtime/scc_core/src/memory.c

@@ -1,4 +1,4 @@
-#include <core_mem.h>
+#include <scc_core_mem.h>
 
 // 判断是否支持非对齐访问（x86/x64 支持）
 #if defined(__i386__) || defined(__x86_64__) || defined(_M_IX86) ||            \

runtime/scc_core/src/stream.c

@@ -1,5 +1,5 @@
-#include <core_log.h>
-#include <core_stream.h>
+#include <scc_core_log.h>
+#include <scc_core_stream.h>
 
 #ifndef __SCC_CORE_NO_MEM_PROBE_STREAM__
 
@@ -8,7 +8,7 @@ static int mem_probe_stream_consume(scc_probe_stream_t *_stream) {
     scc_mem_probe_stream_t *stream = (scc_mem_probe_stream_t *)_stream;
 
     if (stream->curr_pos >= stream->data_length) {
-        return core_stream_eof;
+        return scc_stream_eof;
     }
 
     unsigned char ch = stream->data[stream->curr_pos++];
@@ -24,7 +24,7 @@ static int mem_probe_stream_peek(scc_probe_stream_t *_stream) {
     scc_mem_probe_stream_t *stream = (scc_mem_probe_stream_t *)_stream;
 
     if (stream->probe_pos >= stream->data_length) {
-        return core_stream_eof;
+        return scc_stream_eof;
     }
 
     // 只查看而不移动探针位置
@@ -36,7 +36,7 @@ static int mem_probe_stream_next(scc_probe_stream_t *_stream) {
     scc_mem_probe_stream_t *stream = (scc_mem_probe_stream_t *)_stream;
 
     if (stream->probe_pos >= stream->data_length) {
-        return core_stream_eof;
+        return scc_stream_eof;
     }
 
     // 返回探针位置的字符，并将探针位置向前移动
@@ -111,7 +111,7 @@ static cbool mem_probe_stream_is_at_end(scc_probe_stream_t *_stream) {
     return stream->curr_pos >= stream->data_length;
 }
 
-static void mem_probe_stream_destroy(scc_probe_stream_t *_stream) {
+static void mem_probe_stream_drop(scc_probe_stream_t *_stream) {
     Assert(_stream != null);
     scc_mem_probe_stream_t *stream = (scc_mem_probe_stream_t *)_stream;
 
@@ -125,7 +125,7 @@ static void mem_probe_stream_destroy(scc_probe_stream_t *_stream) {
 
 scc_probe_stream_t *scc_mem_probe_stream_init(scc_mem_probe_stream_t *stream,
                                               const char *data, usize length,
-                                              cbool need_copy) {
+                                              cbool owned) {
     if (stream == null || data == null) {
         LOG_ERROR("param error");
         return null;
@@ -133,22 +133,11 @@ scc_probe_stream_t *scc_mem_probe_stream_init(scc_mem_probe_stream_t *stream,
 
     if (length == 0) {
         LOG_WARN("input memory is empty");
-        need_copy = false;
+        owned = false;
     }
 
-    stream->owned = need_copy;
-    if (need_copy) {
-        char *buf = (char *)scc_malloc(length);
-        if (buf == null) {
-            LOG_ERROR("malloc error");
-            return null;
-        }
-
-        scc_memcpy(buf, data, length);
-        stream->data = buf;
-    } else {
-        stream->data = data;
-    }
+    stream->owned = owned;
+    stream->data = data;
     stream->data_length = length;
     stream->curr_pos = 0;
     stream->probe_pos = 0;
@@ -164,9 +153,30 @@ scc_probe_stream_t *scc_mem_probe_stream_init(scc_mem_probe_stream_t *stream,
     stream->stream.read_buf = mem_probe_stream_read_buf;
     stream->stream.reset = mem_probe_stream_reset;
     stream->stream.is_at_end = mem_probe_stream_is_at_end;
-    stream->stream.drop = mem_probe_stream_destroy;
+    stream->stream.drop = mem_probe_stream_drop;
 
