catcatsboxboxeschildchildrenpersonpeoplemousemicepartypartiesanalysisanalysesphenomenonphenomenamatrixmatricesbusbusesheroheroeswifewivesAbout String Pluralizer
Generating copy, building test data or doing internationalization often requires picking the right singular or plural form for a count. English rules look simple but have many exceptions: consonant plus y becomes ies, s, x, ch and sh take es, and then there are irregular forms plus nouns whose singular and plural are identical. A tool centralizes the rules and exceptions so the codebase does not accumulate scattered conditionals. Implementations typically combine a rule table with an exception dictionary: check the dictionary first and return on a hit, otherwise match rules against the ending and apply them while preserving capitalization. Better ones also handle compound nouns by changing only the head noun, and distinguish countable from uncountable nouns, which in copy determines whether an article or quantifier is appropriate at all. Practical notes: Let an internationalization framework own the mapping from count to form rather than concatenating strings yourself, since many frameworks support per-language plural categories and some languages have more than the simple singular-plural split; hand-built concatenation breaks the moment you add a language. When generating test data, cover zero, one and many, because zero takes the same form as many in many languages rather than the singular, which is the mistake people make most. Also watch uncountable nouns, where attaching a quantifier is simply wrong. Limits: Most implementations only cover English, while other languages have large and intricate rule sets unsuited to a simple table, so use a dedicated internationalization library for them. Rules also cannot capture semantic exceptions where one word names different things in different contexts and pluralizes differently, which requires a hand-maintained list. Proper nouns and brand names must never be auto-changed, so exclude them when processing text in bulk or the output will misspell a brand. Data and privacy: this tool handles ordinary text, normally containing no personal information. Conversion happens in the browser, nothing is uploaded, and no network is needed; for unreleased copy, offline use avoids any leakage. On code organization, funnel this capability into one module or dependency instead of scattering string concatenation everywhere, so every pluralization in user-facing copy goes through the same entry point and adopting an internationalization library later means replacing one place. Add unit tests for the everyday edge cases too: zero, negatives, very large values and uncountable nouns. They cost little and block a large share of the "copy looks wrong after release" problems. Copy defects rarely break functionality and so get skipped in testing, yet they are the ones users notice most directly. First establish whether the text is English or another language: rules suit English while other languages should use an internationalization library, rather than implementing an inevitably incomplete rule set yourself. For maintenance, collect the nouns your product has used into a list kept with the code, add irregular forms to it along with a test as you meet them, and exceptions gradually converge. Also give the count its own placeholder in copy templates instead of hard-concatenating a number and a noun, or every language addition means rework.
How It Works Internally
The implementation is usually an exception dictionary plus a rule table: check the dictionary first and return on a hit, otherwise match the ending against rules. The difficulty is not the rules but the mapping from count to form — some languages have several plural categories rather than a singular-plural split, so hand-built concatenation inevitably breaks when a language is added, and the correct approach is to let an internationalization framework choose the form.
1 item / 2 item(s) / 1 child / 2 child(ren)1 item | 2 items (rule) | 1 child | 2 children (exception list, no -s)Measured on This Page
| Baseline before | 25 of 48 words wrong (52%): bus/gas/virus/status returned unchanged, photo→phooes, cafe→caves, human→humen, quiz→quizes |
|---|---|
| Failure 1 | Singular nouns ending in s were skipped entirely — bus, gas, virus, status, campus all unchanged |
| Failure 2 | The -o rule used a non-capturing group but referenced $1, dropping the leading consonant: photo→phooes, hero→heoes, potato→potaoes |
| Failure 3 | The f rule /fe?$/ applied to every f-ending: cafe→caves, chief→chieves, cliff→clifves, roof→rooves |
| Failure 4 | The man rule hit every -man ending: human→humen, ottoman→ottomen |
| After the fix | All 48 words match the reference; quiz→quizzes (doubled z), soliloquy→soliloquies, appendix→appendixes |
Against the 48-word reference the old code got 25 wrong (52%) and the rewrite gets 0. The design moves from an ordered rule list to default-plus-whitelist: the 11 -o words taking -es, the 14 f words becoming -ves and the 6 -man words that are not -men each live in their own set, so rules no longer fight over precedence. Latin plurals (indices, cacti, formulae) are kept because technical-word users expect them, even though inflect defaults to indexes and cactuses. English plurals have no universal rule and any automatic pluralizer can fail on rare words — this page marks its irregulars explicitly.
How to Use
- Open "String Pluralizer"
- Put one English noun per line on the left
- The plural form appears on the right
- Check the irregular forms look correct
- Copy the whole table with the copy icon
Use Cases
- API fields — Turn user and category into users and categories for REST paths or collection fields.
- Table names — Convert singular model names to plural table names per ORM conventions for consistency.
- Dynamic copy — Show the right form like 1 item / 3 items based on a count in the UI.
- Code generation — Derive collection names automatically when scaffolding CRUD code, reducing typos.
- Learning aid — Quickly verify irregular plurals like mouse and analysis.
- Localize UI messages — Generate correct plural forms for notification strings to handle count-based copy in your app.
- Normalize taxonomy — Convert singular tag or category names to their plural form for consistent taxonomy labeling.
FAQ
Does it handle irregular plurals?
It includes a dictionary of common irregulars like child to children, man to men, foot to feet, but English has many; double-check rare words.
What about words with identical forms?
Words like sheep, fish, and series, which are the same in singular and plural, are kept unchanged rather than gaining an "s."
What if I enter a word that is already plural?
The tool assumes singular input and may double-pluralize an already-plural word, so enter the singular form.
Does it support other languages?
It targets English word-formation only. Languages without morphological plurals, like Chinese, are out of scope.
Why does a -y ending sometimes add s and sometimes become ies?
A consonant + y (city) becomes ies, while a vowel + y (day) just adds s; the tool decides by the preceding letter.
How are -f/-fe endings handled?
Common words like knife to knives and leaf to leaves swap the ending to -ves, but safe and roof simply add s; the tool follows common rules, so verify exceptions.
Does it handle irregular nouns (person → people)?
Yes, a built-in table covers common irregulars (man/men, child/children, mouse/mice). Some loanwords (datum/data, criterion/criteria) may be missing — verify those. Keep proper nouns and abbreviations (SKU, ID) out of pluralisation.
Does it preserve capitalisation?
Yes, the original case pattern is kept: Cat→Cats, CAT→CATS. Whether all-caps abbreviations take a plain s (APIs) or an apostrophe (API's) is a style-manual decision — the tool applies the rule, the choice is yours.