awk -v search="$search_string" '$0 ~ search { gsub(/"/, "", $5); print $1" "$5; }' "$filename" | while read line do result=$(awk -v search="$line" '$0 ~ search { print $3; } ' "$lookup_file"); # Do "something" with $result done

you can try: find $WORK.../.history -type f -printf '%T@ %p ' | sort -nr | cut -f2- | xargs grep 'your_pattern' Decomposed: the find finds all plain files and prints their modification time and path the sort sort sort them numerically - and reverse, so highest number comes first (the latest modified) the cut removes the time from each line the xargs run its argument for each file what get to it input, in this case will run the grep command, so the 1st file what the grep find - was the lastest modified The above not works when the filenames containing spaces, but hopefully this is not your case... The -printf works only with GNU find. For the repetative work, you can split the command to two parts: find $WORK.../.history -type f -printf '%T@ %p ' | sort -nr | cut -f2- > /somewhe

I tried yesterday and I am very close to the solution. Researching the internet , I made a perl script and was able to create multiple interface by extracting data from csv file. But there is small issue I see and not able to figure out why an extra character is coming in the file. Following is the code: #!/usr/bin/perl use strict; use warnings; sub main { # Note: this could be a full file path my $filename = "file1.csv"; open(INPUT, $filename) or die "Cannot open $filename"; # Read the header line. #my $line = <INPUT>; my $line; # Read the lines one by one. while($line = <INPUT>) { chomp($line); # Display the header, just to check things are working.

The underscore is a valid character in variable names. Do: ./"$f" > "${f}_july_15.txt" Your code was trying to use a variable named f_july_15. You need to use braces to delimit the variable name when it's followed by a character that can be part of a variable name.

You're actually deliberately turning off parameter subsitution by enclosing 'ENDFILECONTENT' in quotes. See this excerpt from example 19-7 of the advanced Bash scripting guide on Heredocs, slightly reformatted: # No parameter substitution when the "limit string" is quoted or escaped. # Either of the following at the head of the here document would have # the same effect. # # cat <<"Endofmessage" # cat <<Endofmessage Remove the single quotes around 'ENDFILECONTENT' and BASH will replace the variables as expected.

Always quote your variable substitutions. I.e. not cp $source $target, but cp "$source" "$target". This way they won't be subject to word splitting and pathname expansion. Specify "--" before positional arguments to file operation commands. I.e. not cp "$source" "$target", but cp -- "$source" "$target". This prevents interpreting file names starting with dash as options. And yes, "/" is not a valid character for file/directory names.

I think this little script I hacked together does the trick. If not, it's definitely a starting point! Good luck. #!/bin/bash ############################ ## Methods

if your file contains only domain and email addr, try this line: grep -v '@' file test with your example: kent$ echo "john3525@aol.com brenda392@yahoo.com xyx.net boxers.on.ca helloworlds.co.uk"|grep -v '@' xyx.net boxers.on.ca helloworlds.co.uk

VAR="INPUTFILENAME" # One solution this does not use the VAR: touch INPUTFILENAME{1,2,3,4,5,6,7,8,9,10} # Another for i in `seq 1 20` ; do touch "${VAR}${i}" done And there are several other ways.

I don't know about the syntax error, but this should work: SERVERS="blah1.example.com blah2.example.com blah3.example.com blah4.example.com" for server in $SERVERS do echo $server done EDIT: As noted by Jonathan Leffler in a comment, maybe you are not running the script with bash. Other shells, such as dash, may not recognize the array syntax. If that's the case, you can do: SERVERS=(blah1.example.com blah2.example.com blah3.example.com blah4.example.com) for i in $(seq 0 3) do echo ${SERVERS[$i]} done But if you just want to loop through the names and run an SSH command (ie if having an array won't provide useful functionality), the first method is more straightforward.

An approach would be to have 'template file' in your installation folder like: ex. database.php $active_group = 'default'; $active_record = TRUE; $db['default']['hostname'] = '<?php echo $hostname;?>'; $db['default']['username'] = '<?php echo $username;?>'; $db['default']['password'] = '<?php echo $password;?>'; $db['default']['database'] = '<?php echo $database;?>'; $db['default']['dbdriver'] = 'mysql'; $db['default']['dbprefix'] = '<?php echo $prefix;?>'; $db['default']['pconnect'] = TRUE; $db['default']['db_debug'] = TRUE; $db['default']['cache_on'] = FALSE; $db['default']['cachedir'] = ''; $db['default']['char_set'] = 'utf8'; $db['default']['dbcollat'] = 'utf8_general_ci'; $db['default']['swap_pre'] = ''; $db['default']['autoinit'] = TRUE; $db['default']

Perhaps you could use php like this: running php script (php function) in linux bash And then you something like domdocument(http://php.net/manual/en/class.domdocument.php) to read and the right the xml. Of course this is assumes you have php installed.

You need one more <: ./script -a <<<"$CONTENT" <<< is called a herestring, and takes the following string and passes it as the standard input.

