I want to rename all the files in a folder which starts with 123_xxx.txt to xxx.txt.
For example, my directory has:
123_xxx.txt
123_yyy.txt
123_zzz.txt
I want to rename all files as:
xxx.txt
yyy.txt
zzz.txt
I have seen some useful bash scripts in this forum but I'm still confused how to use it for my requirement.
Let us suppose I use:
for file in `find -name '123_*.txt'` ; do mv $file {?.txt} ; done
Is this the correct way to do it?
You can do it this way:
find . -name '123_*.txt' -type f -exec sh -c '
for f; do
mv "$f" "${f%/*}/${f##*/123_}"
done' sh {} +
No pipes, no reads, no chance of breaking on malformed filenames, no non-standard tools or features.
find . -name "123*.txt" -exec rename 's/^123_//' {} ";"
will do it. No AWK, no for, no xargs needed, but rename, a very useful command from the Perl lib. It is not always included with Linux, but is easy to install from the repos.
In case you want to replace string in file name called foo to bar you can use this in linux ubuntu, change file type for your needs
find -name "*foo*.filetype" -exec rename 's/foo/bar/' {} ";"
you could check 'rename' tool
for example
rename 's/^123_//' *.txt
or (gawk is needed)
find . -name '123_*.txt'|awk '{print "mv "$0" "gensub(/\/123_(.*\.txt)$/,"/\\1","g");}'|sh
test:
kent$ tree
.
|-- 123_a.txt
|-- 123_b.txt
|-- 123_c.txt
|-- 123_d.txt
|-- 123_e.txt
`-- u
|-- 123_a.txt
|-- 123_b.txt
|-- 123_c.txt
|-- 123_d.txt
`-- 123_e.txt
1 directory, 10 files
kent$ find . -name '123_*.txt'|awk '{print "mv "$0" "gensub(/\/123_(.*\.txt)$/,"/\\1","g");}'|sh
kent$ tree
.
|-- a.txt
|-- b.txt
|-- c.txt
|-- d.txt
|-- e.txt
`-- u
|-- a.txt
|-- b.txt
|-- c.txt
|-- d.txt
`-- e.txt
1 directory, 10 files
To expand on Sorpigal's answer, if you want to replace the 123_ with hello_, you could use
find . -name '123*.txt' -type f -exec bash -c 'mv "$1" "${1/\/123_/\/hello_}"' -- {} \;
A slight variation on Kent's that doesn't require gawk and is a little bit more readable, (although, thats debatable..)
find . -name "123*" | awk '{a=$1; gsub(/123_/,""); printf "mv \"%s\" \"%s\"\n", a, $1}' | sh
Using rename from util-linux 2.28.2 I had to use a different syntaxt:
find -name "*.txt" -exec rename -v "123_" "" {} ";"
Provided you don't have newlines in your filenames:
find -name '123_*.txt' | while IFS= read -r file; do mv "$file" "${file#123_}"; done
For a really safe way, provided your find supports the -print0 flag (GNU find does):
find -name '123_*.txt' -print0 | while IFS= read -r -d '' file; do mv "$file" "${file#123_}"; done
You can make a little bash script for that.
Create a file named recursive_replace_filename with this content :
#!/bin/bash
if test $# -lt 2; then
echo "usage: `basename $0` <to_replace> <replace_value>"
fi
for file in `find . -name "*$1*" -type f`; do
mv "'$file'" "${file/'$1'/'$2'}"
done
Make executable an run:
$ chmod +x recursive_replace_filename
$ ./recursive_replace_filename 123_ ""
Keep note that this script can be dangerous, be sure you know what it's doing and in which folder you are executing it, and with which arguments. In this case, all files in the current folder, recursively, containing 123_ will be renamed.
Tried the answer above but it didn't work for me cause i had the string inside folders and files name at the same time so here is what i did the following bash script:
for fileType in d f
do
find -type $fileType -iname "stringToSearch*" |while read file
do
mv $file $( sed -r "s/stringToSearch/stringToReplaceWith/" <<< $file )
done
done
First i began by replacing inside folders name then inside files name.
Here's the only solution to date that works:
find-rename-recursive --pattern '123_' --string '' -- . -type f -name "123_*"
All other solutions don't work for me--some even deleted files!
For details, see https://github.com/l3x/helpers#find-rename-recursive
If the names are fixed you can visit each directory and perform the renaming in a subshell (to avoid changing the current directory) fairly simply. This is how I renamed a bunch of new_paths.json files each to paths.json:
for file in $(find root_directory -name new_paths.json)
do
(cd $(dirname $file) ; mv new_paths.json paths.json)
done
using the examples above, i used this to replace part of the file name... this was used to rename various data files in a directory, due to a vendor changing names.
find . -name 'o3_cloudmed_*.*' -type f -exec bash -c 'mv "$1" "${1/cloudmed/revint}"' -- {} \;
Below, the find command searches recursively in the tree from the directory passed as first parameter (here it is the current directory "."). The following parameters are predicates to filter the files we are looking for. The "-name" predicate is a pattern to which the current file name must match (here we want manage only filenames beginning with '123_' and terminating with '.txt'. The "-type" filters the type of the file ('f' means regular file, 'd' would mean directory...). The "-exec" predicate runs a command line into which '{}' is the current file name and the ';' terminates the command.
find . -name '123_*.txt' -type f -exec bash -c 'name={};mv $name ${name//123_//}' \;
prename s/^123_// 123_*
See prename in the official Perl5 wiki. Copy for your convenience:
prename - A script that renames files according to a regular expression. (Where was the original published?) Originally named rename, it is found mostly as prename because the original one clashes with the rename command from the util-linux package. Numerous forks and reimplementations:
