neng, Non-Extravagant Name Generator

Introduction

neng is a Golang package that can generate random English phrases from nouns, verbs, adverbs and adjectives according to user-specified pattern. It is powered by diverse collection of 41000 words compiled from WordNet Lexical Database. Inspired by Terraria's world name generator, neng is designed to be simple yet versatile name making tool for other projects.

Unsuitable for cryptographic applications.

Showcase

go run github.com/Zedran/neng/examples/phrase@latest

Sample use

Code

package main

import (
    "fmt"

    "github.com/Zedran/neng"
)

func main() {
    gen, _ := neng.DefaultGenerator()

    // <title case + noun> <Present Simple + verb> the <upper + adjective> <upper + noun>.
    phrase, _ := gen.Phrase("%tn %Nv the %ua %un.")

    // A single, transformed verb
    verb, _ := gen.Verb(neng.MOD_PAST_SIMPLE)

    // Transforming an arbitrary word
    word, _ := gen.Transform("involve", neng.WC_VERB, neng.MOD_GERUND|neng.MOD_CASE_TITLE)

    fmt.Printf("Phrase -> %s\nVerb   -> %s\nWord   -> %s\n", phrase, verb, word)
}

Output

Phrase -> Serenade perplexes the STRAY SUPERBUG.
Verb   -> shared
Word   -> Involving

Phrase pattern commands

neng's phrase generation syntax resembles C-style string format specifiers. The user provides a pattern consisting of commands, preceded by the escape character, to insert (and optionally transform) randomly generated words. A regular text can be mixed with command syntax.

Escape character: %

Insertion

Symbol	WordClass	Description
%		Inserts % sign
a	WC_ADJECTIVE	Inserts a random adjective
m	WC_ADVERB	Inserts a random adverb
n	WC_NOUN	Inserts a random noun
v	WC_VERB	Inserts a random verb

WordClass values are required by some of the Generator's methods to recognize parts of speech.

Transformation

Transformations can only be applied to compatible parts of speech.

Symbols are used to request transformations for words within a phrase. Constants of type Mod are designed to work with "single-word" methods.

Symbol	Compatible with	Mod	Description
2	verb	MOD_PAST_SIMPLE	Past Simple (2nd form)
3	verb	MOD_PAST_PARTICIPLE	Past Participle (3rd form)
N	verb	MOD_PRESENT_SIMPLE	Present Simple (now)
c	adjective, adverb	MOD_COMPARATIVE	Comparative (better)
f	any	MOD_CASE_SENTENCE	Sentence case (first letter)
g	verb	MOD_GERUND	Gerund
i	adjective, adverb, noun*	MOD_INDEF	Indefinite adjective (a, an)
_	noun	MOD_INDEF_SILENT	Silent indefinite**
l	any	MOD_CASE_LOWER	lower case
p	noun, verb***	MOD_PLURAL	Plural form
s	adjective, adverb	MOD_SUPERLATIVE	Superlative (best)
t	any	MOD_CASE_TITLE	Title Case
u	any	MOD_CASE_UPPER	UPPER CASE

* MOD_INDEF is not compatible with MOD_PLURAL and MOD_SUPERLATIVE.

** MOD_INDEF_SILENT ensures that the noun is grammatically compatible with an indefinite article (not uncountable, not plural-only), but does not modify it in any way. It is useful in phrase patterns such as %ia %_n, where the indefinite article belongs to the noun, but it stands before the adjective describing the noun. If Generator.TransformWord method receives silent indefinite, it does nothing to the provided word, but it still returns an error in case of incompatibility.

*** MOD_PLURAL is only compatible with verbs when combined with MOD_PAST_SIMPLE or MOD_PRESENT_SIMPLE.

Mod values form two conceptual categories: grammar modifiers and case modifiers. Only one modifier from each category may be applied to any given word. If multiple modifiers of the same kind are specified, the one with the lowest value is applied. The above-mentioned verb transformations with MOD_PLURAL and MOD_INDEF are exceptions to this rule.

State of the vocabulary

Generator's default vocabulary consists of:

• 23000 nouns
• 10000 adjectives
• 6000 verbs
• 2000 adverbs

Original WordNet lists have been thoroughly vetted. I have strived to remove any words that are offensive, too specific (chemistry, medicine) or relate to topics that are considered sensitive, controversial or fear-inducing. However, I am not native English speaker and the database is quite large, so it is likely I have missed something. If you find any unsuitable words, I will be happy to hear from you.

If the embedded database does not meet your requirements, you can provide neng with your own word lists. To ensure the accuracy of transformations, I recommend that your custom vocabulary remains a subset of the embedded one.

Attributions

Refer to NOTICE.md.

License

This software is available under MIT License.