Author: Jonas Bilinkevicius

How to determine whether a string contains binary data or normal readable text

Answer:

Solve 1:

Well, actually there is not really a difference between binary and not. There is 
just a common "agreement" that everything below ASCII code 32 is a "binary" 
character.

1   function IsReadable(st: string): Boolean;
2   var
3     i: Integer;
4   begin
5     Result := false;
6     for i := 1 to Length(st) do
7       if st[i] < #32 then
8         Exit;
9     Result := true;
10  end;



Solve 2:

11  procedure GetVeryBasicFileType(const Path: string;
12    var VeryBasicFileType: LongWord); overload;
13  var
14    Fs: TFileStream;
15    ExtCharCount, NothingCount, SpaceCount, LfCount, i, DidRead: Word;
16    LastB,
17      B: byte;
18    ReadBuffer: array[0..MAX_FILE_DETECT_BUFFER] of byte;
19  begin
20    VeryBasicFileType := VERY_BASIC_FILETYPE_UNKNOWN;
21    if (FileExists(Path)) then
22    begin
23      Fs := TFileStream.Create(Path, fmOpenRead);
24      try
25        if (Fs.Size >= 1) then
26        begin
27          LfCount := 0;
28          ExtCharCount := 0;
29          SpaceCount := 0;
30          NothingCount := 0;
31          LastB := 0;
32          Fs.Seek(0, soFromBeginning);
33          DidRead := Fs.read(ReadBuffer, SizeOf(ReadBuffer));
34          for i := 0 to DidRead - 2 do
35          begin
36            B := ReadBuffer[i];
37            if (B < 1) then
38              Inc(NothingCount)
39            else if (((B < 32) or (B > 127)) and not (B in [$0A, $0D, $21])) then
40              Inc(ExtCharCount)
41            else if ((B = $0A) or (B = $0D)) then
42              Inc(LfCount)
43            else if ((B = $21) and (LastB <> $21)) then
44              Inc(SpaceCount);
45            LastB := B;
46          end;
47          if (NothingCount > (DidRead div 30)) then
48          begin
49            if ((DidRead >= 4) and ((ReadBuffer[0] = $FF) and (ReadBuffer[1] = $FE)) 
50  or
51              ((ReadBuffer[0] = $FE) and (ReadBuffer[1] = $FF))) then
52            begin
53              VeryBasicFileType := VERY_BASIC_FILETYPE_TEXT_UNICODE;
54              Exit;
55            end
56            else
57            begin
58              VeryBasicFileType := VERY_BASIC_FILETYPE_BINARY;
59              Exit;
60            end;
61          end;
62          if (ExtCharCount > (DidRead div 3)) then
63          begin
64            if ((LfCount >= (DidRead div 60)) or (SpaceCount > (DidRead div 6))) then
65              VeryBasicFileType := VERY_BASIC_FILETYPE_TEXT
66            else
67              VeryBasicFileType := VERY_BASIC_FILETYPE_BINARY;
68          end
69          else
70          begin
71            if ((ReadBuffer[0] = $EF) and (ReadBuffer[1] = $BB) and
72  						 (ReadBuffer[2] = $BF) and (LfCount > (DidRead div 60))) then
73              VeryBasicFileType := VERY_BASIC_FILETYPE_TEXT_UTF8
74            else
75              VeryBasicFileType := VERY_BASIC_FILETYPE_TEXT;
76          end;
77        end
78        else
79          VeryBasicFileType := VERY_BASIC_FILETYPE_EMPTY;
80      finally
81        Fs.Free;
82      end;
83    end;
84  end;