Skip to content

Alfred unit converter is a smart calculator for Alfred with support for unit conversions to make it a bit comparable to the Google Calculator and Wolfram Alpha.

Notifications You must be signed in to change notification settings

wolph/alfred-converter

Repository files navigation

Alfred Unit Converter

Alfred unit converter is a smart calculator for Alfred with support for unit conversions to make it a bit comparable to the Google Calculator and Wolfram Alpha.

Issues

If new units and/or other names for units should be added please let me know by creating an issue at: https://github.com/WoLpH/alfred-converter/issues

Installation

To install use the following link:

https://github.com/WoLpH/alfred-converter/blob/master/unit_converter.alfredworkflow

Configuration

The extension can be configured through the environment variables setting in Alfred.

BASE_2: Enable base 2 (binary ) output
BASE_8: Enable base 8 (octal) output
BASE_16: Enable base 16 (hexadecimal) output
DECIMAL_SEPARATOR: Comma or dot separator for decimals
FRACTIONAL_UNITS: "both", "decimal" or "fractional" only
MAX_MAGNITUDE: Maximum orders of magnitude to show. For 1 megabyte in bytes we need 9 orders of magnitude because it's 1 million bytes.
UNITS_BLACKLIST: Units you wish to hide
UNITS_SIDE: Showing the units at the right or the left side

Example queries

1m in cm # Just a simple conversion
2^30 byte # Using powers before conversion
5' # Converting units with special characters
20" # Like above
5 * cos(pi + 2) # Executing mathematical functions
5 * pi + 2 mm in m # Mathematical constants with unit conversion
1 * cos(pi/2) - sin(pi^2) # More advanced mathematical expressions
ln(e^10) # Testing the ln(x) alias of log _e(x)
log(e^10) # The normal log method
5+3^2" in mm # Testing math with unit conversion
1 + 2 / 3 * 4) mm^2 in cm^2 # Unbalanced paranthesis with unit conversion
((1 + 2 / 3 * 4) mm^2 in cm^2 # Unbalanced paranthesis the other way
inf - inf # Not actually possible, but we backtrack to "inf"
0b1010 + 0xA - 050 # Test calculations with hex, oct and binary numbers

https://raw.githubusercontent.com/WoLpH/alfred-converter/master/examples/bytes.png

https://raw.githubusercontent.com/WoLpH/alfred-converter/master/examples/exponent_rounding.png

https://raw.githubusercontent.com/WoLpH/alfred-converter/master/examples/mathematical_functions.png

https://raw.githubusercontent.com/WoLpH/alfred-converter/master/examples/square_metres.png

https://raw.githubusercontent.com/WoLpH/alfred-converter/master/examples/bin_oct_hex.png

The list of units and conversions was downloaded from: http://w3.energistics.org/uom/poscUnits22.xml

About

Alfred unit converter is a smart calculator for Alfred with support for unit conversions to make it a bit comparable to the Google Calculator and Wolfram Alpha.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published