Lesson of 14 July 2008

Use of map and vector of STL

Read a file and count the words

File wordcount/main.cc

:open:
 1/*
 2  A simple program that count words from a file
 3  and print with some statistics.
 4*/
 5
 6#include <iostream>
 7#include <fstream>
 8#include <cmath>
 9
10// from STL
11#include <string>
12#include <map>
13
14/*
15  > main pippo pluto
16
17  argc = 3
18  argv[0] = "main"
19  argv[1] = "pippo"
20  argv[2] = "pluto"
21  
22  Usage of the program
23  
24  > main filename
25  
26 */
27 
28using namespace std ;
29
30int
31main( int argc, char * argv[] ) {
32  string fileName ;
33
34  if ( argc > 2 ) {
35    // too nuch argument, issue an error
36    cerr << "usage: " << argv[0] << " file\n" ;
37    return 1 ;
38  }
39
40  if ( argc == 1 ) {
41    // only one argument, ask to the user the file name
42    cout << "File: " ;
43    cin >> fileName ; 
44  } else {
45    // two argument, get file name from command line
46    fileName = argv[1] ;
47  }
48  
49  // open the file
50  ifstream file( fileName . c_str() ) ;
51  
52  // check if the file exist is opened etc.
53  if ( file . bad() ) {
54    cerr << "error in openeing : " << fileName << "\n" ;    
55    return 1 ; 
56  }
57  
58  map<string,int> wc ; // associative array to store word count
59  
60  // loop to read all the words
61  while ( file . good() ) {
62    string w ;
63    file >> w ;
64    
65    // check if w is present in the associative array wc
66    map<string,int>::iterator i = wc.find(w) ;
67    if ( i == wc.end() ) {
68      // element not found
69      wc[w] = 1 ;
70    } else {
71      // element found!
72      // wc[w]++ ;
73      // i->first  keyword
74      // i->second value
75      i->second++ ;
76    }
77  }
78
79  // close the file
80  file . close() ;
81  
82  // write the output
83  for ( map<string,int>::const_iterator i = wc.begin() ; i != wc.end() ; ++i ) {
84    cout << i -> first << " " << i -> second << "\n" ;
85  }
86}

Read a file and count the words, resort the count and print only the most frequents

Filewordcount2/main.cc

:open:
  1/*
  2  A simple program that count word from a file
  3  and print with some statistics.
  4*/
  5
  6#include <iostream>
  7#include <iomanip>
  8#include <fstream>
  9#include <cmath>
 10#include <cctype>
 11
 12// from STL
 13#include <string>
 14#include <vector>
 15#include <map>
 16
 17/*
 18  > main pippo pluto
 19
 20  argc = 3
 21  argv[0] = "main"
 22  argv[1] = "pippo"
 23  argv[2] = "pluto"
 24  
 25  Usage of the program
 26  
 27  > main filename
 28  
 29 */
 30 
 31 
 32using namespace std ;
 33
 34
 35// comparator for sort (ascending order)
 36typedef struct less_than {
 37  //! return always true
 38  bool operator () (pair<string,int> pa, pair<string,int> pb) const {
 39    if ( pa . second < pb . second ) return true ;
 40    if ( pa . second == pb . second ) {
 41      return pa . first < pb . first ;
 42    }
 43    return false ;
 44  }
 45} ;
 46
 47// comparator for sort (sescending order)
 48typedef struct greater_than {
 49  //! return always true
 50  bool operator () (pair<string,int> pa, pair<string,int> pb) const {
 51    if ( pa . second > pb . second ) return true ;
 52    if ( pa . second == pb . second ) {
 53      return pa . first > pb . first ;
 54    }
 55    return false ;
 56  }
 57} ;
 58
 59less_than    lt ;
 60greater_than gt ;
 61
 62int
 63main( int argc, char * argv[] ) {
 64  string fileName ;
 65
 66  if ( argc > 2 ) {
 67    // too much argument, issue an error
 68    cerr << "usage: " << argv[0] << " file\n" ;
 69    return 1 ;
 70  }
 71
 72  if ( argc == 1 ) {
 73    // only one argument, ask to the user the file name
 74    cout << "File: " ;
 75    cin >> fileName ; 
 76  } else {
 77    // two argument, get file name from command line
 78    fileName = argv[1] ;
 79  }
 80  
 81  // open the file
 82  ifstream file( fileName . c_str() ) ;
 83  
 84  // check if the file exist is opened etc.
 85  if ( file . bad() ) {
 86    cerr << "error in openeing : " << fileName << "\n" ;    
 87    return 1 ; 
 88  }
 89  
 90  map<string,int> wc ; // word count
 91  
 92  // loop to read all the words
 93  while ( file . good() ) {
 94    string w1, w ;
 95    file >> w1 ;
 96    w = "" ;
 97    for ( string::iterator i = w1.begin() ; i != w1.end() ; ++i ) {
 98      // string::value_type is the type of the element of the string
 99      // normally string::value_type == char
100      string::value_type c = tolower(*i) ;
101      // consider only letters
102      if ( isalpha(c) ) w . push_back(c) ;
103    }
104    
105    // check if w is present in the associative array wc
106    map<string,int>::iterator i = wc.find(w) ;
107    if ( i == wc.end() ) {
108      // element not found, create and initailize to 1
109      // only if lenght > 0 
110      if ( w . length() > 0 ) wc[w] = 1 ;
111    } else {
112      // element found!
113      i->second++ ;
114    }
115    
116  }
117
118  // close the file
119  file . close() ;
120  
121  // define a vector to store the results
122  vector<pair<string,int> > vec ;
123  copy( wc.begin(), wc.end(), back_inserter(vec) ) ;
124
125  // the code is equivalent to 
126  // for ( map<string,int>::const_iterator i = wc.begin() ;
127  //       i != wc . end() ; ++i )
128  //   vec . push_back(*i) ;
129  //
130  // the "back_inserter" function return an object which implement
131  // the operator ++ and =. ++ do nothing while
132  // back_inserter(vec) = rhs
133  // is transformed to vec.push_back(rhs)!.
134
135  // sort the vector
136  sort( vec.begin(), vec.end(), gt );
137
138  // write the output
139  int nmax = 10 ;
140  for ( vector<pair<string,int> >::const_iterator i = vec.begin() ;
141        i != vec.end() && nmax-- > 0 ; ++i ) {
142    cout << "word = " << setw(10) << left << i -> first
143         << " counted N. " << i -> second << " times\n" ;
144  }
145}

The lesson in a zip file

The zip file lesson6.zip