CODEWITHSUNDEEP
arrayslinuxbashlistfile-listing
Jaw. M.
Jaw. M.

Reputation: 149

Writing a bash script for checking filenames in a directory by using an array with specific conditions

There is a directory including several XML files.

Each filename has the following format: yyyy-mm-dd-hh-mm-ss-AUFXXXXXXXXX.xml

Some XML filenames have the same number which is coming after AUF because they've been modified. I want to have a bash script which should recognize these kinds of files, keep the origin ones by checking the timestamps and move the modified files to another folder. I'm trying to write bash script to solve this issue. This is what I've got so far:

#!/bin/bash
declare -a filelist
shopt -s extglob

for file in $(ls -tr1 *AUF*); do
  filelist=(${filelist[@]} "$file")
  echo "${file}"
done

part of output:

2019-11-14-17-44-04-AUF19000276.xml
2019-11-15-09-12-01-AUF19000276.xml
2019-11-15-09-27-26-AUF19000276.xml
2019-11-15-09-28-51-AUF19000276.xml
2019-11-18-13-50-34-AUF19000296.xml
2019-11-20-16-45-14-AUF19000300.xml
2019-11-27-12-16-25-AUF19000292.xml
2019-11-27-12-19-50-AUF19000225.xml
2019-11-27-17-11-04-AUF19000300.xml
2019-11-28-09-40-44-AUF19000294.xml
2019-11-29-17-03-33-AUF19000305.xml
2019-11-29-17-04-43-AUF19000306.xml
2019-11-29-17-05-41-AUF19000306.xml
2019-12-02-12-02-20-AUF19000305.xml
2019-12-02-12-03-00-AUF19000305.xml
2019-12-03-09-22-06-AUF19000307.xml
2019-12-04-10-49-03-AUF19000308.xml
2019-12-05-09-23-54-AUF19000310.xml
2019-12-05-09-24-41-AUF19000310.xml
2019-12-09-13-12-31-AUF19000256.xml
2019-12-09-13-59-42-AUF19000256.xml
2019-12-09-15-29-25-AUF19000281.xml
2019-12-09-15-30-13-AUF19000281.xml
2019-12-09-15-34-07-AUF19000284.xml
2019-12-09-15-39-39-AUF18000346.xml
2019-12-09-15-40-21-AUF19000058.xml
2019-12-10-16-19-35-AUF19000312.xml
2019-12-11-11-58-55-AUF19000313.xml

For example: I want to keep the first created file 2019-11-14-17-44-04-AUF19000276.xml and move the other three files with the same number to another directory.

I currently don't know how to check through the array and include the above-mentioned conditions. Glad for any help!

Upvotes: 1

Views: 123

Answers (2)

tshiono
tshiono

Reputation: 22012

Would you please try the following:

dest="another"          # folder to store the modified files
mkdir -p "$dest"
declare -A seen         # associative array to count AUFXXX substring
while IFS= read -r f; do
    auf="${f##*-}"      # extract the substring after "AUF"
    auf="${auf%.*}"     # remove extension
    (( seen[$auf]++ )) && mv -- "$f" "$dest"
                        # if the AUFXXX substring is seen then move the file
done < <(find . -name "*AUF*.xml" -printf "%T@\t%p\n" | sort -n | cut -f 2-)

It extracts the AUFXXX substring from each file and counts up the appearance in the associative array seen. If the value of seen is nonzero, then the file is a modified one and move it to another directory.

Upvotes: 1

tink
tink

Reputation: 15204

I think the following (minus the echo) will do the job ... 

# iterate over the AUF* parts
for i in $( ls | sed -r 's/^.*(AUF[^.]+)\.xml/\1/' | sort -u )
# iterate over the sections, move all but the first (oldest) files
do for j in $( ls -1 *${i}*| tail -n +2 )
    do echo mv $j newdir
    done
done

Remove the echo if the output looks sane =}

Upvotes: 3

Related Questions