Shell Programming and Scripting

I have this script to read a file from a directory,
for FILE in `ls -1 ../dir` ;do
echo Renaming $FILE


with a filename TRO029274465.txt now i want to rename this file to a certain string found inside this textfile.

For example this
TRO029274465.txt contains
H20090812982009081211021025301000000012689925
T200908129820000105700668330000000126899250000000214600004
...

i want to get the string in 9-16
"9820090812"
so that my new file name would be 20090812.98

Been trying on some commands but it seems that i dont get the right format for this one.
pls help. I'm a linux newbie

Thanks,
-dying octupos

Welcome.

First, be careful with the construct of

Code:

for FILE in `ls -1 ../dir`

The shell (default) will break the file names if they contain spaces, tabs, etc. It is also not necessary.

Instead, use:

Code:

for FILE in ../dir/*

You do not specify where in the file the string in question is? Top line in the file? Does the the string need to be greped?

More detail needed...

Yeap the string[9-16] in the top of the line. And no it doesn't need to be grepped.
My apologies.

Code:

for file in `ls dir`
do
    echo $file
    var=`head -1 dir/$file |cut -c 10-19 | sed -r 's/([0-9]{2})([0-9]+$)/\2.\1'/`
    echo $var;
    `mv A/$file dir/$var`
done

Here dir is the directory name. So you can give your directory.

hi drewk,
i already found a command for this one:

name=`cut -c1 12-19 $FILE`

but my problem now is it works only with those file who only have one line or row.
but my file contains many rows

-dying octupos

---------- Post updated at 10:37 PM ---------- Previous update was at 10:20 PM ----------


OMG!! this works very fine!!

thank you very much murugaperumal

If this is bash, there's no need for a command, just use parameter expansion..

Code:

${string:11:8} and ${string:9:2}

First get the string into a variable. You may want to use "string=$(head -1 yourfile)", I just did it differently as an example. Then the parameter expansion is used to cut out portions of the variable using two numbers to locate the substring you want. In the first one, it cuts out 11 characters from the left and leaves in the next 8. In the second one, it cuts out 9 and leaves the next two. Put a dot between them and you have the filename you need.

Code:

# string=H20090812982009081211021025301000000012689925
# echo $string
H20090812982009081211021025301000000012689925
# echo ${string:11:8}.${string:9:2}
20090812.98
#

Your code may be something like this:

Code:

NEWNAME=$(head -1 $FILE)
mv $FILE ${NEWNAME:11:8}.${NEWNAME:9:2}

Code:

echo "H20090812982009081211021025301000000012689925" |awk '{print substr($0,12,6)"."substr($0,10,2)}'