LinuxQuestions.org - [SOLVED] Mass file renaming

- Linux - Newbie (https://www.linuxquestions.org/questions/linux-newbie-8/)

- - Mass file renaming (https://www.linuxquestions.org/questions/linux-newbie-8/mass-file-renaming-4175453776/)

Mass file renaming

i'm looking for how to rename files in mass quantity. basically i am completely moving from windows (ntfs) to CentOs (ext4). in windows i used to give spaces in between file names, now i want to replace spaces with "-" or "_", so that i get the full file name when i hit tab in terminal.

i've searched google and also looked into "similar threads" of LQ, but could not find answer for my specific requirement. either they are for particular directory or have a pattern/incremental in nature, for example renaming *.jpeg files, file1*.jpg file123.jpg etc.

i'm looking for solutions based on "find" command.

for example:

Code:

# find /dc/cbt/ -type f -iname "*bind*"



/dc/cbt/Books/DNS Bind Configuration in RHEL 6.pdf

/dc/cbt/Linux/Install & Configure BIND DNS Server in CentOS.mp4

/dc/cbt/Linux/YouTube/rhel 6 bind primary dns server in redhat linux.flv

i know how to display modified files, but don't know how to write them permanent. that's the problem :doh:

Code:

# find /dc/cbt/ -type f -iname "*bind*" | sed "s/ /_/ig" | tr "[:upper:]" "[:lower:]"



/dc/cbt/books/dns_bind_configuration_in_rhel_6.pdf

/dc/cbt/linux/install_&_configure_bind_dns_server_in_centos.mp4

/dc/cbt/linux/youtube/rhel_6_bind_primary_dns_server_in_redhat_linux.flv

thanks...

How about something like this?

Code:

while read -r name; do

    lower="${name,,}"

    newname="${lower// /_}"

    echo "$name --> $newname"

    # mv "$name" "$newname"

done < <(find /dc/cbt/ -type f -iname "*bind*")

Be very careful with this. Backup your data.
You may also look at the manpage of rename command, which is quite a useful tool for mass renaming

Code:

#!/bin/bash



find /dc/cbt/ -type f -iname "*bind*" | while read line; do

  oldfile=$line

  newfile=${oldfile// /_}

  newfile=${newfile,,}

  echo mv "$oldfile" "$newfile"

done

Once you're sure it's working correctly, remove the "echo" to actually rename the files.

edit: too slow

Thanks millgates & suicidaleggroll. I tried both, and both works great!!! :hattip:

while trying, i realised that certain files have multiple spaces. so i changed a little. i think when both scripts are translated to machine language, they will be same.

Code:

#!/bin/bash

while read -r name; do

    lower="${name,,}"

    newname=`echo "$lower" | sed 's/  */_/g'`

    echo "$name --> $newname"

    mv "$name" "$newname"

done < <(find /dc/cbt/ -type f -iname "*bind*")

Code:

#!/bin/bash

find /dc/cbt/ -type f -iname "*bind*" | while read line; do

  oldfile=$line

  newfile=`echo "$oldfile" | sed 's/  */_/g'`

  newfile=${newfile,,}

  mv "$oldfile" "$newfile"

done

As an alternative to sed you could use bash extglob:

Code:

shopt -s extglob

foo="a  b  c"



# replace spaces with underscores

echo "${foo//+( )/_}"

also, I would recommend to use the $( command ) syntax for command substitution. It handles nesting and double quotes nicely, unlike the backticks.

Quote:

shopt -s extglob

??? :scratch:

Re shopt = shell options ie amend the way the shell works.
See http://linux.die.net/man/1/bash and search for shell builtin commands, then shopt, then extglob within that.
Its about 80% of the way down that page or google for examples etc.