moving files by date

[Mark Vollmers]

I have a folder with a very large number of files in it. I need to go through that and move all of the files that were created from a certain date on (say 1-19-01) to another folder. What is the easiest way to do this? I can't find anything for mv that talks about moving by date, and I would really hate to list them and move individually (or even in handfuls). There could be several hundred files or more. Thanks in advance!

[Curtis Larson]

how about trying this:

here=/tmp/test
there=/tmp/test


for month in Jan Feb # etc
do
(( day = 0 ))
while (( $day < 31 ))
do
(( $day = $day + 1 ))
[ ! -d $there/${month}.${day} ] && mkdir $there/${month}.${day}
there=$there/${month}.${day}

ll $here | tr -s '[:space:]' | awk "/$month $day/"'{print $NF;}' |
while read file
do
mv $here/$file $there/
done
done
done

[James R. Ferguson · Accepted solution]

Hi Mark:

Use the 'find' command with the 'newer' option and a "reference file" that you specify with the date and time transition:

# cd
# touch -mt myfile
# find . ! -newer myfile -exec mv {} \;

See the man pages for 'find'.

...JRF...

[Curtis Larson]

probably want to add:

[ -f $file ] && mv $here/$file $there/

so you don't affect directories and such

[Mark Vollmers]

Thanks for the help! I used James suggestion (Curtis, not that your's wouldn't work, but James' was shorter :) ) Thanks. It saved me a lot of time.

[James R. Ferguson]

Hi Mark (again):

A postscript (pun intended).

'find' is intended to descend a directory structure. In some cases, you may want to limit this action. For instance, if you had subdirectories and files therein, upon which you did not want 'find' to operate, you could add the and options to the syntax I presented above. Thus:

# find . ! -newer myfile -prune -type f -exec mv {} \;

The "-type f" selects only regular files; not directories, etc. The "-prune" keeps 'find' from examining the contents of the subdirectory in the first place.

Regards!

...JRF...