mouse = file('/dev/input/mouse0')
while True:
    status, dx, dy = tuple(ord(c) for c in mouse.read(3))

    def to_signed(n):
        return n - ((0x80 & n) << 1)

    dx = to_signed(dx)
    dy = to_signed(dy)
    print "%#02x %d %d" % (status, dx, dy)

I hope this just made your day!

int ab=10;
int array[ab];

And surprisingly it didn’t work. (I mean it works in C++ and C#)
Anyway I was thinking of ways on how to do this thought of using malloc() however I really didn’t know how hence I did some Google’ing around. And found this neat piece of code.

int *resize_array(int *a, size_t new_size)
{
  int *save;

  save = realloc(a, new_size);
  if (save == NULL) {
    fprintf(stderr, "Memory exhausted\n");
    exit(EXIT_FAILURE);
  }
  return save;
}

int *user_old_array; // the array
int new_array_size=10;
user_old_array = malloc(initial_array_size * sizeof *user_old_array); //resized array

Quite neat isn’t it. I thought I should probably write this somewhere so I won’t forget. Hence the post.

Enjoy!

How to set up a CVS server using SSH :

Step 1 : Install CVS to Fedora. My Fedora distro came with CVS since I chose developer properties during installation. However you may not have it to install simply say;

sudo yum install cvs

Step 2 : Create a cvs user and group! Just follow the code below;

mkdir /home/cvs

Step 3 : Set permissions and add /home/cvs directory to cvs by using the following;

cvs -d /home/cvs init

chmod 700   /home/cvs/CVSROOT

Step 4 : Now you have to add your project to the CVS server. I assume you have a ~/project directory on the server. Just follow the following commands;

cd ~/project

cvs -d /home/cvs checkout project

Step 5 : Add an account to the CVS server (preferably with the same username and password of the user acount you have created for the CVS server that will connect through SSH.) Use the following by replacing <username> and <password>

echo "<username>:<password>" >> /home/cvs/CVSROOT/passwd

Step 6 : Make sure you have created a user to connect through SSH give him CVS group permission.

And that is that. It should work. Though you may need to do some changes in Step 3. To connect you can simply use the following code or just use Eclipse’s IDE ( that is what we are using. )

export CVS_RSH=ssh

cvs -d :ext:username@repository.host.name:/home/cvs co moduleName

I hope it will be helpful. Peace!

You see we learned that we should NOT set any environment variables while using demo u-boot image which was downloaded from here. In fact what we did is we created our own Ångström image (from Ångström site) and simply installed that to our Linux partition of our SDHC card. And it worked beautifully. Though I must say first boot is a drag, very very slow.

Ran to some problems again. It seems BeagleBoard only works with DVI-D LCD’s and we had a HDMI to VGA instead of a HDMI to DVI-D cable.We ended up connecting to it from SSH and installing our programs like that. We really don’t need an LCD any time soon. And to connect to it from SSH we had to learn it’s IP from terminal. So no problems there. And we were very happy with the outcome.

Installing python libraries; as installing python libraries go this one was a little tricky. You see it seems that Ångström doesn’t update its python libs that frequently. So calling the package by

opkg intall python-pycrypto

Didn’t quite work since installed itself to python2.5 directory. Had to install pycrypto manually. I hated every moment of it. I  than uploaded my own program and I was very happy with the outcome. The program worked perfectly.

Stay tuned for more updates on The Black Box project! Remember you can find it athttp://sourceforge.net/projects/theblackbox/ .

Live coding!