Shell Script - various questions...

[ryanVickers]

For one, I have built this little script that finds any of those annoying text backup files ending in ~ in my home directory, and asks to remove them or not. My code so far is this:

#!/bin/bash
echo
echo "=================================="
echo "= Will now display all '~' files ="
echo "=================================="
echo
find $HOME -name "*~"
echo
echo "=================================="
echo "= End of list of '~' files ="
echo "=================================="
echo
echo -n "Delete these files? (y/n): "
read sure
if [ $sure == "y" ]
then
rm $list
echo
echo "================="
echo "= Files deleted ="
echo "================="
echo
elif [ $sure == "n" ]
then
echo
echo "===================================="
echo "= Script aborted - Files were kept ="
echo "===================================="
echo
else
echo
echo "======================="
echo "= Not a proper answer ="
echo "======================="
echo
fi
As you can see, it's not quite finished - it will ask to remove the files, but not be able to because there is nothing to delete! You'll notice it's trying to delete everything in "list". I want that 'list' to be everything that the 'find' command finds. How might I do this?

[netyire]

Thanks for the script!

[ryanVickers]

:lolflag:

any ideas, though!? :p

Oh, if your loking for more, this ones quite simple, but...


#!/bin/bash
echo
echo -n "enter filename part (case sensitive): "
read part
echo
echo "please wait - scanning home directory..."
echo
rm "("$part").findersearch" >/dev/null 2>&1
find $HOME -name "*"$part"*">>"("$part")".findersearch
cat $HOME"/("$part").findersearch"
echo
echo "=========================================="
echo "= Displaying all matches in /home/%user% ="
echo "=========================================="
echo
echo -n "Search for old finder searches?(y/n/x): "
read ans
echo "- - - - - - - - - - - - - - - - - - - - - - - - - - -"
echo
if [ $ans == "y" ]
then
find $HOME -name "*.findersearch"
echo
echo "========================================="
echo "= Displaying all search records ="
echo "========================================="
echo
echo -n "Remove? (y/n): "
read rem
if [ $rem == "y" ]
then
find $HOME -name "*.findersearch" -print0|xargs -0 /bin/rm -f >/dev/null 2>&1
echo
echo "========================================="
echo "= Old search records removed! ="
echo "========================================="
echo
else
echo
echo "========================================="
echo "= Search records were not removed ="
echo "========================================="
echo
exit 0
fi
elif [ $ans == "x" ]
then
find $HOME -name "*.findersearch" -print0|xargs -0 /bin/rm -f >/dev/null 2>&1
echo
echo "========================================="
echo "= Old search records removed! ="
echo "========================================="
echo
elif [ $ans == "n" ]
then
echo
echo "========================================="
echo "= Search 'yes' = false, Exiting now ="
echo "========================================="
echo
else
echo
echo "========================================="
echo "= Incorrect answer, Exiting now ="
echo "========================================="
echo
fi
you enter a part of a fileneme, and it will check if it exists and list all matches that exist (or not) int your home folder. It will also save the output to a text file in home by name of your search with extension "findersearch". Afterwards, you can enter n, y, or x to "search for residual search records; y means yes, then you can delete them if you want, n means no - skip, and x means just find and delete them, no conformation, no lists provided. It surrounds the search name in brackets so searches for an extension (.xml for example) don't become hidden! Just a little thought of mine!

[ryanVickers]

I made a counter that I question the efficiency of, but:


#!/bin/bash
echo
echo "At anytime, hit control + c to quit"
echo
echo "Please specify a filename (or path starting at home): "
read path
echo
echo "Enter number to count to (add 1 to desired number!): "
read num
echo
echo "Enter number to start at: "
read stt
COUNTER=$stt
echo
echo "Enter number to count by: "
read by
echo
echo "Counting from "$stt" to "$num" by "$by"'s NOW!"
while [ $COUNTER -lt $num ]; do
echo $COUNTER $1 1>>$HOME"/"$path
let COUNTER=COUNTER+$by
done
echo
echo "=============================================="
echo "= Counting Completed Successfully! ="
echo "=============================================="
echo "The file can be found at: "$HOME"/"$path
echo -n "The filesize and name is: "
du -h $path
echo

it asks for a file name, where to start, end, what to count by, and displays some finishing data like ~ file size and path to file!

[ryanVickers]

If your woundering about this post, check the first one for an actuall problem! :p

[Bluecircle]

lol, for the ~ backup files, do you know that you can turn the creation of them off in gedit preferences?

