Shell Programming and Scripting

Hi All,

I am new to unix, I had a file with 100 records and I want to copy first 50 records to a new file. Can i know how to write a script for this. Thanks in advance.

Sampath

If by records you mean lines:

Code:

head -n 50 < oldfile > newfile

Thank you Corona, What if i want copy 51 to 100 lines to a new file. Thank you.

Code:

# Create an example file with 1000 lines
# You need a bash or ksh shell for this kind of for-loop.
for ((N=1; N<=1000; N++)) ; do echo "$N" ; done > datafile
# display 50 lines starting at 50
tail -n+50 < datafile | head -n50
50
51
...
99

What's your actual goal with this? It may be better to start using a language like awk, or structuring your data in a different way.

Thank you. I am just playing around it. Is there any better way to do this(I really not aware of awk language), and please tell me where i can find good material about unix. Thanks in advance.

The 'best' way to do something depends on what you need to do and why.

Shell commands often work by combining several programs together. Here we're using the "tail" program to chop off some lines before and the "head" program to chop off some lines after in order to only get the section we want.

awk would probably be its own entire book. But you could do all that in one awk script without having to use several external programs, which may be faster in some cases (but not all cases). awk is a lot less flexible with files and redirection though, it's very task-specific. You often see a bit of a mix -- shell scripts which sometimes use short bits of awk for things awk makes really efficient or easy.

It would help to know what your system and shell is. You'll get different advice and books for Linux systems than HP-UX ones. They have a lot in common but are still quite different.

For example if you wanted to split a big file into a number of equal sized files (bytes or number of lines) the split command can do this for you.