Scripting question

nazs · 04-25-2008, 06:36 PM

Hi All,
I have a script that looks for a file. Then changes the name of the file and then scp's it to another server. The problem is this was set up for one file to be in the directory. Now there is going to be another file with the same name but different information in it. I need to add a way to check inside the file and look on the first line of the file for a sequence of numbers(ex. 123456789). If that file has the sequence of numbers then finish the rest of the script.If it does not match the sequence then exit the script. My existing script is below.

#!/bin/sh -x

pospay=/usr/pa/archive/bw/bin_wsobserve/pospay

archive=/usr/pa/archive/bw/bin_wsobserve/pospay/archive

dest=/usr/pa/archive/bw/bin_wsobserve/pospay/wsobserve

sent=/usr/pa/archive/bw/bin_wsobserve/pospay/wsobserve/sent

cd "$pospay"

#CHECK IF THERE ARE ANY POSITIVE PAY FILES
if [ `find . -type f -maxdepth 1 -name "1455*" | wc -l` -eq 0 ]; then
echo "Exit NO FILES"
exit
else

gunzip 1455*

for file in 1455*
do
#this copies the file to new directory and removes everything after the -

cp "$file" "$dest/${file%%-*}"

gzip 1455*

mv "$pospay"/*.gz "$archive"

cd "$dest"

scp 1455* abcd@xfer.abcd.com:/usr/pa/warner/wbdumpdir/bin_wsobserve

gzip "$dest"/1455*

mv "$dest"/1455* "$sent"

slogin abcd@xfer.abcd.com "cd /usr/pa/warner/wbdumpdir/bin_wsobserve" "touch cur_dump.wsobserve;echo "1455"* >> cur_dump.wsobserve
;/usr/local/bin/bwconcord_send_wsobserve"

done
fi

Thanks,
Nazs

jailbait · 04-25-2008, 07:11 PM

Quote:

Originally Posted by nazs

Hi All,

The problem is this was set up for one file to be in the directory. Now there is going to be another file with the same name but different information in it.

How can you have two files with the same name in the same directory?

--------------------------
Steve Stites

jschiwal · 04-25-2008, 07:47 PM

The first file by the same name will probably be overwritten when the second file is saved. It may be possible if it was created by something that directly manipulates the file system or if the file system becomes currupted. In that case you will need to list the inodes of the file and use the find command to select the file by the inode. The "ls -i" command will list the inode as well as the filename. Using something like "find . -inum 7159974 -exec mv -v '{}' temp/'{}' \;" will allow you to move one of the duplicates out of the way.

However, having two files in the same directory with identical names is something that needs to be prevented. Either add a time stamp or other meaningful pattern to the filename or create them in separate directories whose names are based on some criteria.

nazs · 04-26-2008, 02:25 PM

I am sorry. Let me clarify the same name question. Both files start with 1455.LOC. The program that we use creates this file name and then adds a random ending to it. Example file would look like this 1455.LOC-30421-7787E443609E138318DFFC27C14E80A6.gz
So everything after the dash is random. That way the file does not get over written. But there is no way to tell what will be added on after the dash.

Thanks,
Nazs

angrybanana · 04-27-2008, 12:48 AM

Do you mean something like this?

Code:

if [[ $(head -1 file) == 123456789 ]];then
    echo "do stuff"
else
    exit
fi

another way of doing the same..

Code:

read line < file
if [[ $line == 123456789 ]];then
    echo "do stuff"
else
    exit
fi