Save this to a file e.g. script.sh. #!/bin/bash declare -A A while read -ra __; do A[${__[0]}]=${__[1]} done < "$2" while read -r __; do for I in "${!A[@]}"; do [[ $__ == *"$I"* ]] && { V=${A[$I} __=${__//"$I"/"$V"} } done echo "$__" done < "$1" Then run bash script.sh aaa.txt bbb.txt. To get the output run it as something like bash script.sh aaa.txt bbb.txt > out.txt I prefer to use Bash this time since you have to quote non-literals in Awk. Another solution through Ruby: ruby -e 'f = File.open(ARGV[0]).read; File.open(ARGV[1]).readlines.map{|l| l.split}.each{|a| f.gsub!(a[0], a[1])}; puts f' aaa.txt bbb.txt Another way to modify file directly: #!/bin/bash A=$(<"$1") while read -ra B; do A=${A//"${B[

# with /bin/bash . <(awk '/define/ {print $2"="$4}' FS="'" foo.php) # with /bin/sh declare `awk '/define/ {print $2"="$4}' FS="'" foo.php` Result echo $DB_SERVER # localhost Example

Use LaunchOnlyOnce key in plist file. LaunchOnlyOnce : This optional key specifies whether the job can only be run once and only once. In other words, if the job cannot be safely respawned without a full machine reboot, then set this key to be true.

If "file.sh" is not executable then ./file.sh will not work but bash file.sh will. If "file.sh" does not start with the line #!/bin/bash (or another path to a valid bash interpreter) then ./file.sh will not work but bash file.sh will. Basically, in order for a script to look like an executable file it must: Have execute permission. Start with the line #!/path/to/interpreter.

Honestly, IMO something in your shell configuration is breaking things up. Try doing this to verify it: zsh -f # starts a new shell ignoring your configuration autoload compinit compinit ./my-shell-script [TAB] it completes with files. That is the default. FWIW, if you want to bind a particular completer to a command/alias etc, you can do compdef _jstack jstack # simple _files completion compdef _files my-local-python-script # restrict to some file extensions compdef '_files -g "*.(eps|ps|pdf)"' okular

When i had to do something like that, i used expect and a wrapper script that would fetch a password from a file. I.e. in my password file i'd have something like root@192.168.1.10 ThisIsMyPass user@localhost thisIsMyOtherPass and then have the wrapper script get (it could be simple as grep "root@192.168.1.10" ~/.passwords | cut -d ' ' -f2) Im guessing there are more appropriate methods, but with this one you only need to keep your wrapper and password file protected and you can make your setup script public.

With bash if your lines are in order: while read WHEN && read GXCOORD; do echo "$WHEN : $GXCOORD" done < test.kml Or perhaps do it this way: while read WHEN && read GXCOORD; do echo "$WHEN : $GXCOORD" done < <(exec grep -Fe "<when>" -e "<gx:coord>" test.kml) And also perhaps trim out the tags: while read WHEN && read GXCOORD; do WHEN=${WHEN##*'<when>'} WHEN=${WHEN%%'</when>'*} GXCOORD=${GXCOORD##*'<gx:coord>'} GXCOORD=${GXCOORD%%'</gx:coord>'*} echo "$WHEN : $GXCOORD" done < <(exec grep -Fe "<when>" -e "<gx:coord>" test.kml)

Your code is looking for settings.cfg in its current working directory. This working directory will not be the same when cron executes the job, hence the error You have two "easy" solutions: Use an absolute path to the config file in your script (/home/tomi/amaer/config.cfg) CD to the appropriate directory first in your crontab (cd /home/tomi/amaer/ && python /home/tomi/amaer/controller.py) The "right" solution, though, would be to pass your script a parameter (or environment variable) that tells it where to look for the config file. It's not exactly good practice to assume your config file will always be lying just next to your script. You might want to have alook at this question: http://unix.stackexchange.com/questions/38951/what-is-the-working-directory-when-cron-exe

You could modify the line with sed. catcenter.dat sed -i "/CenterListName/ s/=$/= $(cat center.dat)/" AnalyzeClusterParameterFile$c Read the value first and add it to the second file with sed.

Not precisely. -N does a comparison between the file's atime and mtime, which is not accurate on e.g. ext3 filesystems that are mounted relatime. You should either use the OS's file monitoring facilities or compare the mtime of the file directly.

while read -r account abelpass entapass You are reading abelpass & entapass but passing password1 and password2 ! Didn't you meant to use while read -r account password1 password2 ?

You are not passing the actual file to compile - instead you're passing the .java file as the classpath. Your actual command looks something like this: javac -classpath /some/dir/Program.java while it should be something like this: javac -classpath /path/to/library.jar /some/dir/Program.java I think you're missing the actual classpath somewhere. Most likely your script needs to be similar to this: #!/bin/bash CWD=$(pwd) CLSPATH=/path/to/library.jar SOURCE="$CWD/Program.java" COMMAND="javac -classpath $CLSPATH $SOURCE" eval $COMMAND java Program Also, I'm not sure what the purpose of the eval approach is. Why not just invoke javac? I would further modify the script to be: #!/bin/bash CWD=$(pwd) CLSPATH=/path/to/library.jar SOURCE="$CWD/Program.java" javac -classpath $CLSPATH $

Pure bash based solution: for x in *.pickle; do [[ $x =~ ^([^_]+_[^_]+_)([^_]+)(_.+)$ ]] && mv "$x" "${BASH_REMATCH[1]}$((${BASH_REMATCH[2]} + 1))${BASH_REMATCH[3]}" done

This basic trick may work in your case: echo -e "y y y y n" | ./script or echo -e "yes yes yes yes no" | ./script but it depends on how your script handles the input stream...

