README.md

Pipeable JavaScript - another utility like sed/awk/wc... but with JS! Quickly filter, map and reduce from the command line. Features a streaming API. Inspired by pipeable ruby.

Overview

pjs is a cli tool that can accept input on stdin, or read from a list of files. Its filter, map and reduce options take expressions to be run, in that order, and applies them to the supplied input. The expressions themselves can contain identifiers used by keys in String.prototype, which will automatically be bound to the given line. This lets you save a bit of typing with your one-liners, while still giving you access to all your JS string functions! Check out some of the examples below to see how they translate.

# Return all lines longer than 5 chars
# => lines.filter(function(line) { return line.length > 5; });
ls -1 | pjs -f 'length > 5'

# Count characters in each line
# => lines.map(function(line) { return line.length; });
ls -1 | pjs -m 'length'

# Uppercase and pad each line
# => lines.map(function(line) { return '  ' + line.toUpperCase()"; });
ls -1 | pjs -m '"  " + toUpperCase()'

# Return lines longer than 5 chars, and remove any digits
# => lines
#      .filter(function(line) { return line.length > 5; })
#      .map(function(line) { return line.replace(/\d/g, ''); });
ls -1 | pjs -f 'length > 5' -m 'replace(/\d/g, "")'

The current line and value can also be accessed via the $ variable, and the tool supports json output.

(echo 'foo' && echo 'foobar') | pjs -jm '{name: $, length: length}'
[
{"name":"foo","length":3},
{"name":"foobar","length":6}
]

pjs also includes lodash functions, which can be accessed via the _ object, and chained using $$

echo 'hello' | pjs -m '_.upperFirst($)'
# Hello

echo 'please-titleize-this-sentence' | \
pjs -m '$$.lowerCase().split(" ").map(_.upperFirst).join(" ")'
# Please Titleize This Sentence

as well as Ramda and point-free style

echo 'please-titleize-this-sentence' | \
pjs -m "R.compose(R.replace(/(^|\s)\w/g, R.toUpper), R.replace(/-/g, ' '))"
# Please Titleize This Sentence

Installation

It can be installed via npm using:

npm install -g pipeable-js

Usage

Usage: pjs [options] [files ...]

Functions and expressions are invoked in the following order:
filter, map, reduce

All functions are passed the line ($) and index (i)
Built-in reduce functions: length, min, max, sum, avg, concat
Custom reduce expressions accept: prev, curr, i, array
Includes lodash (_), and can be chained using $$
Supports Ramda (R) and point-free style

Options:

  -h, --help               output usage information
  -V, --version            output the version number
  -i, --ignore             ignore empty lines
  -j, --json               output as json
  -f, --filter <exp>       filter by a boolean expression
  -m, --map <exp>          map values using the expression
  -r, --reduce <func|exp>  reduce using a function or expression

Examples

filter

# Print all odd lines
# awk 'NR % 2 == 1' file
pjs -f 'i % 2 == 0' file

# Print all lines greater than 80 chars in length
# awk 'length($0) > 80' file
pjs -f 'length > 80' file

map

# Remove all digits
# tr -d 0-9 < file
pjs -m "replace(/\d/g, '')" file

# Get second item of each line in csv
# awk -F "," '{print $2}' file
pjs -m 'split(",")[1]' file

reduce

# Count lines in file
# wc -l file
# awk 'END { print NR }' file
pjs -r length file

# Sum all decimal numbers in a file
# awk '{ sum += $1 } END { print sum }' file
# perl -nle '$sum += $_ } END { print $sum' file
pjs -r 'Number(prev) + Number(curr)' file
pjs -r '(+prev) + (+curr)' file
pjs -r sum file

# Concatenate all lines in multiple files
# awk '{printf $0;}' file1 file2
# cat file1 file2 | tr -d '\n'
pjs -r concat file1 file2

mixed

# Print the length of the longest line
# awk '{ if (length($0) > max) max = length($0) } END { print max }' file
pjs -m 'length' -r max file

Comparison

Features	pjs	pythonpy	pru
Language	JavaScript	Python	Ruby
Streaming	Yes	Limited [1]	Yes
Implementation	Streams	Iterables	Generators
Easy JSON output	Yes	No	No
WebscaleTM	YES	No	No

_{[1] Can't perform "tail -f logfile | py -x x"}