Reply to thread

Message

[QUOTE="Alf P. Steinbach, post: 3421617"]
* samoukos:

Please don't shout.

Note that ASCII does not define any greek characters: presumably what 
you mean is using an encoding with one byte per character.

Also, it doesn't seem like the program counts words, but character 
instances (e.g. number of occurences of 'A' in the text).

Indentation is a good idea.  E.g. add four spaces at the start of lines 
inside { and }.

These variables should most probably be declared locally where they're 
used.  It's a good idea to use consistent naming convention.  I.e. 
numberUsed or number_used, not NumberUsed which is unlike the others.

From the code below this is used to store the text.

A std::string is much more safe and convenient.

If sum[i] is the number of occurences of the character encoded as value 
i, then the above array can't count occurences of a character encoded as 
255, because valid indices range from 0 to 254, inclusive.

The assumption of no 255-characters may be valid, or not.

[i]

Level 1 correction: use symbolic names for "magic" numbers such as 
1000000, and use descriptive names such as "text" rather than "alpha".

Level 2 correction: use a locally declared loop variable,

for( int i = 0;  i < textBufferSize;  ++i )
   {
       text[i] = 0;
   }

Level 3 correction: do this initialization in the declaration,

size_t const textBufferSize = 1000000;

char text[textBufferSize] = {0};

Level 4 correction: use a std::string instead,

std::string text;

[i]

Level 1, 2 and 3 corrections as noted for 'alpha' above.

Rest snipped, I think the above comments enough for now.

Cheers & hth.,

- Alf[/i][/i][/i][/i]
[/QUOTE]

Verification