     return (scc_probe_stream_t *)stream;
 }
 
+static void scc_owned_mem_stream_drop(scc_probe_stream_t *_stream) {
+    scc_mem_probe_stream_t *stream = (scc_mem_probe_stream_t *)_stream;
+    mem_probe_stream_drop(_stream);
+    scc_free(stream);
+}
+
+scc_probe_stream_t *scc_mem_probe_stream_alloc(const char *data, usize length,
+                                               cbool owned) {
+    scc_mem_probe_stream_t *stream =
+        (scc_mem_probe_stream_t *)scc_malloc(sizeof(scc_mem_probe_stream_t));
+    if (stream == null) {
+        return null;
+    }
+
+    scc_probe_stream_t *ret =
+        scc_mem_probe_stream_init(stream, data, length, owned);
+    stream->stream.drop = scc_owned_mem_stream_drop;
+    Assert(ret != null);
+    return ret;
+}
+
 #endif /* __SCC_CORE_NO_MEM_PROBE_STREAM__ */

runtime/scc_utils/cbuild.toml

@@ -0,0 +1,5 @@
+[package]
+name = "scc_utils"
+version = "0.1.0"
+
+dependencies = [{ name = "core", path = "../scc_core" }]

runtime/scc_utils/include/scc_hashtable.h

@@ -1,48 +1,48 @@
 /**
- * @file hashmap.h
+ * @file hashtable.h
  * @brief 开放寻址法哈希表实现
  *
  * 提供基于向量容器的哈希表实现，支持动态扩容和墓碑机制
  */
 
-#ifndef __SCC_HASHMAP_H__
-#define __SCC_HASHMAP_H__
+#ifndef __SCC_HASHTABLE_H__
+#define __SCC_HASHTABLE_H__
 
-#include <libcore.h>
+#include <scc_core.h>
 
 /**
  * @enum hp_entry_state_t
  * @brief 哈希表条目状态标识
  */
-typedef enum hashmap_entry_state {
+typedef enum scc_hashtable_entry_state {
     ENTRY_EMPTY,    /**< 空槽位（从未使用过） */
     ENTRY_ACTIVE,   /**< 有效条目（包含键值对） */
     ENTRY_TOMBSTONE /**< 墓碑标记（已删除条目） */
-} hp_entry_state_t;
+} scc_hashtable_entry_state_t;
 
 /**
- * @struct hashmap_entry_t
+ * @struct scc_hashtable_entry_t
  * @brief 哈希表条目结构
  *
  * @note key/value内存由调用者管理，哈希表不负责其生命周期
  */
-typedef struct hashmap_entry {
-    const void *key;        /**< 键指针（不可变） */
-    void *value;            /**< 值指针 */
-    u32 hash;               /**< 预计算的哈希值（避免重复计算） */
-    hp_entry_state_t state; /**< 当前条目状态 */
-} hashmap_entry_t;
+typedef struct scc_hashtable_entry {
+    const void *key;                   /**< 键指针（不可变） */
+    void *value;                       /**< 值指针 */
+    u32 hash;                          /**< 预计算的哈希值（避免重复计算） */
+    scc_hashtable_entry_state_t state; /**< 当前条目状态 */
+} scc_hashtable_entry_t;
 
 /**
- * @struct hashmap_t
+ * @struct scc_hashtable_t
  * @brief 哈希表主体结构
  *
  * 使用开放寻址法实现，采用墓碑标记处理删除操作
  */
-typedef struct smcc_hashmap {
-    SCC_VEC(hashmap_entry_t) entries; /**< 条目存储容器 */
-    u32 count;                        /**< 有效条目数量（不含墓碑） */
-    u32 tombstone_count;              /**< 墓碑条目数量 */
+typedef struct scc_hashtable {
+    SCC_VEC(scc_hashtable_entry_t) entries; /**< 条目存储容器 */
+    u32 count;                              /**< 有效条目数量（不含墓碑） */
+    u32 tombstone_count;                    /**< 墓碑条目数量 */
     /**
      * @brief 哈希函数指针
      * @param key 键指针
@@ -56,7 +56,7 @@ typedef struct smcc_hashmap {
      * @return 相同返回0，不同返回非0
      */
     int (*key_cmp)(const void *key1, const void *key2);
-} hashmap_t;
+} scc_hashtable_t;
 
 /**
  * @brief 初始化哈希表结构
@@ -64,7 +64,7 @@ typedef struct smcc_hashmap {
  *
  * @warning 必须设置hash_func和key_cmp后才能使用
  */
-void hashmap_init(hashmap_t *ht);
+void scc_hashtable_init(scc_hashtable_t *ht);
 
