Member Avatar for rasha80

i want to build program that recive (c code in txt file) then searching for words (int , double, char, float) and count number of vaiables appear according to that. then store the variables in array (e.g. int x; i want to store X in array and so on). i build code that search of word e.g (int) in file and count number of its appearance, but still i dont know how to extract the word after the searched word. (e.g X,Y..etc)

  • 2 Contributors
  • 2 Replies
  • 226 Views
  • 3 Days Discussion Span
  • Latest Post Latest Post by rasha80

Dani AI

Generated

— the stream-extraction approach in the thread finds the token after the type but leaves punctuation (like x;) and can't handle comma-separated declarations, initializers, pointers or arrays. asked for code; below is a short, practical two-pass pattern that solves the common cases without rewriting a compiler.

First pass: read the file into a vector of lines and use a regex to find declarations of the simple types (int, double, char, float). Capture the comma-separated variable list, split it, then "clean" each item by removing initializers (= ...), pointer/ref markers (*/&) and array brackets ([...]) and store the variable name plus its declaration line index.

Second pass: for each collected variable scan the lines from the end toward the top and use a whole-word regex (e.g. "\bname\b") to find the last line that contains that identifier (optionally skip the declaration line). The first match encountered is the last usage; store that full line.

Example (concise C++ sketch):

// read all lines
std::vector<std::string> lines = read_all_lines("file.c");

// decl regex: captures the comma list after the type
std::regex decl(R"(\b(?:unsigned\s+)?(?:int|double|char|float)\b\s*([^;]+);)");

// extract names, clean them (remove =, *, &, [..])
// then for each name build word-pattern "\\b" + name + "\\b"
// scan lines from bottom up, skip declaration index, first regex match is last occurrence

Cautions and tips: strip comments and string literals before regexes (they create false positives). This approach works for simple, well-formed C code but will fail on edge cases: typedefs, macros, function parameters, class/struct members, templates or declarations that span multiple lines. For robust, production-grade analysis use a real parser/tokenizer (libclang/Clang AST, Tree-sitter, or a C lexer).

Recommended Answers

All 2 Replies

Member Avatar for NathanOliver

Show the code you have now and we should be able to help you modify it.

Member Avatar for rasha80

this is the code.. the code recivices file and earch for a word in a file if found it store the line in array.. what i want to do, is searching for a word if found i need to store the word next to the searhed word. e.g. if text file inclue the following text ( int x; int y; cout<<"hi";x=5; x=y+z;) when i searched for int word both x,y stored in array. then after that search for the last place where x and y appeared and store x=y+z. how i can do so??

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


int main()
{
    std::cout  << "Write the path of the file\n" ;
    std::string path ;
    std::cin >> path ;
    std::string line;
    std:: string names[50];
    std:: string output;
    std::ifstream file( path.c_str() ) ;

    if( file.is_open() )
    {
        std::cout << "File '" << path << "' opened.\n" ;

        std::cout << "Write the word you're searching for\n" ;
        std::string word ;
        std::cin >> word ;

        int countwords = 0 ;
        std::string candidate ;
        int i=0;
        while( file >> candidate ) // for each candidate word read from the file 
        {
        //  std::cout<< candidate<<"\n";
            if( word == candidate ) 
            { ++countwords ;
             file >> output;
             //getline (file,line);
             names[i]=output;
             i++;
            //  std::cout<<line;

            }
        }
           std:: cout <<"input size="<< (countwords * 32);
        std::cout << "The word '" << word << "' has been found " << countwords << " times.\n" ;
Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.