AnsweredAssumed Answered

Problem with BINBLOCK

Question asked by RobertLefevre on Sep 3, 2008
Latest reply on Dec 28, 2010 by odanzy

Like • Show 0 Likes0
Comment • 10

We are currently saving a certain file type as a BINBLOCK in VEE.
It is read out by writing a real32 BINBLOCK directly out to a text file.

My problem is that I need to parse this data in C# to an array. I need to know how the BINBLOCK file is constructed.

From what I gather, the first digit is an ascii "#" and the second digit is an ascii number (x) which is to read the next x ascii digits which correspond to a number in some format that I do not know, either what it represents or what format (int32...etc...).

The files are then read in by reading each 4 bytes ( corresponding to a float, since it is a real32 format) and converting them to a float.

Could anybody shed some light as to what is going on with the BINBLOCK format? Any information above and beyond what the VEE help file says will be appreciated.

No one else has this question

Outcomes

10 Replies

daras Sep 3, 2008 3:32 PM
Mark Correct
Correct Answer
You might want to post this to Agilent VRF ( www.agilent.com/find/vrf ) since that is a user group dedicated to VEE questions, but I think you might also want to consider using the VEE Callable Server in C#. VEE Callable Server provides a COM interface to a VEE library which can then be wrapped as an interop assembly to allow calls from .NET languages. So essentially you will write a VEE function in a VEE library to read the binblock file and convert it to a data type that C# can understand and then you will call that function from C#.
Like • Show 0 Likes0
Actions
odanzy Sep 3, 2008 5:47 PM
Mark Correct
Correct Answer
One thing that you need to understand is the byte ordering that the data is saved in. The default in VEE is MSB as this is the default for instruments. This means that if you are reading the on a PC (typically LSB) in C# you will need to swap the byte order. The other option is the change the BINBLOCK byte ordering to LSB. The user must also understand the data type (e.g. float, double, integer, long, etc...).

1. Strip off # character
2. Get length of ASCII string that represents the length of the data (1 character)
3. Read the size of the data block by reading back the number of ASCII characters obtained in step 2
4. Read back the data in chunks that corresponds to the data type in the file.

Note: be sure that you open the file for binary read. Using the BinaryReader class to read binary data from a file opened via a FileStream instance might be one way implement this.
Like • Show 0 Likes0
Actions
RobertLefevre Sep 4, 2008 5:46 AM
Mark Correct
Correct Answer
I was able to complete this as follows:

public static float[] getBinaryList(string File_)
        {
            //Declares
            List<float> toReturn = new List<float>();

            if(File.Exists(File_)) //Check for file existence
            {
                BinaryReader binReader = //Open new BINARY READER
                    new BinaryReader(File.Open(File_, FileMode.Open));

                try
                {
                    //Strip Leading "#"
                    binReader.ReadChar();

                    //Read first digit to get size of rest
                    char[] sz = binReader.ReadChars(1);
                    string s = new string(sz);
                    int size = Int32.Parse(s);

                    //Read next digits to get size of string
                    char[] len = binReader.ReadChars(size);
                    s = new string(len);
                    int length = Int32.Parse(s)/4;

                    //Convert rest of file to a REAL32 number
                    //Using a FLOAT variable
                    for (int x = 0; x < length; x++)
                    {toReturn.Add(binReader.ReadSingle());}
                }

            catch(Exception ex) //Catch Any Thrown Exceptions
            {Console.WriteLine(ex.ToString());}

            finally //Close Stream
            {binReader.Close();}

                //Return array of float
                return toReturn.ToArray();

          }//End if File Exists
                return null; //If error, return null
        }//End Function
Like • Show 0 Likes0
Actions
diefish Dec 23, 2010 12:30 AM
Mark Correct
Correct Answer
I have the same problem now. I use VC++ to control E5062A by VISA, i want to transfer file from instrument to PC,first, i saved a screen in the NA:
viPrintf(vi, ":MMEM:STOR:IMAG \"D:/Image01.bmp\"\n");
then, catch data by:
viPrintf(vi, "%s\n", ":MMEM:TRAN? \"D:/Image01.bmp\"");
in the manual, i know that ":MMEM:TRAN? <filename>" will return a binary data block, but i didn't know how to write viScanf argument?
i had read some example in VEE and VB, but i don't know how to deal with it in VC++
can anybody help me?
thank you!
Like • Show 0 Likes0
Actions
odanzy Dec 23, 2010 10:34 AM
Mark Correct
Correct Answer
Below is the code that I've used with the PNA, it should work for the ENA as well. There is a viScanf modifier that will directly read it back, but this method works with some older instruments as well that don't support the official binary block format.

