Question About Cat And Grep

George2 · 03-08-2006, 06:25 AM

Hello everyone,

Could I use cat and grep commands to do the following task? If it is possible, could anyone help to show the sample please?

Task:

Given a file, calculate the number of lines (in this file) which is not started with '#'.

thanks in advance,
George

macemoneta · 03-08-2006, 06:31 AM

grep -v "^#" somefile | wc -l

PS- Try to do your own homework - this isn;t going to help you pass any tests.

jerril · 03-08-2006, 06:46 AM

Hi George

Try this:
$ grep -c "^[^#]" foo

jer

cs-cam · 03-08-2006, 08:12 AM

So you identified to yourself that the question was for a school-related task and you still answered it? Forgive my ignorance, but why..?

macemoneta · 03-08-2006, 08:42 AM

I've been watching the forums, and while I initially avoided answering those that appeared to be obvious school work I noted that someone always answers, and that occassionally incorrect responses will be posted.

I know that for me, "unlearning" is very hard. At least if one of the responses is correct, then the forum response becomes a multiple choice quiz - the OP will probably have to test each to see which is correct.

For example, there are some side effects to Jerril's response... will the OP discover them? Will the OP figure out how to simplify my response by combining it with Jerril's response? By providing a working (but perhaps not optimal) response, it becomes a basis for learning.

We can't stop people from answering these questions, but at least we can nag the OP into doing the right thing and provide guidance.

When I started in computer science in the 70's, I received a lot of help from others. Later I tutored other students. Part of providing the answer (to me) is to point the questioner on the path to improving on the answer. If the OP discovers the better, simpler way, then they've really learned. If they just copy what they get, they probably won't be continuing in this field anyway.

pixellany · 03-08-2006, 09:38 AM

OP does not fit the pattern of student asking us to do homework. Most of them make one post (thread) and are never seen again.

I don't mind the short questions--it's when they cut and paste the entire assignment (test) into a post that I get annoyed.

George2 · 03-09-2006, 02:43 AM

cs-cam,

Quote:

Originally Posted by cs-cam

So you identified to yourself that the question was for a school-related task and you still answered it? Forgive my ignorance, but why..?

It is not. It is my personal interest to this topic.

regards,
George

George2 · 03-09-2006, 02:45 AM

pixellany,

Quote:

Originally Posted by pixellany

OP does not fit the pattern of student asking us to do homework. Most of them make one post (thread) and are never seen again.

I don't mind the short questions--it's when they cut and paste the entire assignment (test) into a post that I get annoyed.

Many thanks. I used C program to do such simple tasks before. I simply want to know whether there are approaches to improve it -- make things more simple.

regards,
George

George2 · 03-09-2006, 03:29 AM

Thanks macemoneta,

Quote:

Originally Posted by macemoneta

grep -v "^#" somefile | wc -l

PS- Try to do your own homework - this isn;t going to help you pass any tests.

I have tried your method. It works! Cool!

regards,
George

George2 · 03-09-2006, 03:30 AM

Thank you very much, macemoneta!

Quote:

Originally Posted by macemoneta

I've been watching the forums, and while I initially avoided answering those that appeared to be obvious school work I noted that someone always answers, and that occassionally incorrect responses will be posted.

I know that for me, "unlearning" is very hard. At least if one of the responses is correct, then the forum response becomes a multiple choice quiz - the OP will probably have to test each to see which is correct.

For example, there are some side effects to Jerril's response... will the OP discover them? Will the OP figure out how to simplify my response by combining it with Jerril's response? By providing a working (but perhaps not optimal) response, it becomes a basis for learning.

We can't stop people from answering these questions, but at least we can nag the OP into doing the right thing and provide guidance.

When I started in computer science in the 70's, I received a lot of help from others. Later I tutored other students. Part of providing the answer (to me) is to point the questioner on the path to improving on the answer. If the OP discovers the better, simpler way, then they've really learned. If they just copy what they get, they probably won't be continuing in this field anyway.

It is a great answer! You are an admirable man and good teacher. :-)

regards,
George

George2 · 03-09-2006, 03:33 AM

jerril,

Quote:

Originally Posted by jerril

Hi George

Try this:
$ grep -c "^[^#]" foo

jer

I have tried your command. One question about it,

suppose the file content is,

--------------------
[root@localhost root]# more foo.grep
line 111
# line 2
line 3#
line 4
--------------------

When I use part of your command,

--------------------
[root@localhost root]# grep -c "[^#]" foo.grep
4
--------------------

I think it should return 2 since I think this command is looking for lines which do not contain '#'. Why the return value is 4.

regards,
George