Shell Programming and Scripting

hi

i need to create a shell script.
>this needs to run on a directory
> the directory has many sub-directories
>get the newly created directory
>get the file *.xyz from that directory
>find an occurence of word "abc" in that file

What have you tried so far?

zero progress.

---------- Post updated at 05:04 PM ---------- Previous update was at 05:01 PM ----------

infact i have tried a couple of things, but i am stuck on how to loop through a dir and take the latest dir

Can't assume a commercial environment here where directories could come and go in seconds.

If we assume that there is only one new directory in the last 24 hours the problem becomes manageable.
The "tail" in this gash solution ensures that we have only one answer to "find" and can give the most recent (not guaranteed at all) directory.

Code:

NEW_DIR="`find . -type d -ctime -1 -print|tail -1`"

echo "New directory is : ${NEW_DIR}"

The full solution is possible.


Beware: My commercial backup software updates the directory creation time. This timestamp is not a reliable parameter at all.

can u explain the script .. thanks

The unix "find" statement above looks for directories (-type d) with a creation time of under 24 hours (-ctime -1) and outputs the result of the find (-print) to standard output. In this case standard output channel from find is intercepted by the pipeline (|) and passed to the unix tail command which has been asked to provide the last line only from the pipeline (tail -1). The whole command sequence (find . -type d -ctime -1 -print|tail -1) is enclosed in backtick characters which in ksh an most modern shells cause the command sequence to execute. The result of the command execution which we expect to yield a directory name is enclosed in double quote characters to preserve space characters. The end product of the command sequence is placed into the environment variable NEW_DIR which we display on the screen with the unix "echo" command to show out result.

Beware: The unix "find" command has more varieties than Heinz and must be checked against your exact Operating System version.

once in that dir , methyl can u pls complete the below 2 steps

>get the file *.xyz from that directory
>find an occurence of word "abc" in that file

@ methyl : thanks. infact i am so excited to see the scripts you have written; thanks

---------- Post updated at 06:36 PM ---------- Previous update was at 06:21 PM ----------

i gave it a try .. i landed up with the following code

cd "${NEW_DIR}"
grep "abc" *.abc > abc.txt

maybe this looks simple, but it did my job ; let me know if script can be more enhanced.