     //////////////////////////////////////////////////
     // FILE READ FROM PNA BEGINS                    //
     //////////////////////////////////////////////////

     printf("Read File from PNA Begins\n");

     //Ask for file back
     status=viPrintf(vi, "MMEM:TRAN? \"C:\\PNATempImage.png\"\n");

     //Read back # character
     status=viRead(vi, buf, 1, &rtncnt);

     //Read back size of data length
     status=viRead(vi, buf, 1, &rtncnt);
     buf[rtncnt]='\0';

     //Convert size to int
     int size=atoi((const char *)buf);

     //Read back data length
     status=viRead(vi, buf, size, &rtncnt);
     buf[rtncnt]='\0';

     //Convert length to long
     long len=atol((const char *)buf);

     //Allocate data size based on length
     data=(ViPBuf)malloc(len);

     //Read back data
     status=viRead(vi, data, len+1, &rtncnt);
     data[rtncnt]='\0';

     //Flush read buffer
     status=viFlush(vi, VI_READ_BUF);

     //Open File for Binary Write
     fs.open("C:\\temp\\instr_screen.png", ios::out | ios::binary);

     //Write Data
     fs.write((const char *)data, len);

     //Close file
     fs.close();

     printf("Read File from PNA Complete\n");

     //////////////////////////////////////////////////
     // FILE READ FROM PNA COMPLETE                  //
     //////////////////////////////////////////////////
Like • Show 0 Likes0
Actions
diefish Dec 23, 2010 6:08 PM
Mark Correct
Correct Answer
Thank you for Odanzy's reply, i will have a try later.
Merry Christmas!
Like • Show 0 Likes0
Actions
diefish Dec 27, 2010 6:52 PM
Mark Correct
Correct Answer
it work very well, thank you odanzy!
and, i have a try viScanf like this:
     viPrintf(vi, "%s\n", ":MMEM:STOR:IMAG \"D:/Image01.bmp\"");
     viPrintf(vi, "%s\n", ":MMEM:TRAN? \"D:/Image01.bmp\"");
     unsigned char c;
     // Read back '#'
     viScanf(vi, "%c", &c);

     // Read back size of data length
     viScanf(vi, "%c", &c);

     // Convert chat to int
     int cnt = c-'0';

     // Read back length of data
     long num = 0;
     for( int i = 0; i < cnt; i++)
     {
          viScanf(vi, "%c", &c);
          int temp = c - '0';
          num = num*10 + temp;
     }

     //Allocate data size based on length
     unsigned char *ime = new unsigned char[num];

     // Read back data
     for(long i = 0; i < num; i++)
     {
          viScanf(vi, "%c", &c);
          ime[i] = c;
     }

     // Save data to file
     CFile im("ddot.bmp", CFile::modeCreate | CFile::modeReadWrite );
     im.Write(ime, num);
     im.Close();

it can work also, but very slowly. Is there any way to catch data in one time by viScanf?
if i read back trace data in real, i can write viScanf as: viScanf( vi, "%#zb", &rtncnt, floatArray ), how can i catch file( image, snp and so on) in this format( viScanf)? i mean that which modifier and format code i can use? the second argument is a float array pointer?
thank you & best wishes!
Like • Show 0 Likes0
Actions
odanzy Dec 27, 2010 9:20 PM
Mark Correct
Correct Answer
You should be able to replace the section of code in my original post with the code below.

     //////////////////////////////////////////////////
     // FILE READ FROM PNA BEGINS                    //
     //////////////////////////////////////////////////

     printf("Read File from PNA Begins\n");

     //Define Maximum file size to be read back
     long maxSize = 1e6;

     //Allocate array to hold data
     data=(ViPBuf)malloc(maxSize);

     //Give maximum size allowed to read back. Set this to the same value as data array size
     len=maxSize;

     //Ask for file back
     status=viPrintf(vi, "MMEM:TRAN? \"C:\\PNATempImage.png\"\n");

     status=viScanf(vi, "%#b\n", &len, data);

     //Open File for Binary Write
     fs.open("C:\\temp\\instr_screen.png", ios::out | ios::binary);

     //Write Data
     fs.write((const char *)data, len);

     //Close file
     fs.close();

     printf("Read File from PNA Complete\n");

     delete(data);

     //////////////////////////////////////////////////
     // FILE READ FROM PNA COMPLETE                  //
     //////////////////////////////////////////////////
Like • Show 0 Likes0
Actions
diefish Dec 27, 2010 10:08 PM
Mark Correct
Correct Answer
Thank you for odanzy's relay, it work very well.
It just need to deal with array size, it also work very well that i change array type from ViPBuf to float. I think the key is modifier and format code which i didn't very understand .
Thank you & Best wishes!
Like • Show 0 Likes0
Actions
odanzy Dec 28, 2010 10:15 AM
Mark Correct
Correct Answer
The issue with not knowing the array size to allocate is one reason to break the block down in code instead of using the special format modifiers. With reading trace data it is easy to determine exactly how many bites are coming back, but not as simple with files.
Like • Show 0 Likes0
Actions

Problem with BINBLOCK

Outcomes

Related Content

Recommended Content