This does the following:

1 - Search recursively for files whose names match REGEX_A

2 - From this list exclude files whose names match REGEX_B

3 - Open this as a group in textmate (in the sidebar)

And now you can use Command+Shift+F to use textmate own find and replace on this particular group of files.

For advanced regex in the first expression you can use -regextype posix-egrep like this:

mate - `find * -type f -regextype posix-egrep -regex 'REGEX_A' | grep -v -E 'REGEX_B'`

Warning: this is not ment to open files or folders with space os special characters in the filename. If anyone knows a solution to that, tell me so I can fix the line.

This command uses the recursive glob and glob qualifiers from zsh. This will remove all the empty directories from the current directory down.

The **/* recurses down through all the files and directories

The glob qualifiers are added into the parenthesis. The / means only directories. The F means 'full' directories, and the ^ reverses that to mean non-full directories. For more info on these qualifiers see the zsh docs: http://zsh.dotsrc.org/Doc/Release/Expansion.html#SEC87

this will show the names of the deleted directories, and will delete directories that only no files, only empty directories.

Find Word docs by filename in the current directory, convert each of them to plain text using antiword (taking care of spaces in filenames), then grep for a search term in the particular file.

(Of course, it's better to save your data as plain text to make for easier grepping, but that's not always possible.)

Requires antiword. Or you can modify it to use catdoc instead.

By putting the "-not \( -name .svn -prune \)" in the very front of the "find" command, you eliminate the .svn directories in your find command itself. No need to grep them out.

You can even create an alias for this command:

Now you can do things like

using mb it's still readable;) a symbol variation

$ du -ms {,.[^.]}* | sort -nk1

Taken from http://www.earthinfo.org/linux-disk-usage-sorted-by-size-and-human-readable/

somewhat faster version to see the size of our directories. Size will be in Kilo Bytes. to view smallest first change '-k1nr' to '-k1n'.

This command gives you the number of lines of every file in the folder and its subfolders matching the search options specified in the find command. It also gives the total amount of lines of these files.

The combination of print0 and files0-from options makes the whole command simple and efficient.

Makes any files in the current directory (and any sub-directories) group-readable.

Using the "! -perm /g=r" limits the number of files to only those that do not already have this property

Using "+" on the end of the -exec body tells find to build the entire command by appending all matching files before execution, so invokes chmod once only, not once per file.

Obviously, you can replace 'man' command with any command in this command line to do useful things. I just want to mention that there is a way to list all the commands which you can execute directly without giving fullpath.

Normally all important commands will be placed in your PATH directories. This commandline uses that variable to get commands. Works in Ubuntu, will work in all 'manpage' configured *nix systems.

"find . -type d -print0 | xargs -0 chmod 755"

thanks masterofdisaster

With this command you can get a previous or future date or time. Where can you use this? How about finding all files modified or created in the last 5 mins?

touch -t `echo $(date -d "5 minute ago" "+%G%m%d%H%M.%S")` me && find . -type f -newer me

List all directories created since last week?

touch -t `echo $(date -d "1 week ago" "+%G%m%d%H%M.%S")` me && find . -type d -cnewer me

I'm sure you can think of more ways to use it. Requires coreutils package.

unzips all zip files in any subdirectory under the current directory. The zip files are unzipped in their respective subdirs

In the example, uid 0 is root. foo:foo are the user:group you want to make owner and group. '.' is the "current directory and below." -print0 and -0 indicate that filenames and directories "are terminated by a null character instead of by whitespace."

This uses Bash's "process substitution" feature to compare (using diff) the output of two different process pipelines.

The "find" command can be annoying when used inside of a Subversion (or CVS) working directory. Obviously, you can combine this with other predicates and commands to create a more elaborate pipeline:

Note: You can use my "dont-go-there.sh" script to wrap the "find" command and do this automatically at http://forwardlateral.com/blog/2006/02/27/dont-go-there/