Commands using xargs (769)
-
0seq 4|xargs -n1 -i bash -c "echo -n 164.85.216.{} - ; nslookup 164.85.216.{} |grep name"|tr -s ' ' ' '|awk '{print $1" - "$5}'|sed 's/.$//'
-
Nice command to create a list, you can create too with for command, but this is so faster. Show Sample Output
0seq 10 |xargs -n1 echo Printing line -
not my cmd... found on the web Show Sample Output
0perl -e '$i=0;while($i<10){open(WGET,qq/|xargs lynx -dump/);printf WGET qq{http://www.google.com/search?q=site:g33kinfo.com&hl=en&start=$i&sa=N},$i+=10}'|grep '\/\/g33kinfo.com\/' -
Alternately for those without getent or only want to work on local users it's even easier:
cut -d: -f1 /etc/passwd|xargs -n1 passwd -eNote that not all implementations of passwd support -e. On RH it would be passwd -x0 (?) and on Solaris it would be passwd -f.0getent passwd|cut -d: -f1|xargs -n1 passwd -e -
I know this has been beaten to death but finding video files using mime types and printing the "hours of video" for each directory is (IMHO) easier to parse than just a single total. Output is in minutes. Among the other niceties is that it omits printing of non-video files/folders PS: Barely managed to fit it within the 255 character limit :D Show Sample Output
0for item in *;do echo -n "$item - ";find "$item" -type f -print0 | xargs -0 file -iNf - | grep video | cut -d: -f1 | xargs -d'\n' /usr/share/doc/mplayer/examples/midentify | grep ID_LENGTH | awk -F= '{sum+=$2} END {print(sum/60)}'; done | grep -v ' - 0$' -
0git clean -n | sed 's/Would remove //; /Would not remove/d;' | xargs mv -t stuff/
-
0ifconfig -l | xargs -n1 ipconfig getifaddr 2> /dev/null
-
0dpkg -l | grep ^rc | awk '{print $2}' | sudo xargs dpkg -P
-
0ipcs -a | grep 0x | awk '{printf( "-Q %s ", $1 )}' | xargs ipcrm
-
Changed out the for loop for an xargs. It's a tad shorter, and a tad cleaner.
0find . -type f |xargs -I% sed -i '/group name/s/>/ deleteMissing="true">/' % -
0git log -g --pretty=oneline | grep '}: commit' | awk '{print $1}' | head -1 | xargs git checkout -f
-
0pattern='regexp_pattern'; find . -type f -perm +220 ! -name '*.bak' -print0 | xargs -0 egrep -lZ $pattern | xargs -0 sed -i.bak -e "/$pattern/d"
-
0svn up | sed 's/\.//g' | cut -d ' ' -f3 | xargs svn log -r
-
0find . -name file.txt | xargs -e grep "token" -o | wc -l
-
Warning!, if the pattern didn't find anything it shows the total size of dot dir Show Sample Output
0find . -name 'pattern'| xargs du -hc -
find -exec is evil since it launches a process for each file. You get the total as a bonus. Also, without -n sort will sort by lexical order (that is 9 after 10).
