Read the contents of sequential Binary File in C#

Question

1.00/5 (1 vote)

See more:

Hi,

Can anyone help me in the foll. issue:
I need to read data in local variables from a binary file generated sequentially in linux. It is very much difficult for me to read the int's, float's and double values from the binary file. As they generate separate outputs when read as stream and read as sequential.

What I have tried so far:
I have tried BinaryFormatter to convert the binary file to readable stream, but it throws an exception of "The input stream is not a valid binary format. The starting contents (in bytes) are: 26-41-6B-73-68-61-79-20-20-20-20-20-20-20-20-20-20 ..."

I even tried BinaryReader, but only able to read strings/chars, but the numbers come different and have no relation the type declared.

The StreamReader is no good in this case.. This is my assumption as StreamReader will read a stream and not sequentially as I want.

Please help.. Thanks

Posted 24-Nov-13 19:51pm

Akshay 007

Add a Solution

Comments

Sergey Alexandrovich Kryukov 25-Nov-13 2:04am

"Even"?! You must have screwed up something. This is way too typical: blaming some classes for your own lack of skills.
—SA

Akshay 007 25-Nov-13 2:47am

Hi Sergey,

To make things clear I am not using BinaryWriter to create the binary file. As said earlier the file is created on linux with another tool. The method of access is used as sequential for both reading and writing. My task here involves that I need to read that binary file in C# and use these values for further tasks and write the output to another binary file.

The length of my binary file is 40 as per br.BaseStream.Length, the string that is present in the binary file is "Akshay" and was initialized for 30 characters space. The float value inserted is "1000.50" and the int value inserted is "1234567890". All the 3 values are written in same line with tab delimited. When I read it using BinaryReader and as you mentioned about reading the values, I used ReadChars(30) for reading the string of 30 characters, then I used ReadSingle() as an alternative for float as both represent 4 byte floating value, and laslty I used ReadInt32(() as ReadInt64() will go beyond the stream length.

Foll is my code for reading
<pre>
using (var br = new BinaryReader(new FileStream(filename, FileMode.Open, FileAccess.Read, FileShare.Read, 0x10000, FileOptions.SequentialScan)))
{
long pos, length;
length = br.BaseStream.Length;
pos = br.BaseStream.Position;

char[] name = br.ReadChars(30);
pos = br.BaseStream.Position;

float fVal = br.ReadSingle();
pos = br.BaseStream.Position;

Int64 intVal = br.ReadInt64();
pos = br.BaseStream.Position;

}
</pre>
Please help me in understanding where I am making the mistake.

Sergey Alexandrovich Kryukov 25-Nov-13 3:07am

So what? I already answered. Want to write as well? Use BinaryWriter, in a way symmetrical to reading, method-by-method.
Look, do you have fixed format? Can you change it? If you can, just write read and write at the same time, side by side.
Now, why binary? Maybe, if it seems difficult for you (but why?), you could use XML? Them you won't need to read/write anything directly, not at all; you could use DataContract... Or use serialization with BinaryFormatter.
—SA

Akshay 007 25-Nov-13 3:50am

There are some applications that are made keeping Linux in mind, and these applications use binary format as their input. I don't think you are concentrating on the part where I say while writing the binary file the access mode set is sequential. It is similar to the access="SEQUENTIAL" in Fortran.
I will be using Binary Writer to write the output of my tasks, but it has to be in the same format as is the reader.

Sergey Alexandrovich Kryukov 25-Nov-13 3:58am

I don't have to "concentrate" on anything anymore, just because I already have you exact recipe of reading any binary file, no matter what it is. I only added some alternative just in case. What else is unclear?
—SA

Akshay 007 25-Nov-13 20:53pm

Why are you so adamant. If you are a master than download the binary file from https://www.dropbox.com/s/bclgkybavbxpey9/swiss.bin
and try it yourself. Let me also know what different you are doing.
1st value is "Mike Tyson", 2nd value is "1234567890", 3rd value is "1000.50" and last value is "Mike Tyson"

Sergey Alexandrovich Kryukov 25-Nov-13 21:16pm

No, first value if '&', and it cannot be a length of "Mike Tyson". And this name is followed by a string of blank spaces, not digits. Of course, if you have wrong information of file format, you cannot read it as easily. But it does not worth doing. Is someone gives you false information (lies, but not always, it could be a honest mistake), why considering it.

You need to know value types, not values. When format is known (in this case, this is just the set of types of fields), it's not just easy, it's pathetically easy. Or you need to have the code which writes it... But you are not telling what is there.

Enough, let's stop this useless talk...

—SA

Akshay 007 25-Nov-13 21:34pm

Sergey,
The Data Types are CHARACTER of length 30, INTERGER (4 byte), FLOAT (4 byte) and CHARACTER of length 30.
The character is similar to FORTRAN 77, where if assign the variable with length 30 then 30 characters will be printed, and the values at indexes can be alphanumeric or blanks. The key here is that even blanks will be printed.

Sergey Alexandrovich Kryukov 26-Nov-13 0:33am

Not exactly. Both character sequences are prefixed with '&', second one starts with &.Myke Tyson.
Here is how:


using System.IO;
//...
using (BinaryReader reader = new BinaryReader(File.Open(fileName, FileMode.Open))) {
    byte prefix = reader.ReadByte();  '&'
    byte[] charData1 = new char[30];
    reader.Read(charData1, 30, 30); // of course, don't hardcode 30
    // to get string, do something like:
    string someString = new string(System.Text.Encoding.Ascii.GetChars(charData1));
    int int = reader.ReadInt32();
    single r = reader.ReadSingle();
    prefix = reader.ReadByte(); // '&'
    byte[] charData2 = new char[30];
    //... same as above
    //...
} // at this point, reader is closed automatically

That's it. Did not test; it's easy to debug, in no time, looking at what's expected, in some binary editor...

—SA

Sergey Alexandrovich Kryukov 25-Nov-13 3:09am

Don't create a reader from stream, create it from file. Pos is useless, you never use it, same thing about length.
Just read...
—SA

2 solutions

Solution 2

"All the 3 values are written in same line with tab delimited"
And where do you read the tabs? Nowhere. After every br.ReadSomething() call you must advance the position by an additional byte. A br.ReadByte() could help. - Do not do that after the last read operation.
And just calculate:
30 (for the string) + 1 (for the first tab) + 4 (for the float) + 1 (for the second tab) + 4 (for the int) = 40 bytes.

Posted 25-Nov-13 0:33am

Bernhard Hiller

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Sergey Alexandrovich Kryukov · Accepted Answer · 2013-11-24T20:08:00

You really need to know all the detail of the file format. Reading is quite simple:
http://msdn.microsoft.com/en-us/library/system.io.binaryreader.readdouble(v=vs.110).aspx[^],
http://msdn.microsoft.com/en-us/library/system.io.binaryreader.readsingle(v=vs.110).aspx[^],
http://msdn.microsoft.com/en-us/library/system.io.binaryreader%28v=vs.110%29.aspx[^].

If you make a tiny mistake in an offset of each numeric or any other field, you start to read between fields and screw up everything after such shift. So, you need to be accurate.

—SA