Алгоритм распознавания кодировки нужен для автоматического декодирования текста. Этот алгоритм основан на том, что некоторые буквы русского алфавита встречается очень часто, а некоторые редко. Поскольку этот способ статистический, то лучше всего он работает с большими текстами.
| Код | type
TCode = (win, koi, iso, dos);
const
CodeStrings: array [TCode] of String = ('win','koi','iso','dos');
procedure TForm1.Button1Click(Sender: TObject);
var
str: array [TCode] of string;
norm: array ['А'..'я'] of single;
code1, code2: TCode;
min1, min2: TCode;
count: array [char] of integer;
d, min: single;
s, so: string;
chars: array [char] of char;
c: char;
i: integer;
begin
so := Memo1.Text;
norm['А'] := 0.001;
norm['Б'] := 0;
norm['В'] := 0.002;
norm['Г'] := 0;
norm['Д'] := 0.001;
norm['Е'] := 0.001;
norm['Ж'] := 0;
norm['З'] := 0;
norm['И'] := 0.001;
norm['Й'] := 0;
norm['К'] := 0.001;
norm['Л'] := 0;
norm['М'] := 0.001;
norm['Н'] := 0.001;
norm['О'] := 0.001;
norm['П'] := 0.002;
norm['Р'] := 0.002;
norm['С'] := 0.001;
norm['Т'] := 0.001;
norm['У'] := 0;
norm['Ф'] := 0;
norm['Х'] := 0;
norm['Ц'] := 0;
norm['Ч'] := 0.001;
norm['Ш'] := 0.001;
norm['Щ'] := 0;
norm['Ъ'] := 0;
norm['Ы'] := 0;
norm['Ь'] := 0;
norm['Э'] := 0.001;
norm['Ю'] := 0;
norm['Я'] := 0;
norm['а'] := 0.057;
norm['б'] := 0.01;
norm['в'] := 0.031;
norm['г'] := 0.011;
norm['д'] := 0.021;
norm['е'] := 0.067;
norm['ж'] := 0.007;
norm['з'] := 0.013;
norm['и'] := 0.052;
norm['й'] := 0.011;
norm['к'] := 0.023;
norm['л'] := 0.03;
norm['м'] := 0.024;
norm['н'] := 0.043;
norm['о'] := 0.075;
norm['п'] := 0.026;
norm['р'] := 0.038;
norm['с'] := 0.034;
norm['т'] := 0.046;
norm['у'] := 0.016;
norm['ф'] := 0.001;
norm['х'] := 0.006;
norm['ц'] := 0.002;
norm['ч'] := 0.011;
norm['ш'] := 0.004;
norm['щ'] := 0.004;
norm['ъ'] := 0;
norm['ы'] := 0.012;
norm['ь'] := 0.012;
norm['э'] := 0.003;
norm['ю'] := 0.005;
norm['я'] := 0.015;
Str[win] := 'АаБбВвГгДдЕеЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя';
Str[koi] := 'юЮаАбБцЦдДеЕфФгГхХиИйЙкКлЛмМнНоОпПяЯрРсСтТуУжЖвВьЬыЫзЗшШэЭщЩчЧъЪ';
Str[iso] := 'РрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯяа№бёвђгѓдєеѕжізїијйљкњлћмќн§оўпџ';
Str[dos] := 'Ђ ЃЎ‚ўѓЈ„¤…Ґ†¦‡§ЂЁ‰©ЉЄ‹"Њ¬ЌЋ®ЏЇђа'б'в"г"де–ж—з?и™йљк›лњмќнћоџп';
for c := #0 to #255 do
Chars[c] := c;
min1 := win;
min2 := win;
min := 0;
s := so;
fillchar(count, sizeof(count), 0);
for i := 1 to Length(s) do
inc(count[s[i]]);
for c := 'А' to 'я' do
min := min + sqr(count[c] / Length(s) - norm[c]);
for code1 := low(TCode) to high(TCode) do begin
for code2 := low(TCode) to high(TCode) do begin
if code1 = code2 then continue;
s := so;
for i := 1 to Length(Str[win]) do
Chars[Str[code2][i]] := Str[code1][i];
for i := 1 to Length(s) do
s[i] := Chars[s[i]];
fillchar(count, sizeof(count), 0);
for i := 1 to Length(s) do
inc(count[s[i]]);
d := 0;
for c := 'А' to 'я' do
d := d + sqr(count[c] / Length(s) - norm[c]);
if d < min then begin
min1 := code1;
min2 := code2;
min := d;
end;
end;
end;
s := Memo1.Text;
if min1 <> min2 then begin
for c := #0 to #255 do
Chars[c] := c;
for i := 1 to Length(Str[win]) do
Chars[Str[min2][i]] := Str[min1][i];
for i := 1 to Length(s) do
s[i] := Chars[s[i]];
end;
Form1.Caption := CodeStrings[min2] + ' ' + CodeStrings[min1];
Memo2.Text := s;
end; |
|