0find . \( -iname '*.[ch]' -o -iname '*.php' -o -iname '*.pl' \) | xargs wc -l | sort -n -
This is an updated version that some one provided me via another "find" command to find files over a certain size. Keep in mind you may have to mess around with the print values depending on your system to get the correct output you want. This was tested on FC and Cent based servers. (thanks to berta for the update) Show Sample Output
0find / -type f -size +512000 | xargs ls -lh | awk '{ print $5 " " $6$7 ": " $9 }' -
This command shows the size of directories below here, refreshing every 2s. It will also track directories created after running the command (that what the find bit does). Show Sample Output
0watch 'find -maxdepth 1 -mindepth 1 -type d |xargs du -csh' -
on a dpkg managed system this PATTERN will help you generate .deb files from source AND remove all the dev libs you had to install. i hate cluttering up my machine with rouge packages and headers. it would be pretty darn easy on rpm systems as well. i just dont have a rpm managed system to test on right now. NOTE, you sharp ones will notice that it uninstalls the deb you just made! yeah, but the deb is still there to do with it what you want, like re install it. or you can just grep -v after the diff
0dpkg-query -l > 1.lst; sudo apt-get install -y build-essential; ./configure; make; sudo checkinstall -D make install; dpkg-query --list > 2.lst; diff 1.lst 2.lst | grep '^>' | awk '{print $3}' | xargs sudo apt-get remove -y --purge -
The following command finds all the files not modified in the last 5 days under /protocollo/paflow directory and creates an archive files under /var/dump-protocollo in the format of ddmmyyyy_archive.tar
0find /protocollo/paflow -type f -mtime +5 | xargs tar -cvf /var/dump-protocollo/`date '+%d%m%Y'_archive.tar` -
0find -type f -size +0 -mtime +1 -print0|xargs -0r rm -f
-
The find command isn't the important bit, here: it's just what feeds the rest of the pipe (this one looks for all PDFs less than 7 days old, in an archive directory, whose structure is defined by a wildcard pattern: modify this find, to suit your real needs). I consider the next bit the useful part. xargs stats out the byte-size of each file, and this is passed to awk, which adds them all together, and prints the grand total. I use printf, in order to override awk's tendency to swtich to exponential output above a certain threshold, and, specifically "%0.0f\n", because it was all I can find to force things back to digital on Redhat systems. This is then passed to an optional sed, which formats them in a US/UK number format, to make large numbers easier to read. Change the comma in the sed, for your preferred separator character (e.g. sed -r ':L;s=\b([0-9]+)([0-9]{3})\b=\1 \2=g;t L' for most European countries). (This sed is credited to user name 'archtoad6', on the Linuxquestions forum.) This is useful for monitoring changes in the storage use within large and growing archives of files, and appears to execute much more quickly than some options I have seen (use of a 'for SIZE in find-command -exec du' style approach, instead, for instance). I just ran it on a not particularly spectacular server, where a directory tree with over three thousand subdirectories, containing around 4000 files, of about 4 Gigs, total, responded in under a second. Show Sample Output
0find /path/to/archive/?/??/??? -mtime -7 -name "*.pdf" | xargs stat -c "%s"| awk '{sum +=$1}END{printf("%0.0f\n",sum)}'|sed -r ':Label;s=\b([0-9]+)([0-9]{3})\b=\1,\2=g;t Label' -
recursive find and replace. important stuff are grep -Z and zargs -0 which add zero byte after file name so sed can work even with file names with spaces.
0grep -ZlRr -e BAD_SCRIPT_LINE * |xargs -0 sed -i 's/BAD_SCRIPT_LINE//g' -
0ls *.JPG | cut -d . -f 1 | xargs -L1 -i convert -resize 684 {}.JPG {}.jpg
-
Kills all process that belongs to the user that runs it - excluding bash, sshd (so putty/ssh session will be spared). The bit that says grep -vE "..." can be extended to include ps line patterns that you want to spare. If no process can be found on the hitlist, it will print # NOTHING TO KILL. Otherwise, it will print # KILL EM ALL, with the cull list.
0ps -u $USER -lf | grep -vE "\-bash|sshd|ps|grep|PPID" > .tmpkill; if (( $(cat .tmpkill | wc -l) > 0 )); then echo "# KILL EM ALL"; cat .tmpkill; cat .tmpkill | awk '{print $4}' | xargs kill -9; else echo "# NOTHING TO KILL"; fi; cat .tmpkill; rm .tmpkill; - ‹ First < 13 14 15 16 17 > Last ›
What's this?
commandlinefu.com is the place to record those command-line gems that you return to again and again. That way others can gain from your CLI wisdom and you from theirs too. All commands can be commented on, discussed and voted up or down.
Check These Out
Stay in the loop…
- Follow the Tweets.
-
Every new command is wrapped in a tweet and posted to Twitter. Following the stream is a great way of staying abreast of the latest commands. For the more discerning, there are Twitter accounts for commands that get a minimum of 3 and 10 votes - that way only the great commands get tweeted.
» http://twitter.com/commandlinefu
» http://twitter.com/commandlinefu3
» http://twitter.com/commandlinefu10 - Subscribe to the feeds.
-
Use your favourite RSS aggregator to stay in touch with the latest commands. There are feeds mirroring the 3 Twitter streams as well as for virtually every other subset (users, tags, functions,…):
Subscribe to the feed for:
- » all commands
- » commands with 3 up-votes (commandlinefu3)
- » commands with 10 up-votes (commandlinefu10)
- » commands using xargs