CODEWITHSUNDEEP
bashrenamesubdirectory
Ahmed Amer
Ahmed Amer

Reputation: 1

create sub folders from specific characters of a file in linux

I have a files with the below format test_YYYYMMDDHHMMSS.csv

test_20200328223607.csv
test_20190523112250.csv
test_20180201065548.csv

I need to rename file then create a path from its name then add it to the created path with this format 2020/03/28/223607.csv

Example: test_YYYYMMDDHHMMSS.csv => mkdir YYYY/MM/DD from filename then rename file to be HHMMSS.csv and add it to YYYY/MM/DD

Upvotes: 0

Views: 157

Answers (2)

rtx13
rtx13

Reputation: 2610

Based on an earlier answer by @Cyrus:

#!/bin/bash

# define a regular expression for use with the `[[` conditional command later on to match filename patterns
re='^test_(....)(..)(..)(......)\.csv$'

# iterate through all csv files in the current directory
for i in *.csv; do
  # if filename does not match the regular expression, move on to the next loop iteration
  [[ "$i" =~ $re ]] || continue
  # the [[ conditional command leaves match results behind in an array variable named BASH_REMATCH

  # create a directory path using results; nop if directory already exists
  mkdir -p "${BASH_REMATCH[1]}/${BASH_REMATCH[2]}/${BASH_REMATCH[3]}"

  # ... and relocate the file to the (newly created) directory whilst giving it a new name
  mv "$i" "${BASH_REMATCH[1]}/${BASH_REMATCH[2]}/${BASH_REMATCH[3]}/${BASH_REMATCH[4]}.csv"
done

Upvotes: 1

David Stein
David Stein

Reputation: 399

You can use cut to get substrings and extract the directory names that way. There may be a cleaner solution but something like this should work

for i in *.csv
do
  # split the string
  IFS='_'
  read -ra ADDR <<< "$i"

  year=$(echo "${ADDR[1]}" | cut -c1-4)
  ... get other sub strings ...
  path=${year}/${month}/etc
  mkdir -p "$path"
  mv "$i" "${path}"/"${seconds}".csv
done

Upvotes: 0

Related Questions