Posted on 2008-07-08 11:49
RichardHe 閱讀(2212)
評(píng)論(1) 編輯 收藏 引用 所屬分類:
[轉(zhuǎn)]
出自http://hi.baidu.com/yesbaba/blog/item/79c3eb13215da0d2f7039ec9.html
1.1版，命令行下使用如1.0版，
增加更易用的方法：點(diǎn)擊coco.exe，將文件拖入窗口，高亮文件輸出到同名txt文件中
轉(zhuǎn)成html的功能以后不懶的時(shí)候再寫吧
1.0版：命令行下使用，輸出到stdout，
可以重定向到文件，彩色復(fù)制就行了
coco.exe xxx.cpp > whateverfile
記事本打開whateverfile，復(fù)制，在bbs上彩色粘貼
編譯這個(gè)大概需要2百多M內(nèi)存
想直接現(xiàn)在用的，可以在這下載，這是水木的附件，不知道能有效到什么時(shí)候
http://www.newsmth.net/att.php?p.335.193566.17417.exe
論壇可以用它來(lái)實(shí)現(xiàn)代碼高亮顯示
以后再改進(jìn)
正確的代碼coco應(yīng)該不會(huì)分析錯(cuò)，能分析正確的不一定是正確的代碼
讀代碼可以先看spirit最基礎(chǔ)的ch_p,str_p等，還有spirit文檔中的utility
再看如何分析文件
http://hi.baidu.com/yesbaba/blog/item/091ca995d0fe6e49d1135e8b.html
gnu source highlight就可以實(shí)現(xiàn)這個(gè)功能了，我做這個(gè)一來(lái)是學(xué)spirit練練手，
二來(lái)是要做個(gè)課程設(shè)計(jì)，三來(lái)這個(gè)很實(shí)用，平常在水木帖代碼也能用的上
///////////////////////////////////////////////////////////////////////////////
//
// highlight c++ source code in ansi code format
// COCO (COlorful COde) version 1.1
// ibe@newsmth.net
// [ 2007-08-13 ]
//
///////////////////////////////////////////////////////////////////////////////
#include <boost/spirit/core.hpp>
#include <boost/spirit/iterator/file_iterator.hpp>
#include <boost/spirit/utility/confix.hpp>
#include <boost/spirit/utility/escape_char.hpp>
#include <iostream>
#include <fstream>
///////////////////////////////////////////////////////////////////////////////
using namespace boost::spirit;
using namespace std;
////////////////////////////////////////////////////////////////////////////
//
// Types
//types needed for file parsing
////////////////////////////////////////////////////////////////////////////
typedef char char_t;
typedef file_iterator < char_t > iterator_t;
typedef scanner < iterator_t > scanner_t;
typedef rule < scanner_t > rule_t;
////////////////////////////////////////////////////////////////////////////
//To ansi，action函數(shù)，匹配后輸出成ansi代碼的彩色控制符，
//可參考各大高校bbs的asciiart版精華區(qū)，可以輸出成需要的格式，如html
//
////////////////////////////////////////////////////////////////////////////
ofstream outfile;
namespace to_ansi
{
void
black (iterator_t first, iterator_t const &last)
{
outfile << "\033[1;30m";
while (first != last)
outfile << *first++;
outfile << "\033[m";
}
void
red (iterator_t first, iterator_t const &last)
{
outfile << "\033[1;31m";
while (first != last)
outfile << *first++;
outfile << "\033[m";
}
void
green (iterator_t first, iterator_t const &last)
{
outfile << "\033[1;32m";
while (first != last)
outfile << *first++;
outfile << "\033[m";
}
void
yellow (iterator_t first, iterator_t const &last)
{
outfile << "\033[1;33m";
while (first != last)
outfile << *first++;
outfile << "\033[m";
}
void
blue (iterator_t first, iterator_t const &last)
{
outfile << "\033[1;34m";
while (first != last)
outfile << *first++;
outfile << "\033[m";
}
void
magenta (iterator_t first, iterator_t const &last)
{
outfile << "\033[1;35m";
while (first != last)
outfile << *first++;
outfile << "\033[m";
}
void
cyan (iterator_t first, iterator_t const &last)
{
outfile << "\033[1;36m";
while (first != last)
outfile << *first++;
outfile << "\033[m";
}
void
white (iterator_t first, iterator_t const &last)
{
outfile << "\033[1;37m";
while (first != last)
outfile << *first++;
outfile << "\033[m";
}
void
echo (iterator_t first, iterator_t const &last)
{
while (first != last)
outfile << *first++;
}
}
////////////////////////////////////////////////////////////////////////////
//
// cpp lex
//c++的詞法描述，有了comment_p就方便多了，識(shí)別函數(shù)名還沒(méi)實(shí)現(xiàn)
////////////////////////////////////////////////////////////////////////////
namespace cpp_lex
{
rule_t comment = comment_p ("/*", "*/")
| comment_p ("http://")