 /**
  * @brief 插入/更新键值对
@@ -73,7 +73,7 @@ void hashmap_init(hashmap_t *ht);
  * @param value 值指针
  * @return 被替换的旧值指针（无替换返回NULL）
  */
-void *hashmap_set(hashmap_t *ht, const void *key, void *value);
+void *scc_hashtable_set(scc_hashtable_t *ht, const void *key, void *value);
 
 /**
  * @brief 查找键对应值
@@ -81,7 +81,7 @@ void *hashmap_set(hashmap_t *ht, const void *key, void *value);
  * @param key 查找键指针
  * @return 找到返回值指针，未找到返回NULL
  */
-void *hashmap_get(hashmap_t *ht, const void *key);
+void *scc_hashtable_get(scc_hashtable_t *ht, const void *key);
 
 /**
  * @brief 删除键值对
@@ -91,7 +91,7 @@ void *hashmap_get(hashmap_t *ht, const void *key);
  *
  * @note 实际采用墓碑标记方式删除
  */
-void *hashmap_del(hashmap_t *ht, const void *key);
+void *scc_hashtable_del(scc_hashtable_t *ht, const void *key);
 
 /**
  * @brief 销毁哈希表
@@ -99,17 +99,18 @@ void *hashmap_del(hashmap_t *ht, const void *key);
  *
  * @note 仅释放哈希表内部内存，不会释放key/value内存
  */
-void hashmap_drop(hashmap_t *ht);
+void scc_hashtable_drop(scc_hashtable_t *ht);
 
 /**
- * @typedef hashmap_iter_fn
+ * @typedef scc_hashtable_iter_fn
  * @brief 哈希表迭代回调函数类型
  * @param key 当前键指针
  * @param value 当前值指针
  * @param context 用户上下文指针
  * @return 返回非0停止迭代
  */
-typedef int (*hashmap_iter_fn)(const void *key, void *value, void *context);
+typedef int (*scc_hashtable_iter_fn)(const void *key, void *value,
+                                     void *context);
 
 /**
  * @brief 遍历哈希表所有有效条目
@@ -117,6 +118,7 @@ typedef int (*hashmap_iter_fn)(const void *key, void *value, void *context);
  * @param iter_func 迭代回调函数
  * @param context 用户上下文指针
  */
-void hashmap_foreach(hashmap_t *ht, hashmap_iter_fn iter_func, void *context);
+void scc_hashtable_foreach(scc_hashtable_t *ht, scc_hashtable_iter_fn iter_func,
+                           void *context);
 
-#endif /* __SCC_HASHMAP_H__ */
+#endif /* __SCC_HASHTABLE_H__ */

runtime/scc_utils/include/scc_strpool.h

@@ -8,8 +8,8 @@
 #ifndef __SCC_STRPOOL_H__
 #define __SCC_STRPOOL_H__
 
-#include "hashmap.h"
-#include <libcore.h>
+#include "scc_hashtable.h"
+#include <scc_core.h>
 
 /**
  * @struct strpool_t
@@ -18,14 +18,14 @@
  * 组合哈希表和专用内存分配器实现的高效字符串存储池
  */
 typedef struct strpool {
-    hashmap_t ht; /**< 哈希表用于快速查找已存储字符串 */
-} strpool_t;
+    scc_hashtable_t ht; /**< 哈希表用于快速查找已存储字符串 */
+} scc_strpool_t;
 
 /**
  * @brief 初始化字符串池
  * @param pool 字符串池实例指针
  */
-void init_strpool(strpool_t *pool);
+void scc_strpool_init(scc_strpool_t *pool);
 
 /**
  * @brief 驻留字符串到池中
@@ -36,7 +36,7 @@ void init_strpool(strpool_t *pool);
  * @note 返回值生命周期与字符串池一致
  * @note 重复插入相同字符串会返回已有指针
  */
-const char *strpool_intern(strpool_t *pool, const char *str);
+const char *scc_strpool_intern(scc_strpool_t *pool, const char *str);
 
 /**
  * @brief 销毁字符串池
@@ -45,6 +45,25 @@ const char *strpool_intern(strpool_t *pool, const char *str);
  * @warning 销毁后已获取的字符串指针将失效
  * @note 会自动释放所有驻留字符串内存
  */
-void strpool_destroy(strpool_t *pool);
+void scc_strpool_drop(scc_strpool_t *pool);
+
+/**
+ * @typedef scc_hashtable_iter_fn
+ * @brief 哈希表迭代回调函数类型
+ * @param key 当前键指针
+ * @param value 当前值指针
+ * @param context 用户上下文指针
+ * @return 返回非0停止迭代
+ */
+typedef int (*scc_strpool_iter_fn)(const char *key, char *value, void *context);
+
+/**
+ * @brief 遍历字符串表所有有效条目
+ * @param ht 字符串表实例指针
+ * @param iter_func 迭代回调函数
+ * @param context 用户上下文指针
+ */
+void scc_strpool_foreach(scc_strpool_t *pool, scc_strpool_iter_fn iter_func,
+                         void *context);
 
