我完全是出自个人兴趣学习一下我们的词法，在假期有空去了解这个庞大的机器gcc的运行
词法分析的本质就是有一个自动机，首先我们可以识别出这个东西是标识符还是其他的东西
我们希望偏重代码的讲解，在lex.c里面，我们是实现了我们的代码
首先，我们进行c——lex_one_token，c_lex_direct和c_classify_number函数完成了所有符号的识别，
_c_lex_directr是符号的识别，cpp_classify_number实现对数字的
进入起始状态的代码如图所示:
cpp_token *_cpp_lex_direct(cpp_reader *pfile){fresh_line;buffer = pfile->buffer;skipped_white;c=*buffer->cur++;}
代码路径
switch (c){case ' ': case 't': case 'f': case 'v': case ' ':result->flags |= PREV_WHITE;skip_whitespace (pfile, c);goto skipped_white;case 'n':if (buffer->cur < buffer->rlimit)CPP_INCREMENT_LINE (pfile, 0);buffer->need_line = true;goto fresh_line;
如果是这四个，如果走到n的时候，就是fresh_line
然后确符号为数字
case '0': case '1': case '2': case '3': case '4':case '5': case '6': case '7': case '8': case '9':{struct normalize_state nst = INITIAL_NORMALIZE_STATE;result->type = CPP_NUMBER;lex_number (pfile, &result->val.str, &nst);单独的处理函数warn_about_normalization (pfile, result, &nst);break;}
然后进入lex_number
case '.':result->type = CPP_DOT;if (ISDIGIT (*buffer->cur)){struct normalize_state nst = INITIAL_NORMALIZE_STATE;result->type = CPP_NUMBER;lex_number (pfile, &result->val.str, &nst);warn_about_normalization (pfile, result, &nst);}else if (*buffer->cur == '.' && buffer->cur[1] == '.')buffer->cur += 2, result->type = CPP_ELLIPSIS;else if (*buffer->cur == '*' && CPP_OPTION (pfile, cplusplus))buffer->cur++, result->type = CPP_DOT_STAR;break;
类似的，我们符号的第一个是'L',''U',u','R'
确定符号为字符和字符串，符号的第一个字符是""或者""
确定符号为标识符的东西，在识别到"L""u""U""R"的时候我们就是标识符。然后其他的情况是放进函数进行判别
分析运算符和分隔符
如果/:
case ''':case '"':lex_string (pfile, result, buffer->cur - 1);//数字处理break;case '/':/* A potential block or line comment. */comment_start = buffer->cur;//获取后面的c = *buffer->cur;if (c == '*'){if (_cpp_skip_block_comment (pfile))cpp_error (pfile, CPP_DL_ERROR, "unterminated comment");}//遍历我们的东西else if (c == '/' && (CPP_OPTION (pfile, cplusplus_comments)|| cpp_in_system_header (pfile))){/* Warn about comments only if pedantically GNUC89, and notin system headers. */if (CPP_OPTION (pfile, lang) == CLK_GNUC89 && CPP_PEDANTIC (pfile)&& ! buffer->warned_cplusplus_comments){cpp_error (pfile, CPP_DL_PEDWARN,"C++ style comments are not allowed in ISO C90");cpp_error (pfile, CPP_DL_PEDWARN,"(this will be reported only once per input file)");buffer->warned_cplusplus_comments = 1;}if (skip_line_comment (pfile) && CPP_OPTION (pfile, warn_comments))cpp_error (pfile, CPP_DL_WARNING, "multi-line comment");}else if (c == '='){buffer->cur++;result->type = CPP_DIV_EQ;break;}else{result->type = CPP_DIV;break;}
考虑"<"
case '<':if (pfile->state.angled_headers){lex_string (pfile, result, buffer->cur - 1);if (result->type != CPP_LESS)break;}result->type = CPP_LESS;if (*buffer->cur == '=')buffer->cur++, result->type = CPP_LESS_EQ;else if (*buffer->cur == '<'){buffer->cur++;IF_NEXT_IS ('=', CPP_LSHIFT_EQ, CPP_LSHIFT);}else if (CPP_OPTION (pfile, digraphs)){if (*buffer->cur == ':'){buffer->cur++;result->flags |= DIGRAPH;result->type = CPP_OPEN_SQUARE;}else if (*buffer->cur == '%'){buffer->cur++;result->flags |= DIGRAPH;result->type = CPP_OPEN_BRACE;}}
其他的算法一样，我们就不再赘述
然后是lex_identifier函数
static cpp_hashnode *lex_identifier (cpp_reader *pfile, const uchar *base, bool starts_ucn,struct normalize_state *nst){cpp_hashnode *result;const uchar *cur;unsigned int len;unsigned int hash = HT_HASHSTEP (0, *base);cur = pfile->buffer->cur;if (! starts_ucn)while (ISIDNUM (*cur)){hash = HT_HASHSTEP (hash, *cur);cur++;}pfile->buffer->cur = cur;if (starts_ucn || forms_identifier_p (pfile, false, nst)){/* Slower version for identifiers containing UCNs (or $). */do {while (ISIDNUM (*pfile->buffer->cur)){pfile->buffer->cur++;NORMALIZE_STATE_UPDATE_IDNUM (nst);}} while (forms_identifier_p (pfile, false, nst));result = _cpp_interpret_identifier (pfile, base,pfile->buffer->cur - base);}else{len = cur - base;hash = HT_HASHFINISH (hash, len);result = CPP_HASHNODE (ht_lookup_with_hash (pfile->hash_table,base, len, hash, HT_ALLOC));}/* Rarely, identifiers require diagnostics when lexed. */if (__builtin_expect ((result->flags & NODE_DIAGNOSTIC)&& !pfile->state.skipping, 0)){/* It is allowed to poison the same identifier twice. */if ((result->flags & NODE_POISONED) && !pfile->state.poisoned_ok)cpp_error (pfile, CPP_DL_ERROR, "attempt to use poisoned "%s"",NODE_NAME (result));/* Constraint 6.10.3.5: __VA_ARGS__ should only appear in thereplacement list of a variadic macro. */if (result == pfile->spec_nodes.n__VA_ARGS__&& !pfile->state.va_args_ok)cpp_error (pfile, CPP_DL_PEDWARN,"__VA_ARGS__ can only appear in the expansion"" of a C99 variadic macro");}return result;}
这里我们主要那就是满足条件的可以
·然后就是数字的词法分析实现
static voidlex_number (cpp_reader *pfile, cpp_string *number,struct normalize_state *nst){const uchar *cur;const uchar *base;uchar *dest;base = pfile->buffer->cur - 1;do{cur = pfile->buffer->cur;/* N.B. ISIDNUM does not include $. */while (ISIDNUM (*cur) || *cur == '.' || VALID_SIGN (*cur, cur[-1])){cur++;NORMALIZE_STATE_UPDATE_IDNUM (nst);}pfile->buffer->cur = cur;}while (forms_identifier_p (pfile, false, nst));number->len = cur - base;dest = _cpp_unaligned_alloc (pfile, number->len + 1);memcpy (dest, base, number->len);dest[number->len] = ' ';number->text = dest;}/* Create a token of type TYPE with a literal spelling. */static voidcreate_literal (cpp_reader *pfile, cpp_token *token, const uchar *base,unsigned int len, enum cpp_ttype type){uchar *dest = _cpp_unaligned_alloc (pfile, len + 1);memcpy (dest, base, len);dest[len] = ' ';token->type = type;token->val.str.len = len;token->val.str.text = dest;}
发表评论