[FuturePast]

list=`find $HOME -name "*~"`
though the resulting list will not then be displayed.

[McNils]

list=`find $HOME -name "*~"`
though the resulting list will not then be displayed.

Note that files with spaces will not be deleted with aproach.

Do an other find after with a delete flag instead of rm $list.

find $HOME -name "*~" -delete

[ryanVickers]

Wow, thanks, this is really detailed! I thought there might be a way to turn them off, but ther stuff, like emerald and stuff, I've foud lots of them! Run this yourselves and see!
Ok, here's my most recent code:


#!/bin/bash
echo
echo "=============================================="
echo "= Will now display all '~' files ="
echo "=============================================="
echo
find $HOME -name "*~"
list=`find $HOME -name "*~"`
echo
echo "=============================================="
echo "= End of list of '~' files ="
echo "=============================================="
echo
echo -n "Delete these files? (y/n): "
read sure
if [ $sure == "y" ]
then
rm $list
echo
echo "=============================================="
echo "= Files deleted, unless there was an error ="
echo "= in which case there was nothing to delete! ="
echo "=============================================="
echo
elif [ $sure == "n" ]
then
echo
echo "=============================================="
echo "= Script aborted - Files were kept ="
echo "=============================================="
echo
else
echo
echo "=============================================="
echo "= Not a proper answer! - Files were kept ="
echo "=============================================="
echo
fi

I didn't use the -delete thing because it seems to just do it with out conformation of if I wanted them removed or not, and what I've got here seems to find them all.

[FuturePast]

#!/bin/bash
echo
echo "=================================="
echo "= Will now display all '~' files ="
echo "=================================="
echo
find $HOME -name "*~"
echo
echo "=================================="
echo "= End of list of '~' files ="
echo "=================================="
echo
echo -n "Delete these files? (y/n): "
read sure
if [ $sure == "y" ]
then
find $HOME -name "*~" -print0|xargs -0 /bin/rm -f
echo
echo "================="
echo "= Files deleted ="
echo "================="
echo
elif [ $sure == "n" ]
then
echo
echo "===================================="
echo "= Script aborted - Files were kept ="
echo "===================================="
echo
else
echo
echo "======================="
echo "= Not a proper answer ="
echo "======================="
echo
fi

[ryanVickers]

Of course! I can't beleive I didn't think of that?! :p Thakns, I thought there must be some better way...

[ryanVickers]

Now, I just made this (you have to install "pi" to use it)


#!/bin/bash
echo
echo -n "How many decimal places of pi would you like? "
read dec
echo
echo "Will now 'evaluate' pi to "$dec" decimal places..."
echo
pi $dec >>pi_to_$dec
echo "Finished - File can be found at" $HOME"/pi_to_"$dec
echo
there seems to be something wrong with it - large numbers seem to fill up my ram and crash it quite quickly. I tryed 1 million and it was done in about 2 seconds, so I tried 1 billion and it just exploded! :p Whats wrong?! Why would this take more than 1Mb of ram at most!!!???

Just had a thought - I think it's storing everything it generates in ram until ther very end! How do I make it write to disk after avery few numbers?

[ryanVickers]

I made another script along the "cleaner" line, it shows all the stuff in the trash and you cal clear it too. It's most useful for viewing everything in the trash without oping it and going show hidden files and all that.


#!/bin/bash
echo
echo "============================================="
echo "= Will now display Trash contents ="
echo "============================================="
echo
find $HOME"/.Trash/" -name "*"
echo
echo "============================================="
echo "= End of list of Trash contents ="
echo "============================================="
echo
echo -n "Empty the Trash? (y/n): "
read sure
if [ $sure == "y" ]
then
find $HOME"/.Trash/" -name "*" -print0|xargs -0 /bin/rm -f >/dev/null 2>&1
echo
echo "============================================="
echo "= Answer is yes - Trash was cleared ="
echo "============================================="
echo
elif [ $sure == "n" ]
then
echo
echo "============================================="
echo "= Script aborted - Trash was left alone ="
echo "============================================="
echo
else
echo
echo "============================================="
echo "= Not proper answer! - Trash was left alone ="
echo "============================================="
echo
fi
Quite similar. It may look like it will remove the trash folder, but it doesn't :)

Edit: I just realized it deletes the files, but it will not remove folders! :p And if there are files in folders, it sucks them out and leavs the folders behind! I thnk that's kinda cool :) - but it also makes it not a teribly good trash "emptier" ;)

I also updated my "finder search engine" (way) above :)

Once again, just above this post is an actuall problem insted of just posts of scripts :)