 #endif /* __SCC_STRPOOL_H__ */

runtime/scc_utils/include/scc_utils.h

@@ -0,0 +1,9 @@
+#ifndef __SMCC_UTILS_H__
+#define __SMCC_UTILS_H__
+
+#include "kllist.h"
+#include "scc_hashtable.h"
+#include "scc_strpool.h"
+#include <scc_core.h>
+
+#endif /* __SMCC_UTILS_H__ */

runtime/scc_utils/src/hashtable.c

@@ -1,10 +1,10 @@
-#include <hashmap.h>
+#include <scc_hashtable.h>
 
 #ifndef SCC_INIT_HASHMAP_SIZE
 #define SCC_INIT_HASHMAP_SIZE (32)
 #endif
 
-void hashmap_init(hashmap_t *ht) {
+void scc_hashtable_init(scc_hashtable_t *ht) {
     scc_vec_init(ht->entries);
     ht->count = 0;
     ht->tombstone_count = 0;
@@ -21,17 +21,18 @@ static int next_power_of_two(int n) {
     return n + 1;
 }
 
-static hashmap_entry_t *find_entry(hashmap_t *ht, const void *key, u32 hash) {
+static scc_hashtable_entry_t *find_entry(scc_hashtable_t *ht, const void *key,
+                                         u32 hash) {
     if (ht->entries.cap == 0)
         return NULL;
 
     u32 index = hash & (ht->entries.cap - 1); // 容量是2的幂
     u32 probe = 0;
 
-    hashmap_entry_t *tombstone = NULL;
+    scc_hashtable_entry_t *tombstone = NULL;
 
     while (1) {
-        hashmap_entry_t *entry = &scc_vec_at(ht->entries, index);
+        scc_hashtable_entry_t *entry = &scc_vec_at(ht->entries, index);
         if (entry->state == ENTRY_EMPTY) {
             return tombstone ? tombstone : entry;
         }
@@ -53,25 +54,27 @@ static hashmap_entry_t *find_entry(hashmap_t *ht, const void *key, u32 hash) {
     return NULL;
 }
 
-static void adjust_capacity(hashmap_t *ht, int new_cap) {
+static void adjust_capacity(scc_hashtable_t *ht, usize new_cap) {
     new_cap = next_power_of_two(new_cap);
     Assert(new_cap >= ht->entries.cap);
 
-    SCC_VEC(hashmap_entry_t) old_entries;
+    SCC_VEC(scc_hashtable_entry_t) old_entries;
     old_entries.data = ht->entries.data;
     old_entries.cap = ht->entries.cap;
 
     // Not used size but for gdb python extention debug
     ht->entries.size = new_cap;
     ht->entries.cap = new_cap;
-    ht->entries.data = scc_realloc(NULL, new_cap * sizeof(hashmap_entry_t));
-    scc_memset(ht->entries.data, 0, new_cap * sizeof(hashmap_entry_t));
+    ht->entries.data =
+        scc_realloc(NULL, new_cap * sizeof(scc_hashtable_entry_t));
+    scc_memset(ht->entries.data, 0, new_cap * sizeof(scc_hashtable_entry_t));
 
     // rehash the all of the old data
     for (usize i = 0; i < old_entries.cap; i++) {
-        hashmap_entry_t *entry = &scc_vec_at(old_entries, i);
+        scc_hashtable_entry_t *entry = &scc_vec_at(old_entries, i);
         if (entry->state == ENTRY_ACTIVE) {
-            hashmap_entry_t *dest = find_entry(ht, entry->key, entry->hash);
+            scc_hashtable_entry_t *dest =
+                find_entry(ht, entry->key, entry->hash);
             *dest = *entry;
         }
     }
@@ -80,7 +83,7 @@ static void adjust_capacity(hashmap_t *ht, int new_cap) {
     ht->tombstone_count = 0;
 }
 
