Keep only certain parts of a text file

[bcooperizcool]

Ok, you were very helpful with my last question, (I think this ok to post as a new topic...), but how do I keep certain parts of a text file? Eg.


/var/mobile/Applications/324-sfdg/App.app
/var/mobile/Applications/324-sfdg/App.app/blah.extension
/var/mobile/Applications/324-sfdg/App.app/icon.png
/var/mobile/Applications/dsfg-234/OtherApp.app/
/var/mobile/Applications/dsfg-234/OtherApp.app/title.png
/var/mobile/Applications/dsfg-234/OtherApp.app/idk.extension

I want to only keep the lines that begin with "/var/mobile/Applications" and end with ".app". How would I do this? (this will be inside a bash script) Also would be nice to remove the extra lines.

Thanks again for all your help so far! :D

[worksofcraft]

...
I want to only keep the lines that begin with "/var/mobile/Applications" and end with ".app". How would I do this? (this will be inside a bash script) Also would be nice to remove the extra lines.

Thanks again for all your help so far! :D

can you not use grep for this ?
something like:

grep "/var*.app"

[bcooperizcool]

How would I use that for my text file?


grep "/var*.app" < file.txt > newfile.txt

?
or am I missing something?
Thanks!

[worksofcraft]

How would I use that for my text file?


grep "/var*.app" < file.txt > newfile.txt

?
or am I missing something?
Thanks!

You could do it that way.
You can also use sed.

Actually my regular expression there isn't quite right... you wanted to have the /var at the beginning of line and the .app at the end so it be more like:


grep "^/var/mobile/Application.*\.app$" file.txt > newfile.txt

^ means beginning of line
$ means end of line
. means anything
* means any number of times
\. means .

hope that helps :)

[bcooperizcool]

Thank you! :D (The explanation helped as well).
How can you do that with sed? I know how to remove, but not keep.

[ghostdog74]

Thank you! :D (The explanation helped as well).
How can you do that with sed? I know how to remove, but not keep.

there are many other ways to do it

awk


awk '/^\/var\/mobile\/Application.*\.app$/' file


sed


sed -n '/^\/var\/mobile\/Application.*\.app$/p' file


ruby


ruby -ne 'print if /^\/var\/mobile\/Application.*\.app$/' file


even just using the bash shell(3.2++)



while read -r line
do
if [[ $line =~ "^\/var\/mobile\/Application.*\.app$" ]];then
echo $line;
fi
done < file

[bcooperizcool]

Thanks! I think that covers it for this question! Until my next question, thanks!

[worksofcraft]

Thanks! I think that covers it for this question! Until my next question, thanks!

YW :)
but don't forget to use "thread tools" to mark the thread as solved ;)