help me plz with binary

graemef · 06-07-2006, 06:31 AM

Quote:

Originally Posted by dimah

Don't work!

Doesn't really help those trying to help you does it...

What doesn't work? Does it fail to compile? Does it run but fail to read? What doesn't work?

Here is the code segment from the Qt web site documentation for QT version 3.1

Code:

QFile f(fileName);
f.open(IO_ReadOnly);
char buffer[256];
int l = 256;
l=f.readBlock(buffer,l);

ppanyam · 06-07-2006, 10:44 AM

SEG-Y files are huge, in general. and you dont read them 4 bytes at a time. It is least efficient way of doing it. You have to read them in binary mode, so you open the file in binary mode.

Read 3200 bytes and assign it to a datastructure ccalled EBCDICHeader, then read 400 bytes and allocate it to BinaryHeader.
From Binaryheader, get length of seismic trace(No. Of samples * sample size + 240 bytes for traceheader). Keep reading this chunk of bytes until all the data is done.

You cant do it in a simple 5 line program. Large data structures are involved.

Panyam Prabhakar

dimah · 06-08-2006, 03:38 AM

Thanks a lot, but i really did't understand how to get normal data from this crazy symbols.
For example, this code (i think) must read 4 bytes into buffon;
So i see only that crazy delirium!

////////////////////////////////////////////////////
for (int i=0;i<1000;i++){
char buffon[1024];
int k=4;
l=f.readBlock(buffon,k);
cout << buffon << endl;}
////////////////////////////////////////////////////
What's wrong?

ppanyam · 06-08-2006, 11:01 AM

The data is in binary format. Thats it. You have to convert it to int or float or what-ever-it-is and then "cout" it. Unless ofcourse your data is ascii characters.

If you are trying to print first 3200 bytes of the SEG-Y file, they are characters alright, but they are in EBCDIC format. So you just cant print them.

If you want to print your EBCDIC header, try this

Code:

dd if=colda1.sgy of=colda1.ebc ibs=3200 count=1 conv=ascii

I am not really sure, I have understood your problem. You have to give more details and also explain what you are trying to achieve with your code.

ppanyam

dimah · 06-08-2006, 11:09 PM

ppanyam,
Am i right that after 3600 bytes there are the numerals?
And i need to find 21 byte.
After that i need to find last byte.
Of course i need to take the normal numerals (12 or 324) but not that abra-cadabra.
I think that's all i need to.

So my 21 byte will be 1st (conditionally) and i need to get the last numeral.
I home my English is enough to explain what i want.
Thanks for you participation.

ppanyam · 06-10-2006, 11:14 AM

Dear dilmah

I am on a vacation now. I will send a perl script to do it when I get back to work next week. You are right in assumption that there are numbers after 3600 bytes. I presume you are looking for inline or xline number. But I have not understood what you meant by

Quote:

and i need to get the last numeral.

ppanyam

ppanyam · 06-11-2006, 06:32 AM

I assume you are able to read your 4 bytes from 21 to 24 th locations. Then you can use this and try to print your number

/* read 4 bytes into char buffer[4]; */
fread(.....;
/* add this to your code */
union u{
char ch[4];
int lno;
}

union u myU;
strcpy(myU.ch,buffer);
printf("myInteger is %d.\n",myU.lno);

..

I havent run this, but it should work.

ppanyam

dimah · 06-15-2006, 11:28 PM

I ran it and got that number:
1077952576
What is that number?

ppanyam · 06-16-2006, 12:30 AM

Can you run the following command and send the output (zipped) to me by email?

dd if=yourdata.sgy of=topdata.dat ibs=64000 count=4

I have to look into the header before I can comment what that number is. Looks like there is a Byte order issue. Maybe not

ppanyam

ppanyam · 06-16-2006, 01:41 AM

Sorry... My email id is ppanyam@gmail.com