-void *hashmap_set(hashmap_t *ht, const void *key, void *value) {
+void *scc_hashtable_set(scc_hashtable_t *ht, const void *key, void *value) {
     if (ht->count + ht->tombstone_count >= ht->entries.cap * 0.75) {
         int new_cap = ht->entries.cap < SCC_INIT_HASHMAP_SIZE
                           ? SCC_INIT_HASHMAP_SIZE
@@ -89,7 +92,7 @@ void *hashmap_set(hashmap_t *ht, const void *key, void *value) {
     }
 
     u32 hash = ht->hash_func(key);
-    hashmap_entry_t *entry = find_entry(ht, key, hash);
+    scc_hashtable_entry_t *entry = find_entry(ht, key, hash);
 
     void *old_value = NULL;
     if (entry->state == ENTRY_ACTIVE) {
@@ -107,21 +110,21 @@ void *hashmap_set(hashmap_t *ht, const void *key, void *value) {
     return old_value;
 }
 
-void *hashmap_get(hashmap_t *ht, const void *key) {
+void *scc_hashtable_get(scc_hashtable_t *ht, const void *key) {
     if (ht->entries.cap == 0)
         return NULL;
 
     u32 hash = ht->hash_func(key);
-    hashmap_entry_t *entry = find_entry(ht, key, hash);
+    scc_hashtable_entry_t *entry = find_entry(ht, key, hash);
     return (entry && entry->state == ENTRY_ACTIVE) ? entry->value : NULL;
 }
 
-void *hashmap_del(hashmap_t *ht, const void *key) {
+void *scc_hashtable_del(scc_hashtable_t *ht, const void *key) {
     if (ht->entries.cap == 0)
         return NULL;
 
     u32 hash = ht->hash_func(key);
-    hashmap_entry_t *entry = find_entry(ht, key, hash);
+    scc_hashtable_entry_t *entry = find_entry(ht, key, hash);
 
     if (entry == NULL || entry->state != ENTRY_ACTIVE)
         return NULL;
@@ -133,17 +136,18 @@ void *hashmap_del(hashmap_t *ht, const void *key) {
     return value;
 }
 
-void hashmap_drop(hashmap_t *ht) {
+void scc_hashtable_drop(scc_hashtable_t *ht) {
     scc_vec_free(ht->entries);
     ht->count = 0;
     ht->tombstone_count = 0;
 }
 
-void hashmap_foreach(hashmap_t *ht, hashmap_iter_fn iter_func, void *context) {
+void scc_hashtable_foreach(scc_hashtable_t *ht, scc_hashtable_iter_fn iter_func,
+                           void *context) {
     for (usize i = 0; i < ht->entries.cap; i++) {
-        hashmap_entry_t *entry = &scc_vec_at(ht->entries, i);
+        scc_hashtable_entry_t *entry = &scc_vec_at(ht->entries, i);
         if (entry->state == ENTRY_ACTIVE) {
-            if (!iter_func(entry->key, entry->value, context)) {
+            if (iter_func(entry->key, entry->value, context)) {
                 break; // enable callback function terminal the iter
             }
         }

runtime/scc_utils/src/strpool.c

@@ -0,0 +1,32 @@
+#include <scc_strpool.h>
+
+void scc_strpool_init(scc_strpool_t *pool) {
+    pool->ht.hash_func = (u32 (*)(const void *))scc_strhash32;
+    pool->ht.key_cmp = (int (*)(const void *, const void *))scc_strcmp;
+    scc_hashtable_init(&pool->ht);
+}
+
+const char *scc_strpool_intern(scc_strpool_t *pool, const char *str) {
+    void *existing = scc_hashtable_get(&pool->ht, str);
+    if (existing) {
+        return existing;
+    }
+
+    usize len = scc_strlen(str) + 1;
+    char *new_str = scc_malloc(len);
+    if (!new_str) {
+        LOG_ERROR("strpool: Failed to allocate memory for string");
+        return NULL;
+    }
+    scc_memcpy(new_str, str, len);
+
+    scc_hashtable_set(&pool->ht, new_str, new_str);
+    return new_str;
+}
+
+void scc_strpool_drop(scc_strpool_t *pool) { scc_hashtable_drop(&pool->ht); }
+
+void scc_strpool_foreach(scc_strpool_t *pool, scc_strpool_iter_fn iter_func,
+                         void *context) {
+    scc_hashtable_foreach(&pool->ht, (scc_hashtable_iter_fn)iter_func, context);
+}

