I am very new with linux usage maybe this is my first time so i hope some detailed help please.
I have more than 500 files in multiple directories on my server (Linux) I want to change their extensions to .xml using bash script
I used a lot of codes but none of them work some codes i used :
for file in *.txt
do
mv ${file} ${file/.txt}/.xml
done
or
for file in *.*
do
mv ${file} ${file/.*}/.xml
done
i do not know even if the second one is valid code or not i tried to change the txt extension beacuse the prompt said no such file '.txt'
I hope some good help for that thank you
Explanation
For recursivity you need Bash >=4 and to enable ** (i.e. globstar) ;
First, I use parameter expansion to remove the string .txt, which must be anchored at the end of the filename (%) :
the # anchors the pattern (plain word or glob) to the beginning,
and the % anchors it to the end.
Then I append the new extension .xml
Be extra cautious with filename, you should always quote parameters expansion.
Code
This should do it in Bash (note that I only echothe old/new filename, to actually rename the files, use mv instead of echo) :
shopt -s globstar # enable ** globstar/recursivity
for i in **/*.txt; do
[[ -d "$i" ]] && continue; # skip directories
echo "$i" "${i/%.txt}.xml";
done
If its a matter of a one or two sub-directories, you can use the rename command:
rename .txt .xml *.txt
This will rename all the .txt to .xml files in the directory from which the command is executed.
If all the files are in same directory, it can be done using a single command. For example you want to convert all jpg files to png, go to the related directory location and then use command
rename .jpg .png *
I wanted to rename "file.txt" to "file.jpg.txt", used rename easy peezy:
rename 's/.txt$/.jpg.txt/' *.txt
man rename will tell you everything you need to know.
Got to love Linux, there's a tool for everything :-)
passing command line argument for dir path
#!/bin/sh
cd $1
names_1=`ls`
for file in ${names_1}
do
mv ${file} ${file}.jpg
done
Related
I have video files with this structure : GX**#### where #### is the number of the video.
But sometimes videos are splitted in 2 or 3 files, for instance : GX01#### and GX02####
My problem is that to organise my folders I prefer rename them like this : GX####-1 and GX####-2.
So i ask you the question if I can create a script/command to rename automatically my files to do this ? I can use .bat windows files or .sh linux files. My main problem is that i don't know a command to rename files simply (i saw many solutions which rewrite the entire file). May be if you know just this I will can do my script.
Thanks you in advance.
You can loop through the files and use PE parameter expansion to slice and jumble the strings and mv to rename the files.
Here is an example, first let us create an empty files using touch, but first we need to create a new directory (folder as you call it) and go into that newly created directory, using mkdir and cd.
mkdir newdirectory && cd newdirectory
Now create the empty files using touch.
touch GX01#### GX02#### GX03#### GX04####
Now check the empty files using ls
ls *
The output should be like
GX01#### GX02#### GX03#### GX04####
Now that we have created empty files, we can now do the renaming part.
for file in GX*; do
file1="${file#*??}"
file2=${file1#*??}
file3=${file1%*$file2}
echo mv -v "$file" "${file%*$file1}${file2}"-"$file3"
done
The output should be on stdout (which is your screen) because of the echo.
mv -v GX01#### GX####-01
mv -v GX02#### GX####-02
mv -v GX03#### GX####-03
mv -v GX04#### GX####-04
If you're satisfied with what you see then remove the echo, so mv can actually rename the files. To show how the slicing is done here is an example.
file=GX01####
file1="${file#*??}"
file2=${file1#*??}
file3="${file1%*$file2}"
printf '%s\n' "$file1" "$file2" "$file3"
The output should be something like.
01####
####
01
The ? means a single string from the shell.
A word of caution and advice, make a backup of the files you want to edit just in case the shell made a mistake while you're renaming the files. :-)
Also you should probably use shell globing like nullglob see
http://mywiki.wooledge.org/glob#nullglob
See man bash and look for Parameter Expansion.
PAGER='less +/^[[:space:]]*parameter\ expansion' man bash
Some online resources, with detailed explanation and examples.
http://mywiki.wooledge.org/BashFAQ/073
https://wiki.bash-hackers.org/syntax/pe
You could echo the original and new file names:
for f in GX*; do
echo "$f" $(sed 's/^GX\(..\)\(.*\)/GX\2-\1/' <<< "$f")
done
which should output:
GX01#### GX####-01
GX02#### GX####-02
then use mv -n instead of echo to rename the files.
