layout: post comments: true title: 为GCC添加中文关键字 description: 通过修改gcc编译器源码, 添加中文关键词. Allow gcc to support unicode identifiers. date: 2017-11-24 01:40:00 -0700
知乎原文地址 作者:@狗屎咖啡
GCC不支持UTF-8字符，可以自己添加：
也可以 用脚本转换源代码，再编译：
https://gcc.gnu.org/wiki/FAQ#utf8_identifiers
但是强烈建议打上支持UTF-8的补丁。
实例 swizl/gcn
添加关键字
在 gcc/c-family/c-common.c 中
const struct c_common_resword c_common_reswords[] 下添加
复制一行原关键字，再将字符串该成中文
例如
{ "asm", RID_ASM, D_ASM },
{ "汇编", RID_ASM, D_ASM },
{ "auto", RID_AUTO, 0 },
{ "自动", RID_AUTO, 0 },
{ "bool", RID_BOOL, D_CXXONLY | D_CXXWARN },
{ "布尔", RID_BOOL, D_CXXONLY | D_CXXWARN },
{ "break", RID_BREAK, 0 },
{ "断", RID_BREAK, 0 },
{ "case", RID_CASE, 0 },
{ "例", RID_CASE, 0 },
{ "catch", RID_CATCH, D_CXX_OBJC | D_CXXWARN },
{ "抓", RID_CATCH, D_CXX_OBJC | D_CXXWARN },
{ "char", RID_CHAR, 0 },
{ "字", RID_CHAR, 0 },
添加中文宏关键字
在 libcpp/directives.c 中
1 添加宏定义A，仿照D，加一个参数name_cn. 因为原始GCC不支持UTF-8
当然已经打了支持UTF-8的补丁，没有这个问题。兼容起见，还是按照这样改。
#define D(name, t, o, f) static void do_##name (cpp_reader *);
DIRECTIVE_TABLE
#undef D
#define D(n, tag, o, f) tag,
enum
{
DIRECTIVE_TABLE
N_DIRECTIVES
};
#undef D
#define D(name, t, origin, flags) \
{ do_##name, (const uchar *) #name, \
sizeof #name - 1, origin, flags },
static const directive dtable[] =
{
DIRECTIVE_TABLE
};
#undef D
#define D(name, t, origin, flags) #name,
static const char * const directive_names[] = {
DIRECTIVE_TABLE
NULL
};
#undef D
改成
#define A(name_cn, name, t, o, f)
#define D(name, t, o, f) static void do_##name (cpp_reader *);
DIRECTIVE_TABLE
#undef D
#undef A
#define A(n_cn, n, tag, o, f) tag ## _CN,
#define D(n, tag, o, f) tag,
enum
{
DIRECTIVE_TABLE
N_DIRECTIVES
};
#undef D
#undef A
#define A(name_cn, name, t, origin, flags) \
{ do_##name, (const uchar *) name_cn, \
sizeof (name_cn) - 1, origin, flags },
#define D(name, t, origin, flags) \
{ do_##name, (const uchar *) #name, \
sizeof #name - 1, origin, flags },
static const directive dtable[] =
{
DIRECTIVE_TABLE
};
#undef D
#undef A
#define A(name_cn, name, t, origin, flags) name_cn,
#define D(name, t, origin, flags) #name,
static const char * const directive_names[] = {
DIRECTIVE_TABLE
NULL
};
#undef D
#undef A
2 添加宏关键字
在 #define DIRECTIVE_TABLE 中添加
复制一行原宏关键字，该D为A，添加中文名 字符串。
例如
#define DIRECTIVE_TABLE \
D(define, T_DEFINE = 0, KANDR, IN_I) /* 270554 */ \
A("定义", define, T_DEFINE, KANDR, IN_I) /* 270554 */ \
D(include, T_INCLUDE, KANDR, INCL | EXPAND) /* 52262 */ \
A("含", include, T_INCLUDE, KANDR, INCL | EXPAND) /* 52262 */ \
D(endif, T_ENDIF, KANDR, COND) /* 45855 */ \
A("了如", endif, T_ENDIF, KANDR, COND) /* 45855 */ \
D(ifdef, T_IFDEF, KANDR, COND | IF_COND) /* 22000 */ \
A("如定义", ifdef, T_IFDEF, KANDR, COND | IF_COND) /* 22000 */ \
D(if, T_IF, KANDR, COND | IF_COND | EXPAND) /* 18162 */ \
A("如", if, T_IF, KANDR, COND | IF_COND | EXPAND) /* 18162 */ \
D(else, T_ELSE, KANDR, COND) /* 9863 */ \
A("另", else, T_ELSE, KANDR, COND) /* 9863 */ \
D(ifndef, T_IFNDEF, KANDR, COND | IF_COND) /* 9675 */ \
A("如未定义", ifndef, T_IFNDEF, KANDR, COND | IF_COND) /* 9675 */ \
D(undef, T_UNDEF, KANDR, IN_I) /* 4837 */ \
A("消定义", undef, T_UNDEF, KANDR, IN_I) /* 4837 */ \
D(line, T_LINE, KANDR, EXPAND) /* 2465 */ \
A("行", line, T_LINE, KANDR, EXPAND) /* 2465 */ \
D(elif, T_ELIF, STDC89, COND | EXPAND) /* 610 */ \
A("另如", elif, T_ELIF, STDC89, COND | EXPAND) /* 610 */ \
D(error, T_ERROR, STDC89, 0) /* 475 */ \
A("错误", error, T_ERROR, STDC89, 0) /* 475 */ \
D(pragma, T_PRAGMA, STDC89, IN_I) /* 195 */ \
A("杂注", pragma, T_PRAGMA, STDC89, IN_I) /* 195 */ \
D(warning, T_WARNING, EXTENSION, 0) /* 22 */ \
A("告警", warning, T_WARNING, EXTENSION, 0) /* 22 */ \
D(include_next, T_INCLUDE_NEXT, EXTENSION, INCL | EXPAND) /* 19 */ \
A("含下个", include_next, T_INCLUDE_NEXT, EXTENSION, INCL | EXPAND) /* 19 */ \
D(ident, T_IDENT, EXTENSION, IN_I) /* 11 */ \
D(import, T_IMPORT, EXTENSION, INCL | EXPAND) /* 0 ObjC */ \
A("导入", import, T_IMPORT, EXTENSION, INCL | EXPAND) /* 0 ObjC */ \
D(assert, T_ASSERT, EXTENSION, DEPRECATED) /* 0 SVR4 */ \
A("断言", assert, T_ASSERT, EXTENSION, DEPRECATED) /* 0 SVR4 */ \
D(unassert, T_UNASSERT, EXTENSION, DEPRECATED) /* 0 SVR4 */ \
A("消断言", unassert, T_UNASSERT, EXTENSION, DEPRECATED) /* 0 SVR4 */ \
D(sccs, T_SCCS, EXTENSION, IN_I) /* 0 SVR4? */
make bootstrap && make install 之后，配合支持UTF-8的补丁，就可以愉快地用中文关键字了。