;
rule_t whitespace = space_p
;
rule_t include = str_p ("#include") >> *space_p >>
(comment_p ("<", ">") | comment_p ("\"", "\""))
;
rule_t preprocessor = (include | "##" | "#define" | "#error" | ("#if" >> space_p)
| "#ifdef" | "#ifndef" | "#else" | "#elif"
| "#endif" | "#line" | "#pragma" | "#undef" | "#"
| "__LINE__" | "__FILE__" | "__DATE__" | "__TIME__"
| "_cplusplus" | "__STDC__")
>> space_p
;
rule_t keyword_ = str_p ("asm") | "auto" | "bool" | "break" | "case"
| "catch" | "char" | "class" | "const" | "const_cast"
| "continue" | "default" | "delete" | "do" | "double"
| "dynamic_cast" | "else" | "enum" | "explicit"
| "extern" | "false" | "float" | "for" | "friend"
| "goto" | "if" | "inline" | "int" | "long" | "mutable"
| "namespace" | "new" | "operator" | "private"
| "protected" | "public" | "register" | "reinterpret_cast"
| "return" | "short" | "signed" | "sizeof" | "static"
| "static_cast" | "struct" | "switch" | "template"
| "this" | "throw" | "true" | "try" | "typedef" | "typeid"
| "typename" | "union" | "unsighed" | "using" | "virtual"
| "void" | "volatile" | "wchar_t" | "while"
;
rule_t keyword = keyword_ >> space_p
;
rule_t identifer = (alpha_p | '_') >> *(alnum_p | '_')
;
rule_t operators = punct_p - '`' - '@' - '$' - '\\'
;
rule_t number = real_p
;
rule_t str = confix_p ("\"", *c_escape_ch_p, "\"")
;
rule_t charcter = confix_p("\'", *c_escape_ch_p, "\'")
;
rule_t constant = number
| str
| charcter
;
};
////////////////////////////////////////////////////////////////////////////
//
// Main program
//
////////////////////////////////////////////////////////////////////////////
int
main (int argc, char *argv[])
{
string filepath;
if (2 > argc)
{//把要處理的文件拖到窗口內(nèi)
cout << "drag file to this windows\n";
string filepath_input;
getline(cin, filepath_input);
filepath_input.erase(filepath_input.end()-1); //去掉結(jié)尾的'\"'
filepath_input.erase(filepath_input.begin()); //去掉開頭"
for (int i = 0; filepath_input[i] != 0; i++) {
filepath.push_back(filepath_input[i]);
if (filepath_input[i] == '\\')
filepath.push_back('\\');
}
}else{
// for console usage
filepath = argv[1];
}
iterator_t first (filepath);
if (!first) {
std::cerr << "Unable to open file!\n";
return -1;
}
// Create an EOF iterator
iterator_t last = first.make_end ();
string filepath_output = filepath+".txt";
outfile.open(filepath_output.c_str());
// A simple rule詞法對(duì)應(yīng)的顏色在這里改就可以
rule_t r = *(
cpp_lex::comment[&to_ansi::cyan]
| cpp_lex::constant[&to_ansi::yellow]
| cpp_lex::preprocessor[&to_ansi::red]
| cpp_lex::keyword[&to_ansi::green]
| cpp_lex::whitespace[&to_ansi::echo]
| cpp_lex::operators[&to_ansi::magenta]
| cpp_lex::identifer[&to_ansi::white]
)
;
//
// Parse
/*The parse_info structure
The functions above return a parse_info structure parameterized by the iterator type passed in.
The parse_info struct has these members:parse_info
stop Points to the final parse
position (i.e The parser recognized and processed the input up to this
point)
hit True if parsing is successful.
This may be full: the parser consumed all the input, or partial: the
parser consumed only a portion of the input.
full True when we have a full match (i.e The parser consumed all the input).
length The number of characters
consumed by the parser. This is valid only if we have a successful
match (either partial or full).
*/
parse_info < iterator_t > info = parse (first, last, r);
// This really shouldn't fail...
if (info.full)
std::cout << "\nParse succeeded!\n";
else
std::cout << "\nParse failed!\n";
std::cout << "highlight file saved in " << filepath_output << endl;
string end;
getline (cin, end);//按回車鍵推出窗口
return 0;
}
這只是個(gè)詞法高亮程序，還不能識(shí)別語(yǔ)法，如函數(shù)等，用正則也能做，但boost.spirit更簡(jiǎn)單