problem with text parsing function

[hojoff79]

I am writing a program to search through some text for information. It looks through a line like this

<div class="lvPrice">$50,000</div>

and pulls out the information contained in that line (it transforms the whole line to a cstring, goes through and finds the "<div class="lvPrice">" part then goes through and finds the "<" immediately after this and then prints the cstring elements inbetween them.

The code works when the line is not proceeded by any spaces, but when the line has some spaces, it somehow throws off the printing of the value so it does the wrong thing. For example if the line is indented 1 space as such

<div class="lvPrice">$50,000</div>

then the value printed out will start 1 character earlier in the sequence. so instead of printing "$50,000" like it does in the program with no spaces, the line with 2 spaces would print ">$50,00". Why is this happening and what can I do to fix it??? Thanks for your help in advance

here is the code of the function that is doing the process

// function to parse a file and add it to a new file
void parseandwrite (string writefile, string searchword)
{
     
     string line;
     
     ifstream myfile (writefile.c_str());
  if (myfile.is_open())
  {
  cout << "VIEW OF FILE " << writefile << endl;
    while (! myfile.eof() )
    {
      getline (myfile,line);
      
      //convert the line to a cstr for easier use
      char * linechar, * swordchar;
      linechar = new char [line.size()+1];
      strcpy(linechar, line.c_str());
      swordchar = new char [searchword.size()+1];
      strcpy(swordchar, searchword.c_str());
      
      
      
      //search through the line for the term and end or info and write info
      int linesize, index = 0, found, end;
      char read;
      
      linesize = line.size();
            
      for (int k=0; k<linesize; k++)
      {
          // search for the word using the char by char check and incrementing index
          if (linechar[k] == swordchar[index])
          {
                          index++;
          }
          else index = 0;
          
          //check to see if the word has been found
          if (index == searchword.size())
          {
                    found = index;
                    for (int r=found; r<linesize; r++)
                    {
                        if (linechar[r] == '<')
                        {
                                        end = r;
                                        break;
                        }
                    }
                    cout << "Price:";
                    for (int g=found; g<end; g++)
                    {
                        cout<< linechar[g];
                    }
                    cout << endl;
          }
      }
            
      
      
      
    }
    myfile.close();
  }

  else cout << "Error opening the file" << endl; 

}

[treuss]

I am writing a program to search through some text for information. It looks through a line like this

<div class="lvPrice">$50,000</div>

and pulls out the information contained in that line (it transforms the whole line to a cstring, goes through and finds the "<div class="lvPrice">" part then goes through and finds the "<" immediately after this and then prints the cstring elements inbetween them.

Why C-Strings? Use std::string::find and std::string::substr instead.

[Philip Nicoletti]

As treuss mentioned, it will be easier to sue std::string functions.

If you stay with your current setup : you are leaking memory within
the loop (two new's ... no delete's).

[andrey_zh]

Do you know what is a finite state machine(automaton)? You could use it to parse a string

[D_Drmmr]

I am writing a program to search through some text for information. It looks through a line like this

<div class="lvPrice">$50,000</div>

You might want to consider using an xml parser. There are some good, free C++ libraries available.

[hojoff79]

I have been using cstrings so that i can using the char[index] index find what character the > ends at and what index the < starts and make a string with everything inbetween (aka, the actual information). I would not know how to do this with string functions (aka, how to single out the information inbetween the <div class="lvPrice"> and the next <

[Philip Nicoletti]

As treuss mentioned, you would use substr to get the string
between the ">" and the "<" ...

#include <iostream>
#include <fstream>
#include <string>

using namespace std;

// function to parse a file and add it to a new file
void parseandwrite (const string & writefile, const string & searchword)
{
    string line;
     
    ifstream myfile (writefile.c_str());

    if (myfile.is_open())
    {
        cout << "VIEW OF FILE " << writefile << endl;
        while ( getline (myfile,line) )
        {
            size_t pos1 = line.find(searchword);        // look for the search word

            if (pos1 != string::npos)                   // if found ...
            {
                pos1 = line.find(">",pos1);             // find the >

                if (pos1 != string::npos)               // if found ... 
                {
                    size_t pos2 = line.find("<",pos1);  // find the <

                    if (pos2 != string::npos)           // if found ...
                    {
                        cout << searchword << " : " << line.substr(pos1+1,pos2-pos1-1) << "\n";
                    }
                }
            }
        }
    }
    else
    {
        cout << "Error opening the file" << endl; 
    }
}


int main()
{
    parseandwrite ("test.txt", "Price");

    return 0;
}