Shell Programming and Scripting

Hi,

This is my first thread on this forum.

I have an urgent requirement where I have to write a shell script to
- scan through the mails one by one,
- pick up the time when that mail is received,
- parse through the contents of that mail and pick up relevant information and load it into the database, and finally
- copy this mail to some other location.

The time the mail in received in obtained from the 1st line

for e.g.
From abc@xyz.com Mon Jan 15 11:18:18 2007

I need to pick up the time - "Mon Jan 15 11:18:18 2007"

After parsing through the contents (subject of the mail), which I think I can manage, I need to copy the mail into a different location, so that this mail is not processed once again.

This process would need to run every night.

Can this be done using shell scripts ??

Alternately, is there any way the mails can be split into individual mails and copied into separate files? I would need the header information in this as well as I need to pick the timestamp of the mail-received time.

Any help would be appreciated. I am short of time.

Many Thanks.

Ashish

You are on unix therefore everything is possible ... well not really but have a look here

Many Thanks for the response Andryk.

A few questions ...
I had worked with a sample of 3 mails.

1) I am not sure is `seq150` works in Unix. So I had used
for i in 1 2 3 instead of for i `seq 3`in the 1st line of your script.
2) In your script, in line no - 3,
echo s $i file.$i >> forpipe, I am not sure why was $i also echoed. For e.g. if there are 3 mails, then the file "forpipe" would look
like
s 1 file.1
s 2 file.2
s 3 file.3

Now when I had run your script, I got the following files created - file.1, fil2.2,file.3,1,2,3.

When I removed $1 from the script (line no 3), I got 3 files - file.1,file.2 and file.3.

3) The script runs fine if the number of mails = contents in the file forpipe.
for e.g.
the contents in the file file - forpipe is
s file.1
s file.2
s file.3
x

The number of mails = 3.

However when the number of mails = 4, and there are 3 files listed in forpilpe, the files are still created, but the mail 1-3 are not deleted. Am not sure why.

In my case, the mails can come any time, and a new mail could have come after the creation of forpipe, while the mails are being processed. I cannot process the email again, because each email is expected to be unique.

Is there a way out ?

Apologies for being so lengthly ... I hope I am making sense.

Many Thanks.

My apologize, that script was for linux when i was asked to resubmit someone's mail to his new office ...
I just wanted to show that You can split a unix mailbox provided You know the number of mail in the mailbox

Code:

let i=0
while [ $i -lt $number_of_mails ]
do
        let i=$i+1
        echo  s $i file.$i >> forstdin
       # Add more cmds here if needed...
done

echo x >> forstdin
# redirect standard input for the 'mail' utility
mail -f /var/spool/mail/usr_mbox < fortsdin

For the line that says 'echo s $i file.$i >> forstdin', it just relies on linux's mail utility that says to store mail at index n to file.n (s n file.n) no typo there , BTW You are on which unix flavor ?

Just copy one mailbox to one directory and try it there, it will split up the mailbox to file.1 file.2 ..., if You need to delete mail just dive into mail's man page You will not get lost, thats for sure

Thanks for the modified script.

I am using Unix Korn shell on OS 5.8.

I am not sure if I have explained clearly.

Assuming there are 3 mails, and I use 'echo s $i file.$i >> forstdin', I get 6 files instead of 3. The files created are file.1,file.2,file.3,1,2,3. However when I use 'echo s file.$i >> forstdin', 3 files are created - file.1, file.2 and file.3, which seems okay.

The user command 's' save the mail into a file and deletes the mail, which is fine. However, as I had mentioned below the mails are not deleted when the number of files mentioned in 'fortsdin' is less than the number of mails.
I guess this has to do with the way unix works.

Thanks,
Ashish