Bash has very powerful job control with its Ctrl-Z, fg and bg
commands. A lesser known (but equally powerful) feature is disown. disown
will detach a process from its controlling terminal, allowing it to continue
running after you have exited your terminal.
cursor: pointer;
git add -N file
This lets git know that you intend to add a file at some later point.
This makes the content of the file show up in ie. git diff
Lets you stage only parts of the changes made to a file for committing, instead of the entire contents.
Git add --patch FILE
Responses:
y yes; stage this change
n no; don't stage this change
s split; cut the change in smaller parts
e edit; manually define the part to split
VBoxManage setextradata global GUI/Customizations noMenuBar,noStatusBar
kwriteconfig --file kdeglobals --group Icons --key Theme gnome
kwriteconfig --file kdeglobals --group General --key widgetStyle gtk
gconftool -t boolean -s /apps/gksu/sudo-mode true
sudo usermod -a -G GROUP [USER]
Group is 'sudo' on Debian, 'wheel' on Mandriva
Usually speical tools like gksudo does the job fine. However, if that's not
available (or broken) you can use sudo XAUTHORITY="$XAUTHORITY"
DISPLAY="$DISPLAY" [command] instead.
cron has an @reboot "time" that you can use, which makes the command run
once during boot. This is very handy for running commands after a reboot as a
user (in particular if you don't have root access and thus can't edit
/etc/rc.local). Example (in your crontab - crontab -e):
@reboot /home/myuser/initCommand
apt-get and aptitude can't install local packages, and installation with
dpkg -i will not install any additional dependencies (requiring an additional
apt-get install -f afterwards to resolve the deps).
The gdebi tool can install a local .deb package and resolve, download and
install any dependencies (and in case you're wondering, gdebi is not a
graphical tool, that's gdebi-gtk, gdebi's gtk front-end).
Use the --sparse parameter for rsync. This will let rsync handle
those files efficiently. Note however that --sparse can not be used
with --inplace.
netselect - find the fastest server from a list of servers
unshare - remove access to i.e. networking for a single process
nethogs - simple network statistics
mu provides a full-text index of maildirs, and can
easily be used from within mutt. The tool itself is available
from http://www.djcbsoftware.nl/code/mu/ (in Debian just
install maildir-utils).
I use the following cronjobs to generate the index
# Update mail index every two hours
07 */2 * * * mu index --maildir=~/mail/ --quiet
# Remove old search results every morning
11 5 * * * rm -rf "/home/zerodogg/.mu/results"
And the following code inside of mutt, F7 then lets you enter a
search query, and F8 will go to the search results.
macro index <F7> "<shell-escape>mu find -c -l ~/.mu/results " "mu find"
macro index <F8> "<change-folder-readonly>~/.mu/results\n" "display mu find results"
perl -MData::Dumper -MTerm::ReadLine -e '$r = Term::ReadLine->new(1);while(defined($_ = $r->readline("\$ "))){$ret=Dumper(eval($_));$err=$@;if($err ne ""){print $err;}else{print $ret;}}'
Alias (ie. for use in .bashrc):
alias 'perl-repl'='perl -MData::Dumper -MTerm::ReadLine -e '\''$r = Term::ReadLine->new(1);while(defined($_ = $r->readline("\$ "))){$ret=Dumper(eval($_));$err=$@;if($err ne ""){print $err;}else{print $ret;}}'\'''
kill -CHLD <PID>
set t_md=