Primers Coming Up

Thru my work I'm thrown into alot of technologies that i dont nearly know enough about and as with alot of tech related things, the education scene is basic basic basic..GURU with little or no gradiation, so what I'm going to do is post what i learn when i learn it and where i learn it from and hopefully it'll be useful for someone else, and I'll also take the opportunity to rehash stuff I've already done.

ATM I'll probably be doing Bash scripting, Perl Scripting, XML, and whatever UNIX stuff comes up whenever I'm writing, but for now and for a relativly simple start; X display fowarding...

This is the setup: Linux/Unix based "client" and "server"; in my case I have headless systems that i fiddle with from time to time, but after a while vim just becomes a pain, and as for viewing html files etcetc copying things back and forth is a pain in the ass.

The solution is already there; From its beginning the X server has always been client server based on some level. Basically what were gonna do is tell the server to use the client as an X display; this is controlled thru an environment variable, strangly called DISPLAY

Lets assume were using SSH. *nix SSHd has a simple system for fowarding and routing the relevent X ports (these wont show up in your "tunnels" tab if your using something like putty, but you shouldnt be using putty anyway)

Theres two versions of this fowarding ability, X and Y. in a nutshell, X is compressed and encrypted. Y isnt. On slow links, Y is probably your best bet, also some servers dont support the X flag. Also also, some sysadms disable this X fowarding.

Anyway, to the point. 

login to the server thusly

client# ssh user@server -X

then when you get a shell check to see if DISPLAY isnt already set (some servers are good enough to do this for you)

server# echo $DISPLAY

if your lucky it'll say localhost:10.0

otherwise it'll probably say :0.0 or something similar

If your in this unlucky situation, just enter

server# DISPLAY=localhost:10:0

then try it out, by running the clock and forking it into the background (&)

server# xclock &

Job done. Later

Another Uni Project

If anyone is interested in Erlang B Calculations, very relevent to any communications or engineering students, I've written a little quick piece of code to calculate them.

There are several levels of functionality in the code.
Erlang B itself only has 2 variables, System load in Erlangs, and the number of "trunks" (read: servers/call center operators/phone lines), and its output is a blocking probability from 0 to 1

All three of these variables or none atall can be defined at runtime;

• The desired blocking probability can be input to stop the calculation at that point. (default 0)
  
• The Load can be defined (See Erlang A) (default 1)
• The maximum trunks to be calculated (default 100)

The code uses the unistd.h library for argument parsing so is more or less unix only (or cygwin alternativly) and long doubles for more or less everything inside the code.

Having tested the limits, it kinda conks out then calculating large (read 1000 erlangs) on large trunks (got as far as 1234 trunks, then died)

When i get a bit of time i might optimise the factorial part so it doesnt run thru the entire factorial sequence for each number.

Anyway, the code is here. I'm not wasting my time laying out code on blogger.

Embedded C GPS Project

Afternoon folks, I'm supposed to be studying but dont have the heart to, so I'm documenting a recent project from Uni.

The remit was to be able to parse RS232 data coming in from a GPS unit and reformat it for a LCD display. I dont have the part numbers handy but I was programming on a 18F series PIC that supported C.

Most of the ancillary code is more platform dependant, such as working with the PIC interrupts etc, so for the purposes of this code snippit, assumme that a NMEA sentence (I used RMC and some RMB, but never really finished that bit) stored as a character buffer, and a structure, as defined, to store relevent data in.

PLEASE read up about NMEA sentence structure before continuing

typedef struct message
{
//date
int day, month, year;

//Time
int hour, min, sec;

//lat
int lat_deg;
float lat_min;
char lat_ref;

//lng
int lng_deg;
float lng_min;
char lng_ref;

}message;

Since NMEA sentences are comma separated values, I kinda cheated and iterated thru the string, replacing the commas with terminating characters and recording the next positions as character pointers.

for(i=0;i<BUFFERSIZE;i++){
if(end)buffer[i]=0; //wipe the rest of the sentence

if(buffer[i]==42&&check==1){ //asterix
check=0;
if(checksum==chr2hex(&buffer[i+1])) valid=1;
else valid=0;

i+=2; //get to the end of the checksum

end=1; //this is the end of the current sentence
}
if(check)checksum^=buffer[i];
if(buffer[i]==36){ //dollarsign

buffer[i]=0;

check=1;

AddToList(&(buffer[i+1])); //ignore first character

}
if(buffer[i]==44){
buffer[i]=0; //replace all commas with nulls
AddToList(&(buffer[i+1])); //add next position to list of words
}
}


The first character(dollar sign) is ignored because it is never needed beyond this point.

The NMEA sentence structure includes a asterix delimited checksum, i.e everything after the dollarsign and before the asterix is progressivly XOR'd and the hex value representation of this result is concatenated on the end of the sentence before transmission.

the chr2 hex function simply converts two ASCII characters to their Hex value equivalent.

AddToList, strangly enough, adds the pointer passed to it to a wordlist, which is an array of character pointers.

Now we have a list of pointers, because i used a character pointer array, and ended all the strings with nulls, we essentially now have individual strings for each part of the NMEA sentence, that can be addressed directly. eg:
{ //GPRMC
getTime(words[1],&incoming);
getDate(words[9],&incoming);
getLat(words[3],&incoming,*words[3+1]);
getLng(words[5],&incoming,*words[5+1]);
}

In this instance, the get functions all take two arguments, what to read from, and where to put it.

For anyone whos interested the full code is here

I guess i better do some work then.