weTagScanner.h

Go to the documentation of this file.

/*
    webEngine is the HTML processing library
    Copyright (C) 2009 Andrew Abramov aabramov@ptsecurity.ru

    This file is part of webEngine

    webEngineis free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.

    webEngineis distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
    along with webEngine.  If not, see <http://www.gnu.org/licenses/>.
*/
#ifndef __TAGSCANNER_H__
#define __TAGSCANNER_H__

#include <string>
#include <vector>
#include <stdlib.h> // wchar_t
#include <string.h> // strcmp, etc.

using namespace std;

class WeInStream {
public:
    virtual char GetChar() = 0;
    virtual void PushBack(char c) = 0;
    virtual int GetPos() = 0;
};

class WeStrStream : public WeInStream
{
    char* p;
    const char* start;
    const char* end;
public:
    WeStrStream(const char* src): p((char*)src), start(src), end(src + strlen(src)) {}
    virtual char GetChar() { return p < end? *p++: 0; }
    virtual void PushBack(char c) {if(p > start) {p--;}}
    virtual int GetPos() { return p - start; };
};

enum WeScannerToken {
    wstError = -1,
    wstEof = 0,
    wstTagStart,
    wstTagEnd,
    wstAttr,
    wstWord,
    wstSpace,
    wstData,
    wstCommentStart, wstCommentEnd, // after "<!--" and "-->"
    wstCDataStart, wstCDataEnd,     // after "<![CDATA[" and "]]>"
    wstPiStart, wstPiEnd,           // after "<?" and "?>"
    wstEntityStart, wstEntityEnd    // after "<!ENTITY" and ">"
};

class WeTagScanner {
public:

    WeTagScanner(WeInStream& is):
            input(is),
            input_char(0),
//            value_length(0),
//            tag_name_length(0),
//            attr_name_length(0),
            got_tail(false) { c_scan = &WeTagScanner::ScanBody; }
    virtual ~WeTagScanner(){};

      WeScannerToken    GetToken() { return (this->*c_scan)(); }
      const char*       GetValue();
      const char*       GetAttrName();
      const char*       GetTagName();

      virtual char   ResolveEntity(const char* buf, int buf_size) { return 0; }
      virtual int    GetPos() { return input.GetPos(); };

private:
    /* types */
    static const int MAX_TOKEN_SIZE = 1024;
    static const int MAX_NAME_SIZE = 128;

    typedef WeScannerToken (WeTagScanner::*fnScan)();
    fnScan        c_scan; // current 'reader'

    /* methods */
    // content 'readers'
    WeScannerToken  ScanBody();
    WeScannerToken  ScanHead();
    WeScannerToken  ScanComment();
    WeScannerToken  ScanCdata();
    WeScannerToken  ScanPi();
    WeScannerToken  ScanTag();
    WeScannerToken  ScanEntityDecl();

    char            SkipWhitespace();
    void            PushBack(char c);

    char            GetChar();
    char            ScanEntity();

    bool            IsWhitespace(char c);

    void            AppendValue(char c);
    void            AppendAttrName(char c);
    void            AppendTagName(char c);

    /* data */
#ifndef __DOXYGEN__
    WeScannerToken  token;

    vector<char>    value;
    //int             value_length;

    vector<char>    tag_name;
    //int             tag_name_length;

    vector<char>    attr_name;
    //int             attr_name_length;

    WeInStream&     input;              
    char            input_char;         

    bool            got_tail;           
#endif //__DOXYGEN__
};
#endif //__TAGSCANNER_H__

---

	©2009 Positive Technologies	Generated on Tue Jul 14 18:27:30 2009 for WebEngine by		1.5.7