cheatsheets/command_line.md

title: Command line stuff

List (ls)

Usage:

ls [options] [paths]

Format:

-1   One entry per line
-l   Long view
-o   Long view (without groups)
-C   Multicolumn (sorted horizontally)
-x   Multicolumn (sorted vertically)

-F   Add / after directories
-G   Color

Options:

-R   Recurse
-a   Include hidden (dotfiles)
-A   Include hidden (but not . and ..)

Sorting:

-r   reverse order
-S   sort by size
-t   sort by time modified
-u   sort by time accessed
-U   sort by time created
-c   sort by time status was changed

-h   Human-readable size (3k)

Find (find)

Usage:

find <path> <conditions> <actions>

Conditions:

-name "*.c"

-user jonathan
-nouser

-type f            # File
-type d            # Directory
-type l            # Symlink

-depth 2           # At least 3 levels deep
-regex PATTERN

-newer   file.txt
-newerm  file.txt        # modified newer than file.txt
-newerX  file.txt        # [c]hange, [m]odified, [B]create
-newerXt "1 hour ago"    # [t]imestamp

Condition flow:

\! -name "*.c"
\( x -or y \)

Actions:

-exec rm {} \;
-print
-delete

Examples:

find . -name '*.jpg'
find . -name '*.jpg' -exec rm {} \;

find . -newerBt "24 hours ago"

Tail

Usage:

tail [-F | -f | -r] [-bN | -cN | -nN] [file ...]

Modes:

-f     # follow
-F     # follow by filename (accounts for log rotation)
-r     # Reverse order

Options:

-bN    # N*512 bytes
-cN    # N bytes
-nN    # N lines

+N     # Start from line N

Sudo

Listing:

-l           # List allowed commands

Options:

-A           # Use $SUDO_ASKPASS
-b           # Run in background
-E           # Preserve environment
-H           # use target's $HOME
-n           # Don't prompt for password
-P           # Preserve group vector
-S           # Read password from stdin

File descriptors:

-C fd        # Close all open file descriptors

Prompt:

-p prompt    # Custom prompt (-p "%p password:")

Interactive:

-i [cmd]     # Interactive shell without variables
-s [cmd]     # Interactive shell

-u user      # run as this user
-g group     # run as this group

Timestamp:

-v           # revalidate timestamp for 5 mins
-k           # invalidate timestamp
-K           # just like -k

wc (Word count)

-c   # Bytes
-l   # Lines
-m   # Characters (incl multi-byte)
-w   # Words

Search-and-replace in all files

perl -p -i -e 's/hello/HELLO/g' **/*

Grep

-A num      # Print `num` lines of training context

-G          # --basic-regexp (default)
-E          # --extended-regexp
-P          # --perl-regexp

-f file     # --file (Get patterns for file)
-F          # --fixed-strings

-h          # --no-filename
-H          # --with-filename

-l          # --files-with-matches (just print filenames)
-L          # --files-without-match

-r, -R      # --recursive
-v          # --invert-match
-i          # --ignore-case

Synonyms:

egrep  =>  grep -E
fgrep  =>  grep -F