If loop not working as intended for perl scrpt

Question

I'm trying to create a Perl script that automatically check for changes in a data.list every hour, username/project/tota/datas/data.list is where the file is and then execute the remaining codes, which in my case, is to count the number of lines in that data.list and upload using CURL to online database.

#Look for changes every 6 hours
If (changes to data.list)
{
 count number of lines,
 upload data
}
else ( no change )
{
 do nothing
}

The part whereby counting lines and uploading data is done. just need to find a way to automatically search for changes in data.list Thanks all

Edited new : i got this answer from cas,

use Digest::SHA qw(sha256_hex);
my $filename = 'username/project/tota/datas/data.list';
my $old_mtime = 0;
my $old_size = 0;
my $old_digest = '';
while(1) { # loop forever
 my ($dev, $ino, $mode, $nlink, $uid, $gid, $rdev, $size,
 $atime, $mtime, $ctime, $blksize, $blocks) = stat($filename);
 if ($mtime != $old_mtime || $size != $old_size) {
 # slurp in entire file and get its sha256 hash
 local $/;
 open(my $fh, "<", $filename) or die "couldn't open $filename: $!\n";
 my $digest = sha256_hex(<$fh>);
 close($fh);
 if ($digest ne $old_digest) {
 print "change detected";
 $old_digest = $digest; 
 $old_mtime = $mtime;
 $old_size = $size
 }
 } else {
 print "no change detected";
 };
 sleep 3600; # sleep 1 hour between iterations of the loop
}

problem with this is whenever i run it, it always start with change detected and then it will go into no change detected every hour, any idea how can i change this part

• cas: Do you need to know if individual files have been added to or deleted from the directory? or just a count of the files? If you need to keep track of individual files, do you also want to just upload any new files, or re-upload everything?
• nobody: I would suggest that you keep the previous version of data.list to be able to spot the changes. Than run diff --suppress-common-lines data.list.old data.list. There are many parameters for diff that can help you customize the output to your preferences (e.g. --line-format), In PERL, you can catch the output of diff and perform actions you need to.
• Hermann: When reacting to file changes – are you aware of inotify (example here)?
• HelpMeAndroidStudio: @Hermann hi, cas's is working the same as ur exmaple ! thanks tho ! anyway i edited my question, care to check it out ?

Answer 1

If you only need to know IF a file has changed and don't need to know exactly WHAT has changed, the easiest way is to compare a hash of the current file against a hash of the previous file.

This can be further optimised to avoid an expensive hash computation by first comparing file timestamps and file sizes.

e.g. using SHA256 from Digest::SHA:

use Digest::SHA qw(sha256_hex);
my $filename = 'username/project/tota/datas/data.list';
my $old_mtime = 0;
my $old_size = 0;
my $old_digest = '';
while(1) { # loop forever
 my ($dev, $ino, $mode, $nlink, $uid, $gid, $rdev, $size,
 $atime, $mtime, $ctime, $blksize, $blocks) = stat($filename);
 if ($mtime != $old_mtime || $size != $old_size) {
 # slurp in entire file and get its sha256 hash
 local $/;
 open(my $fh, "<", $filename) or die "couldn't open $filename: $!\n";
 my $digest = sha256_hex(<$fh>);
 close($fh);
 if ($digest ne $old_digest) {
 # the file has changed. upload it
 #....your curl upload code here...
 # don't forget to update the old_* variables
 $old_digest = $digest; 
 $old_mtime = $mtime;
 $old_size = $size
 }
 } else {
 # either the file hasn't changed or someone evil has modified it while
 # making sure the file size and mtime remains the same.
 # you'd need something like Digest::SHA to detect that :-)
 };
 sleep 3600; # sleep 1 hour between iterations of the loop
}

You can install Digest::SHA from CPAN, or it's probably already packaged for your distro. On Debian, it's in the libdigest-sha-perl package.

---

I'm not sure why you even need or want to do this in perl. If all you want to do is check if a file has changed every hour, then you'd be better off just running a simple shell script like the following from cron:

#!/bin/bash
# run this as "/path/to/script-name.sh /path/to/data.list"
filename="1ドル"
checksumfile='/var/tmp/data.list.sha256'
# cd to the directory containing data.list
cd "$(dirname "$filename")"
if [ ! -e "$checksumfile" ] || ! sha256sum --quiet -c "$checksumfile" ; then
 # upload your file with curl
 # ... your curl code here ...
 
 # generate sha256 checksum file
 sha256sum "$(basename "$filename")" > "$checksumfile"
 # make sure it's RW by everyone
 chmod a+rw "$checksumfile"
fi

• HelpMeAndroidStudio: not files, its the data.list , i want to see if there are more data added into that list every hour
• cas: in that case, you can probably just use the file's timestamp and not bother with counting the lines. Do you need to upload the entire file again or just the changes?
• HelpMeAndroidStudio: upload the entire data.list file again, im new at this , so im not sure how to do
• cas: 1. ask the sysadmins to install it for you. 2. if it's a personal script that's only going to be run by you, then install it into your home directory with cpan Digest::SHA
• HelpMeAndroidStudio: cool buddy ! thanks for the help ! much appreciated, i am going to type out now on my linux and tty