tools/cbuild/cbuild.py

@@ -608,9 +608,10 @@ class GccCompiler(Compiler):
         flags = {
             BuildMode.TEST: [
                 "-DTEST_MODE",
-                "-O2",
+                "-O0",
                 "-g",
                 "--coverage",
+                "-fprofile-update=atomic",
                 "-Wall",
                 "-Wextra",
             ],
@@ -647,9 +648,10 @@ class ClangCompiler(Compiler):
         flags = {
             BuildMode.TEST: [
                 "-DTEST_MODE",
-                "-O2",
+                "-O0",
                 "-g",
                 "--coverage",
+                "-fprofile-update=atomic",
                 "-Wall",
                 "-Wextra",
             ],
@@ -861,7 +863,7 @@ class PackageBuilder:
         """打印依赖树"""
         self.context.resolver.print_tree()
 
-    def tests(self, filter_str: str = ""):
+    def tests(self, filter_str: str = "", timeout: int = 30):
         """运行测试"""
         targets = [
             t for t in self.context.get_targets() if t.type == TargetType.TEST_EXEC
@@ -888,7 +890,7 @@ class PackageBuilder:
         for target in targets:
             logger.info("运行测试: %s", target.name)
             try:
-                result = subprocess.run(target.output, check=True, timeout=30)
+                result = subprocess.run(target.output, check=True, timeout=timeout)
                 if result.returncode == 0:
                     print(f"  ✓ 测试 {target.name} 通过")
                     passed += 1
@@ -1064,13 +1066,16 @@ def create_parser():
     # test 命令
     test_parser = subparsers.add_parser("test", help="运行测试")
     add_common_args(test_parser)
+    test_parser.add_argument("--timeout", "-t", type=int, default=3, help="测试时间")
     test_parser.add_argument("--filter", default="", help="过滤测试")
     test_parser.set_defaults(mode=BuildMode.TEST)
 
     # clean 命令
     clean_parser = subparsers.add_parser("clean", help="清理构建产物")
     add_common_args(clean_parser)
-    clean_parser.add_argument("--all", action="store_true", help="清理所有模式")
+    clean_parser.add_argument(
+        "-a", "--all", action="store_true", default=True, help="清理所有模式"
+    )
 
     # tree 命令
     tree_parser = subparsers.add_parser("tree", help="显示依赖树")
@@ -1131,7 +1136,7 @@ def main():
         builder.run()
     elif args.command == "test":
         builder.build([TargetType.TEST_EXEC])
-        builder.tests(getattr(args, "filter", ""))
+        builder.tests(args.filter, args.timeout)
     elif args.command == "clean":
         if hasattr(args, "all") and args.all:
             # 清理所有模式

tools/wc.py

@@ -0,0 +1,53 @@
+"""统计目录下C/C++文件的行数(write by AI)"""
+
+import os
+
+
+def count_lines(file_path):
+    """统计单个文件的代码行数"""
+    try:
+        with open(file_path, "rb") as f:  # 二进制模式读取避免编码问题
+            return sum(1 for _ in f)
+    except UnicodeDecodeError:
+        print(f"警告：无法解码文件 {file_path}（可能不是文本文件）")
+        return 0
+    except Exception as e:
+        print(f"读取 {file_path} 出错: {str(e)}")
+        return 0
+
+
+def scan_files(directory, exclude_dirs=None):
+    """扫描目录获取所有C/C++文件"""
+    if exclude_dirs is None:
+        exclude_dirs = [".git", "venv", "__pycache__", ".old"]  # 默认排除的目录
+
+    c_files = []
+    for root, dirs, files in os.walk(directory):
+        # 跳过排除目录
+        dirs[:] = [d for d in dirs if d not in exclude_dirs]
+
+        for file in files:
+            if file.endswith((".c", ".h")):
+                full_path = os.path.join(root, file)
+                c_files.append(full_path)
+    return c_files
+
+
+def main():
+    """main function"""
+    target_dir = input("请输入要扫描的目录路径（留空为当前目录）: ") or "."
+
+    files = scan_files(target_dir)
+    total_lines = 0
+
+    print("\n统计结果：")
+    for idx, file in enumerate(files, 1):
+        lines = count_lines(file)
+        total_lines += lines
+        print(f"{idx:4d}. {file} ({lines} 行)")
+
+    print(f"\n总计: {len(files)} 个C/C++文件，共 {total_lines} 行代码")
+
+
+if __name__ == "__main__":
+    main()