Std :: MAP first experience

/ *

This Exercise is in the field of bibliometric (Words and text) analyysis. You will be supplied with A Text

File, Which Consists of Several Paragraphs of English Text. Your Task is to Write a Program Which Will

Analyze the text, And Output A Range of Statistics About The Text.

Your Program Should Do The Following:

• A List of all the Words That Occur, in Alphabetical ORDER. You Must Not Print The Same Word Twice.

• Print Out The 20 MOST Common Words in The Text, TOGETHER WITH The Number of Instances of Each Word,

With the Most Common Word At the top.

* /

//

// readxt.cpp:.

// by mythma

// List the words in the article in the form of a standard alphabet, can not be repeated

/ / And 20 prints with the highest frequency

// compiler: g

// Cannot compile this procedure with VC6

//

#include

#include

#include

#include

#include

#include

Using namespace std;

Const string str_interval = "/ n ------------------------------------------- ---------------- ";

Map GMAP;

Vector Gmmstr;

Bool Addwordtolist (String Strword)

{

IF (strWord.empty ())

Return False;

String straTemp = strword;

// Upper to Lower

For (int i = 0; i

{

Strtemp [I] = TOLOWER (Strtemp [i]);

}

// Remove Head Puncture and Number

While (strTemp [0] <'a' || strTemp [0]> 'Z')

{

IF (strTemp.size ()> 1)

Strtemp = startemp.substr (1, strTemp.size () - 1);

Else

Return False;

}

// Remove Tail puncture and number

While (strTemp [stratemp.size () - 1] <'a' || strTemp [startemp.size () - 1]> 'z'

{

IF (strTemp.size ()> 1)

Strtemp = startemp.substr (0, strTemp.size () - 2);

Else

Return False;

}

Map :: item = gmap.find (strTemp);

// Add to map if exist

IF (it == gmap.end ())

GMap.Insert (Map :: value_type (strTemp, 1);

// increase if not exist

Else

(* it). Second;

Return True;

}

Void Outputwordslist ()

{

Cout << str_interval

<< "/ n-- The total number of words in the file is:"

<< gmap.size ()

<< ", arrangements according to the letters"

<< Str_Interval << endl;

INT n = 0;

For (Map :: item = gmap.begin (); it! = gmap.end (); IT)

{

N;

Cout.width (15);

Cout.Flags (ios :: left);

Cout << IT-> first.c_str ();

IF (5 == n)

{

Cout << Endl;

n = 0;

}

}

Cout << str_interval << endl;

}

Bool CMP (Const Pair & P1, Const Pair & P2)

{

Return p1.second> p2.second;

}

void outputcount ()

{

Vector > wd (gmap.begin (), gmap.end ());

Sort (wd.begin (), wd.end (), cmp);

Cout << str_interval

<< "/ n - There are several words with the most frequent frequency:"

<< Str_Interval;

INT i = 0;

For (Vector :: item = wd.begin ();

It! = wd.end () && i <20; IT, i)

{

Cout.width (15);

Cout.Flags (ios :: left);

Cout << Endl

<< IT-> first.c_str ()

<< "----"

<< IT-> second;

}

Cout << str_interval;

}

Int main (int Argc, char * argv [])

{

Char * strpath;

IF (argc == 2)

StrPath = argv [1];

Else

StrPath = "c: //words.txt";

// read file

IFStream Infile (STRPATH);

While (! infile.eof ())

{

String strword;

Infile >> strword;

AddWordtolist (strword);

Outputwordslist ();

Outputcount ();

Return 0;

}