parse_num is a small library to parse atoms with a 'numeric' representation into the equivalent scalar.

For example, given the atom seventy_five, then parse_num:parse(seventy_five) returns the scalar 75.

parse_num works with representations ranging from zero to 10^9 - 1 (or, one less than one quadrillion).

1. parse_num:parse(one) %% returns 1

2. parse_num:parse(two) %% returns 2

3. parse_num:parse(twice) %% also returns 2

4. parse_num:parse(two_thousand_three_hundred) %% returns 2300

5. parse_num:parse(twenty_three_hundred) %% also returns 2300

6. parse_num:parse(nine_billion_eight_hundred_and_seventy_million_six_hundred_thousand_and_one) ;; returns 9870600001

parse_num works with both 'grammatically correct' representations (e.g. two_thousand_three_hundred above) and 'colloquial' representations (e.g. twenty_three_hundred).

parse_num is forgiving: for example, you do not need to correctly employ and (for which there exists specific rules).

parse_num is useful for e.g. Domain-Specific Language development; I wrote it for just such a project merlin but thought to release it on its own in case others find it useful.

Rails provides a similar facility that uses Ruby symbols, and leverages it to good effect in their API.

Please log any defects in the Issue Tracker.

• floating-point numbers
• negative numbers

Copyright (C) 2011 Edward Garson

Distributed under the terms of the MIT license as described by the included file MIT-LICENSE.