Skip to content

Latest commit

 

History

History
76 lines (53 loc) · 2.48 KB

README.md

File metadata and controls

76 lines (53 loc) · 2.48 KB
Raw

Description

The jump utility can be used to assign convenient names to frequently used strings, typically directory paths. The related j shell command can then be used to cd to a named path. In combination, these utilities make it very easy to jump from one place to another on the command line.

A few simple examples:

# Name a few paths.
jump --add docs $HOME/my/favorite/docs
jump --add mp3s $HOME/my/music/from/70s/80s/90s/and/today
jump --add words /usr/share/dict/words

# Go to docs directory and do some work. If desired, jump can
# be used in command-interpolation, as shown in the grep example.
j docs
grep ^aa `jump words` > words_starting_with_aa

# Easily cd somewhere else.
j mp3s

Usage overview:

jump --add NAME STRING [--literal]   # Add a named path or literal string
jump --rm NAME                       # Delete a named string
jump --mv OLD_NAME NEW_NAME          # Change name of a string

jump NAME                            # Print a named string
jump                                 # List all named strings

jump --help                          # Print help and exit

j NAME                               # cd to a named path

The jump utility will add paths in their normalized, absolute form and will handle common shell conventions (., .., and ~). Use the --literal option to bypass the default path normalization.

Named paths are stored in .jumprc in the user's home directory.

Installation

The script's only dependency is Python 2.7 or higher.

Download the jump script to a directory in your PATH and make it executable. For example:

# Modify the path to the jump script as desired.
URL='https://raw.githubusercontent.com/hindman/jump/master/jump'
curl "$URL" -o $HOME/bin/jump
chmod u+x $HOME/bin/jump

Add a function like this to your Bash profile:

# cd to a named path.
j() {
    [[ $# -eq "1" ]] && cd "$($HOME/bin/jump --cd "$1")"
}

For command-completion, also add this:

_jcomplete() {
    # Use `jump` to write matching completions to a temp file,
    # one completion-entry per line.
    local curr_word=${COMP_WORDS[COMP_CWORD]}
    local tmp_file="$(mktemp /tmp/jump-completions.XXXXXX)"
    jump --complete "$curr_word" >| "$tmp_file"
    # Load that temp file into the COMPREPLY array, and clean up.
    IFS=$'\n' read -d '' -r -a COMPREPLY < "$tmp_file"
    rm -f "$tmp_file"
}

complete -F _jcomplete j