Skip to main content
Version: 4.0 (next)

Pluralize and singularize nouns

Orientation​

Use Pluralize or Singularize when the application has an English noun but not a count. Use ToQuantity when the count and noun belong together. Humanizer’s vocabulary handles common irregular and uncountable English nouns; an application can add domain-specific rules before it begins formatting text.

Example​

This isolated executable covers a built-in English irregular noun, a quantity written as words, one application-specific irregular pair, and caller-authored Polish plural forms:

Program.cs
using System.Globalization;
using Humanizer;

var culture = CultureInfo.GetCultureInfo("en-US");
CultureInfo.CurrentCulture = culture;
CultureInfo.CurrentUICulture = culture;

AssertEqual("people", "person".Pluralize());
AssertEqual("person", "people".Singularize());
AssertEqual("two people", "person".ToQuantity(2, ShowQuantityAs.Words));

Vocabularies.Default.AddIrregular("cactoid", "cactoidae", matchEnding: false);
AssertEqual("cactoidae", "cactoid".Pluralize());
AssertEqual("cactoid", "cactoidae".Singularize());

var polishForms = new PluralizationForms(
    singular: "plik",
    other: "pliku",
    one: "plik",
    few: "pliki",
    many: "plików");
AssertTrue(polishForms.TryPluralize(5m, CultureInfo.GetCultureInfo("pl"), out var polish));
AssertEqual("plików", polish);
AssertTrue(polishForms.TrySingularize("pliki", out var singular));
AssertEqual("plik", singular);

Console.WriteLine("people; two people; cactoidae; plików; plik");

static void AssertEqual(string expected, string? actual)
{
    if (actual != expected)
        throw new InvalidOperationException($"Expected '{expected}', got '{actual}'.");
}

static void AssertTrue(bool value)
{
    if (!value)
        throw new InvalidOperationException("Expected success.");
}

Tell Humanizer what the input means​

Pluralize(inputIsKnownToBeSingular: false) avoids pluralizing an input that may already be plural. The matching Singularize flag avoids a second singularization. Singularize(skipSimpleWords: true) specifically skips the simple trailing-s rule for a one-word input; irregular and other configured rules still apply.

ToQuantity accepts:

  • ShowQuantityAs.None for only the correctly inflected noun.
  • ShowQuantityAs.Numeric for a formatted numeric count.
  • ShowQuantityAs.Words for a localized number word followed by the noun.

Numeric format strings and providers control the displayed number. They do not turn the English inflector into a localized noun engine.

Pluralize a localized noun for a count​

On Humanizer 4, PluralizationForms.TryPluralize selects a caller-supplied form using the explicit culture's Unicode CLDR decimal cardinal rule. Every supported culture supplies this rule. If the selected form was not supplied, the operation returns false; it does not guess morphology or fall back to English. Use PluralizationForms.Invariant for a word that never changes, or explicitly supply equal forms when categories share spelling. The encoded decimal scale is significant, so a culture can select different forms for 1m and 1.0m.

CLDR categories are count-selection tags, not complete grammatical categories. PluralizationForms models one noun's singular and plural forms. Call TrySingularize on that form set to resolve any exact authored form back to its singular spelling. Matching is Unicode NFC-normalized, ordinal, and case-sensitive.

The selector follows the Unicode CLDR plural-rule contract and its decimal operands.

Extend the vocabulary​

Vocabularies.Default supports AddIrregular, AddUncountable, AddPlural, and AddSingular. Prefer an irregular or uncountable entry over a broad regular expression. Set matchEnding: false when a rule must match the entire noun instead of every word ending with the same letters.

In Humanizer 4, AddAcronym uses the same vocabulary to preserve the canonical casing of a domain acronym during string humanization. Registration is case-insensitive, but the registered spelling controls the output. Acronyms must contain letters only.

Pitfall​

Vocabulary changes, including acronym registrations in Humanizer 4, are process-global. Register domain words once during startup and test them in an isolated process; do not add rules per request. ToQuantity treats exactly 1 and -1 as singular. Fractional values, NaN, and infinity use the plural form.

The existing Pluralize, Singularize, vocabulary, and ToQuantity APIs continue to pluralize and singularize English nouns. Every supported culture supplies the cardinal rule used by PluralizationForms.

Version notes​

Core inflection is available across the documented release set. Individual vocabulary rules and overloads have changed, and early Humanizer 3.0.x temporarily lacked quantity overloads restored by 3.0.8. The verified source runs against the selected package. AddAcronym is available on Humanizer 4. Culture-explicit PluralizationForms are new in v4.