jumping from svn to git

If you are comfortable with svn but want to switch to git for a variety of reasons, this post is for you!
The major difference is that git has distributed repositories whereas svn has a central repository.So, instead of checking out a working copy from the central repo,in git we clone the repo to our local machine. The immediate advantage is that even when you are disconnected from the network, you can keep committing your changes to the local repo. When you get back the connectivity, just push the changes in the local repo to the remote repo.

This is a feature I missed a lot in svn.(while working in train/flight where connectivity is not available). Below is an illustration of a typical work flow:

1. Cloning the remote repo(~= cheking out a copy in svn) : for the first time
   git clone repo_url
2. make changes
3. get changes of others from the remote repo(~= svn update)
   git pull
4. commit your changes to the local repo
   git commit -a
5. commit the changes in the local repo to the remote repo
   git push

Steps 4 and 5 are roughly equivalent to svn commit

replacing text using regular expressions

Regular expressions can be a very powerful tool for searching and replacing. Today I learnt how to replace text using regular expressions. The problem I faced was:
I got these 4 equations from the internet: Basically they give you the 4 parameters for a plane equation Ax+By+Cz=D passing thru 3 points (x1,y1,z1),(x2,y2,z2),(x3,y3,z3)

A = y1 (z2 - z3) + y2 (z3 - z1) + y3 (z1 - z2)
B = z1 (x2 - x3) + z2 (x3 - x1) + z3 (x1 - x2)
C = x1 (y2 - y3) + x2 (y3 - y1) + x3 (y1 - y2)
D = x1 (y2 z3 - y3 z2) + x2 (y3 z1 - y1 z3) + x3 (y1 z2 - y2 z1)

I had to use these equations in my code. But my data structures were quite different. The 3 points were in an array of 3 points having fields x,y,z.

So the required replacements were as like:
x1 with poly->points[0].x
y2 with poly->points[1].y
z3 with poly->points[2].z

so on...(total 9 pairs!)
Also, replace the whitespace by * at appropriate places:

Smart soln:
Open the replace dialog in your editor. Select the "regular expressions" options.
I used netbeans but I think most of the good editors in linux support regular expressions

in find, enter: ([x-z])([1-3])
in replace, enter: poly->points[$2-1].$1
($1 stands for the text matching the contents of the 1st parenthesis in the find regexp. and so on..)

open the replace dialog again(to replace whitespace by * at appropriate places)
in find, enter: ([x-z])(\s)(\()
in replace, enter: $1*$3
This one needs some explanation. \s stands for whitespace. ( needs to be escaped because ( is also a token for regular expression defination syntax.

to replace whitespaces in eqation D, do:
in find, enter: ([x-z])(\s)(p)
in replace, enter: $1*$3

the results:

A = poly->points[1-1].y*(poly->points[2-1].z - poly->points[3-1].z) + poly->points[2-1].y*(poly->points[3-1].z - poly->points[1-1].z) + poly->points[3-1].y*(poly->points[1-1].z - poly->points[2-1].z);

B = poly->points[1-1].z*(poly->points[2-1].x - poly->points[3-1].x) + poly->points[2-1].z*(poly->points[3-1].x - poly->points[1-1].x) + poly->points[3-1].z*(poly->points[1-1].x - poly->points[2-1].x);

C = poly->points[1-1].x*(poly->points[2-1].y - poly->points[3-1].y) + poly->points[2-1].x*(poly->points[3-1].y - poly->points[1-1].y) + poly->points[3-1].x*(poly->points[1-1].y - poly->points[2-1].y);

D = poly->points[1-1].x*(poly->points[2-1].y*poly->points[3-1].z - poly->points[3-1].y*poly->points[2-1].z) + poly->points[2-1].x*(poly->points[3-1].y*poly->points[1-1].z - poly->points[1-1].y*poly->points[3-1].z) + poly->points[3-1].x*(poly->points[1-1].y*poly->points[2-1].z - poly->points[2-1].y*poly->points[1-1].z);


(People might say that writing 1 as 2-1 is inefficient but I'm sure that the c compiler does constant folding while generating the target code)