Hello All,

Basically I'd like a command that I can run in a folder lets say Products- inside that folder has another set of folders of prodcut names, a- b- c- d etc.

In each of those(a b c d) -they are versions created 1, 2 ,3, 4 etc.

Each version is created on different days-

I'd like to list from folder products- all the products(a,b,c,d) that have versions created within the last 3 months. so the out put would to a txt file would be

a
e
h
j - all of which have Versions created within the last 3 months.

Anyway to do this?

Hi,
is this a homework? What have you got so far?

Those 'versions' -- are they files? Directories? Is there anything else in directories a, b, c, d...?

lol not homework- just giving an example.

yes theres .zip files and xml files in the version folders( 1 2 3 4) not in the a b c d. The only thing in a b c d is the 1 2 3 4 versions.

How about something like this?

#!/bin/bash

for fname in "$1"/*; do
if [ ! -d "$fname" ]; then continue; fi
if [ "$(find "$fname" -mindepth 1 -maxdepth 1 -mtime -90 | wc -l)" -gt 0 ]; then
echo "$fname"
fi
done


--

Some explanation would be perfect, I'm a linux noob!

fname would be which variable? the main folder?

we are echoing to fname variable instead of a .txt file?

why the greater than 0?

Looks good! but for me to use it I should understand it haha

Ok. The entire script is a loop.
It creates a variable called fname and loops over all values listed after the in keyword.
The expression "$1"/* is a glob that will be expanded by bash into the list of all files in the directory
"$1". The "$1" variable itself refers to the first command line argument of the script, so if I name the file "script"
and execute it as

the "$1" will be "./Products". If the Products directory contains directories a, b and c, the for line
will be expanded to

Note that unlike some other methods, such as pasting the output of ls, the * glob will expand into a set of correctly delimited filenames even in the
case when the filenames contain spaces.

The first thing I do with each fname is checking that it is a directory

The "!" sign is a negation (logical not), the -d is true if its argument exists and is a directory. The line therefore means
"if $fname is not a directory, skip the rest of the loop and continue with the next iteration"

Next part is using find (see man find for a complete manual).

[code]find "$fname" -mindepth 1 -maxdepth 1 -mtime -90[code]

This will search all files/directories in $fname that fit the following conditions:
-mtime -90 - was modified 90 or less days ago
-mindepth 1 -- min. level to be searched -- this prevents the directory $fname itself from being included in the list
-maxdepth 1 -- max. level to be searched -- this prevents find from searching the subdirectories of $fname recursively

The output of find is then piped to wc which counts the entries returned by find. Both commands are enclosed in $( ), which means
"run the commands in parentheses and paste their output here":

The output (which is the number of files in $fname that were modified in the last 90 days) is then compared with zero in an if construct
which means "if there's more than zero files modified in the last 90 days, print $fname"

So, that's about it. Hope this makes it more clear

Oh, sorry if this is more detailed than what you wanted... It's just that it is a long boring afternoon in a lonely office and I am trying to keep myself busy by writing long posts )

I think I should take you to dinner...

IT WORKS! I'm just going to edit it with an echo command to a txt file... see if I can do that

THANKS SO MUCH, saved me the time of manually having to do this