explanation for: find . -type f -name "*.xml" -exec grep -i "some text" {} \; -print

[tomerbd]

Hi,

I use

find . -type f -name "*.xml" -exec grep -i "some text" {} \; -print

in order to search with decendants folders for files with this content.
however i dont understand this command.
why exec and not pipe | ?
what is the {}
{} \;
how come the print knows to print the resolutions? why does the print has a - before it?

Thanks for anyone who could help me.

[Rubberman]

Hi,

I use



in order to search with decendants folders for files with this content.
however i dont understand this command.
why exec and not pipe | ?
what is the {}
{} \;
how come the print knows to print the resolutions? why does the print has a - before it?

Thanks for anyone who could help me.

1. The {} is a symbol that is replaced with each file found that matches the -name argument.
2. The -name argument tells find to only match the files found with the name pattern. BTW, you need to use single quotes and not double quotes as shown. Ie, -name '*.xml' and not -name "*.xml"
3. I would assume that you just want a list of files that contain the string specified by grep? Then you should execute the command like this: find . -type f -name '*.xml' -exec grep -i -l "some text" {} \;
4. The -exec tells find to execute the specified command for each file that matches -name 'pattern', replacing the {} with each file name.

Doing it the way I show above, you will get a list of the .xml files that contain the string you are searching for, ignoring the upper/lower case differences. That's the -i argument to grep. The -l argument just tells grep to output the file name if the pattern is found in the file. I use this construct very frequently to find files with a specific pattern in an entire directory tree.

[tomerbd]

Thanks alot that was very helpful.
Isn't pipe used to take the output of one command as input to another? so why do we use exec and not pipe to pipe the output of the find into the grep?

[Rubberman]

Thanks alot that was very helpful.
Isn't pipe used to take the output of one command as input to another? so why do we use exec and not pipe to pipe the output of the find into the grep?

Because find only returns file names, normally. Piping the output from find to grep would be like using the -name argument in find. IE, "find . -type f -name '*.xml'" is more or less the same as "find . -type f | grep '\.xml$'". Since you want to search within each file, you need to run grep against each candidate file, hence the -exec argument to find.