I have a bunch of files named like this:
file.jpg?sw=450&sh=450
I want to batch rename them removing that awful extension and get this:
file.jpg
I tried this script:
for file in *'.jpg?sw=450&sh=450'; do mv "$file" "${file%}".jpg; done
and also this script:
for file in *'.jpg?sw=450&sh=450'; do mv "$file" "${file%}'.jpg?sw=450&sh=450'".jpg; done
What happens is I get this result:
file.jpg?sw=450&sh=450.jpg
Bash is one way to do it, although I might consider use of rename, which renames based on regular expressions. Consider:
$ rename 's/\.jpg\?.*$/.jpg/' *.jpg\?*
This says:
Given the files that match the glob *.jpg\?*
Replace, in the filenames, occurrences of .jpg that have a question mark following them and all characters after to the end: /\.jpg\?.*$/
With the simple string .jpg
I'm working on some text files, i have to remove a specific string
ex: "1234_NAME=TRUE", (including double quotes and comma)
Using Vim editor if i pass
:%s/"1234_NAME=TRUE",//g
But In this way i will be doing manually. But i want to automate the job by using some script.
I tried using sed tool, but it failed.
sed 's/"1234_NAME=TRUE",//g' hello.txt
I'm not expert on scripting. i know command is not correct. what is the solution for the above command.
And one more question:
Instead of passing file names manually, can we add a small loop which will take all text files in a directory one by one and remove the string.
Answer as suggested by #codeforester require some fix:
for file in *.txt
do
[[ -f "$file" ]] || continue
semicolon was the issue.
and Thanks to #SlePort for the answer.
Regards,
GBiradar
This might work for you:
sed -i.bak 's/"1234_NAME=TRUE",//g' *.txt
-i for in place editing
.bak: the extension of the backed up files
Your command is correct. To loop through your files, you can use a simple loop:
for file in *.txt; do # assuming your files are have .txt extension, modify accordingly
[[ -f "$file" ]] || continue # skip if not a regular file
sed 's/"1234_NAME=TRUE",//g' "$file" > "$file.modified" && mv "$file.modified" "$file"
# you can also use the `-i` flag to make a backup of the file and then overwrite the original
# `-i ''` will skip the backup and just overwrite the file
# sed -i .bak 's/"1234_NAME=TRUE",//g' "$file"
done
Make sure you have a backup of your files before running the above code.
I'm trying append to word "dicom" to the front of many filenames in a set of folders. The folders all begin with "s" (referred to by "s*" in the script below), and each contain many files (specified by "*" below)--I'd like all of these files to be changed using this bash script. I tried to run this:
for file in /Volumes/USB_AIB/DICOMFunCurrentBatch/MOVr1unzip/s*/*
do
mv $file dicom${file%%}
done
but got thousands of lines of the following error (once for each file within each folder--this is just an example of one of them):
mv: rename /Volumes/USB_AIB/DICOMFunCurrentBatch/MOVr1unzip/s307_1/29217684 to dicom/Volumes/USB_AIB/DICOMFunCurrentBatch/MOVr1unzip/s307_1/29217684: No such file or directory
Any ideas on how to fix it?
I don't you have a valid path as dicom/Volumes/USB_AIB/DICOMFunCurrentBatch/MOVr1unzip/s307_1/, why do you add dicom at the beginning?
maybe you want to append dicom to the end of the $file?
mv "$file" "${file}_dicom"
or something like that.
the following variable expansion ${file%%} is strange because it does nothing:
${parameter%%word} : remove the longest matching suffix pattern.
to move the file into a directory the path should exists, to create the path:
mkdir -p "$(dirname "${newfilename}")"
Maybe what you are trying to do:
for file in /Volumes/USB_AIB/DICOMFunCurrentBatch/MOVr1unzip/s*/*
do
mv "$file" "${file%/*}/dicom${file##*/}"
done
Hi I have a file that sorts some code and reformats it. I have over 200 files to apply this to with incremental names run001, run002 etc. Is there a quick way to write a shell script to execute this file over all the files? The executable creates a new file called run001an etc so just running over all files containing run doesnt work, how do i increment the file number?
Cheers
how about:
for i in ./run*; do
process_the_file $i
done
which is valid Bash/Ksh
To be more specific with run### files you can have
for file in dir/run[0-9][0-9][0-9]; do
do_something "$file"
done
dir could simply be just . or other directories. If they have spaces, quote them around "" but only the directory parts.
In bash, you can make use of extended patterns to generate all number matches not just 3 digits:
shopt -s extglob
for file in dir/run+([0-9]); do
do_something "$file"
done