You can remove files from the argument list as you use them, with the shift command: if (( $# % 2 )); then echo Please enter an even number of filenames exit 1 fi while (( $# )); do src=$1 dst=$2 cp "$src" "$dst" shift 2 done

I would use cat together with here-doc syntax for this: cat <<EOF > /etc/nsswitch.conf group: compat shadow: compat hosts: files mdns4_minimal [NOTFOUND=return] dns mdns4 ... EOF The statement above will overwrite or create the file with the contents between the first line and EOF. In the form above even variables like group: $group would be expanded by bash. If you don't want this, then use <<'EOF' (note the single quotes ' around the EOF)

try this: awk '/localhost> localhost>/{print $1 >"file2";next}{$2=$3=$NF=""}7' file this will output text you want, and generate a "file2" for the localhost> localhost> case. didn't test, I hope there is no typoes.

Looks to me like those sed commands arguments are not on the same line and line shift is not escaped. Try: sed -e "s/$SEEDVALUE/$RANDOMVAR/" -e "s/$EXPIREMENT/$QUAN/" -e "s/$DISTANCE/$DIST/" -e "s/$OUTPUTFILE/$outputText/" < $TEMPLATEFILE > $outputFile Also, to get more debugging information, it's often useful to use the following as interpreter: #!/bin/bash -x

It's because this plugin is for old version of spree which was for Rails 2.x and this command syntax is deprecated by now. I think it would be better for you to use something more up-to-date, like https://github.com/nebulab/spree-subscriptions.

The following would give you a list of packages for which you want the reports: grep "^packages" config.txt | cut -d= -f2 | tr ',' ' ' Based on this, you can loop for values in the list: filename="config.txt" for i in $(grep "^packages" $filename | cut -d= -f2 | tr ',' ' '); do for file in $(find /home/user/ftpuser -maxdepth 1 -name "*.[ew]ar" -type f); do echo /apps/oracle/jrockit/4.1.0-1.6.0_37-R28.2.5-x86_64/bin/java -jar ../windup-cli-0.6.8/windup-cli.jar -javaPkgs ${i} - input ../ftpuser/ -output ../reports/ "${file}" cp "${file}" /home/user/ftpuser/scanned/ done done

Change: if ["$(( $(date +"%s") - $(stat -c "%Y" $i) ))" -gt "259200" ]; then to: if [ "$(( $(date +"%s") - $(stat -c "%Y" $i) ))" -gt "259200" ]; then Spaces around the [ are required.

If you have saved your patterns in the file /root/my-wordfile, then you can use the following to find "all" the files under /home grep -rf /root/my-wordfile /home -r would search it recursively and -f would say the patterns are in /root/my-wordfile(separated by newline) Or if you want to search files with specific extension or names, you could use find as below: For example if you want to search all *.c files find /home -name *.c -exec grep -Hnf /root/my-wordfile {} ; Here -H would list the file name, -n would display line number.

If the echo prints ${Upper} blank, probably it contains only a space or a tab, change the if this way if [[ "${Upper}" =~ '^[[:blank:]]*$' ]]; then echo "VAR is empty" fi

Are you looking for this?: $ ls -1 INFO1 - INFO2 - INFO3.doc INFO1 - INFO4 - INFO6.doc l.sh $ ./l.sh <li> <h5>INFO1</h5> <p>INFO2</p> </li> <li> <h5>INFO1</h5> <p>INFO4</p> </li> The script: #!/bin/bash for i in *.doc do # split file-name into array IFS='- ' read -a array <<< "$i" echo "<li>" echo "<h5>${array[0]}</h5>" echo "<p>${array[1]}</p>" echo "</li>" done update If filename looks like this: First name blabla - Second part bla boa - third part, unimportant.doc Change the IFS-parameter to a single - instead. Make this change: IFS='-' read -a array <<< "$i" output: $ ./l.sh <li> <h5>First name blabla &l

this awk one-liner should do : awk -F'$' '{x+=$2+0}7;END{print "Total: "x}' file it prints the whole file, and add a line Total: 98 at the end.

After months of looking for an answer I have finally found the solution. It was in a comment on an answer in some other thread I can't even remember. Hope this can help others out there. Your bash script should look like this and will connect to the sftp server, prompt the user for the password, and then execute the remaining commands. #!/bin/bash sftp user@server <<! cd /the/remote/directory lcd /your/local/directory put/get some.file !

Not quite sure whether this helps or not: This calls the program script five times: for n in 1 2 3 4 5 do ./program $n input.ini done Then in program, refer to the first parameter $n as $1. The second parameter input.ini is $2.