Search Results

Single bracket test matches strings not patterns. Use a case statement which allows pattern matching.

case $col2 in
*/*) do stuff
do more stuff
;;

*\"*) ...

This will place the window of the new instance of the Terminal.app in front.

open -F -n -b com.apple.Terminal

Just exclude the .Trash directory in your find statement.

find "$HOME" -path '*.Trash*' -prune -o \( -name sox -print \)

I would go though all these files and quote your variables.

NFSv4 ACLs can to the job but that depends on whether your OS supports them.

Use a case statement:


read -r -p "Is this correct? [yes/No] " response
case $response in
[yY] | [yY][Ee][Ss] )
do something
;;

[nN] | [n|N][O|o] )
echo...

What have you tried?

Use ed:


for file in *.txt
do
printf "%s\n" 1i "geneID $file" . w | ed -s "$file"
done

I think you are looking for something like this

x=0
while [ "$x" -le 12 ]
do date +'%Y%m' -d "$x month ago"
((x++))
done


Though this doesn't assign values to a variable. Does your...

Enclose the variable in strong and weak quotes.

sed '/'"$DB_Name"'\;/d' FILE

Using RE's search for the backlash character then the following n character

sed 's/[\][n]/,/g' filename

Why not
awk '/host/{print $0"\n"}' host.properties
or in sed
sed '/host/!d;G' host.properties

Of course there is no need to match a pattern if the file only needs double spacing.

To double space the file:

sed G FILE

I counted 10 variable that are not defined.


if [ ! -d $Dest ]
then
mkdir -p $Dest
fi


This test will fail if the last name in the path of Dest resolves to another file type. I would...

If I understand your question correctly then a good starting point would be

man hier

A variation of RudiC's solution that should work in OS X. If your file names have spaces in the name then

awk 'FNR==3 && /[a-z]*-[a-z]*-[a-z]*-/ {printf"%s\n", FILENAME}' file* | xargs -I {} echo...

You should read the INSTALL and README files before you attempt to compile open source software. You are missing the dependency lib.ICE- configure: error: Can't find libICE because configure cannot...

Phil Harvey's exiftool is a great tool for reading and writing metadata. It requires that you have perl installed. You may need to compile exiftool ( I'm not sure whether compiled binaries are...

I'm assuming that you are using a Bourne like shell. You could use parameter expansion.

filename=20150217-v1-PH_THOMSON_Reuters_Bangkok_Conceptual_Ready_P20_170215

filename=${filename%??_*} #...

I'm going to suggest that you start by editing your post. You should use proper capitalization and punctuation. Write your requests in complete sentences and check your grammar.

As a "one liner" this would look like scrambled eggs.

awk '/AA|BB|CC/ {fcount++}
/XY|RS|PQ/ {scount++}
END { printf("%s\n%s\n",
"Type1 = " fcount, "Type2 = " scount)
...

Try the following->

sed '/.*_id .*/s/double/string/' FILE

You somewhat "jump through hoops" with csh. The following should work (tested in tcsh).

set v=`cat file`
set i=1
while ( $i < = $#v )
echo $v[$i]
@ i = $i + 1
end

Using BSD sed: substitute the carrage return character with nothing; terminate the last line of the file with the new-line character; remove any blank lines and save the results in a new file.

...

echo -n "${var1}," >> log_file
.
.
echo -n "${var2}," >> log_file
.
.

or

printf "${var1}," >> log_file
.
.
printf "${var2}," >> log_file