Accept small words as a config option

blakeembrey · Oct 3, 2023 · c2c96b6 · c2c96b6
1 parent 4befb41
commit c2c96b6
Show file tree

Hide file tree

Showing 2 changed files with 78 additions and 24 deletions.
diff --git a/packages/title-case/src/index.spec.ts b/packages/title-case/src/index.spec.ts
@@ -65,6 +65,7 @@ const TEST_CASES: [string, string][] = [
     "Is Human Activity Responsible for the Climate Emergency? New Report Calls It ‘Unequivocal.’",
   ],
   ["лев николаевич толстой", "Лев Николаевич Толстой"],
+  ["Read foo-bar.com", "Read foo-bar.com"],
 ];
 
 describe("swap case", () => {

diff --git a/packages/title-case/src/index.ts b/packages/title-case/src/index.ts
@@ -1,37 +1,90 @@
-const SMALL_WORDS =
-  /\b(?:an?d?|a[st]|because|but|by|en|for|i[fn]|neither|nor|o[fnr]|only|over|per|so|some|tha[tn]|the|to|up|upon|vs?\.?|versus|via|when|with|without|yet)\b/i;
-const TOKENS = /[^\s:–—-]+|./g;
-const WHITESPACE = /\s/;
+const TOKENS = /\S+|./g;
 const IS_MANUAL_CASE = /\p{Ll}(?=[\p{Lu}])|\.\p{L}/u; // iPhone, example.com, U.N., etc.
-const ALPHANUMERIC_PATTERN = /[\p{L}\d]/u;
+const ALPHANUMERIC_PATTERN = /[\p{L}\d]+/gu;
 
-export function titleCase(input: string, locale?: string[] | string) {
+const WORD_SEPARATORS = new Set(["—", "–", "-", "―", "/"]);
+
+const SMALL_WORDS = new Set([
+  "an",
+  "and",
+  "as",
+  "at",
+  "because",
+  "but",
+  "by",
+  "en",
+  "for",
+  "if",
+  "in",
+  "neither",
+  "nor",
+  "of",
+  "on",
+  "or",
+  "only",
+  "over",
+  "per",
+  "so",
+  "some",
+  "that",
+  "than",
+  "the",
+  "to",
+  "up",
+  "upon",
+  "v",
+  "vs",
+  "versus",
+  "via",
+  "when",
+  "with",
+  "without",
+  "yet",
+]);
+
+export interface Options {
+  smallWords?: Set<string>;
+  locale?: string | string[];
+}
+
+export function titleCase(
+  input: string,
+  options: Options | string[] | string = {},
+) {
   let result = "";
   let m: RegExpExecArray | null;
 
+  const { smallWords = SMALL_WORDS, locale } =
+    typeof options === "string" || Array.isArray(options)
+      ? { locale: options }
+      : options;
+
   // tslint:disable-next-line
   while ((m = TOKENS.exec(input)) !== null) {
     const { 0: token, index } = m;
 
-    if (
-      // Ignore already capitalized words.
-      !IS_MANUAL_CASE.test(token) &&
-      // Ignore small words except at beginning or end.
-      (!SMALL_WORDS.test(token) ||
-        index === 0 ||
-        index + token.length === input.length) &&
-      // Ignore URLs.
-      (input.charAt(index + token.length) !== ":" ||
-        WHITESPACE.test(input.charAt(index + token.length + 1)))
-    ) {
-      // Find and uppercase first word character, skips over *modifiers*.
-      result += token.replace(ALPHANUMERIC_PATTERN, (m) =>
-        m.toLocaleUpperCase(locale),
-      );
-      continue;
-    }
+    // Ignore already capitalized words.
+    if (IS_MANUAL_CASE.test(token)) {
+      result += token;
+    } else {
+      result += token.replace(ALPHANUMERIC_PATTERN, (m, i) => {
+        // Ignore small words except at beginning or end.
+        if (
+          index > 0 &&
+          index + token.length < input.length &&
+          smallWords.has(m)
+        ) {
+          return m;
+        }
 
-    result += token;
+        // Only capitalize words after a valid word separator.
+        if (i > 1 && !WORD_SEPARATORS.has(input.charAt(index + i - 1))) {
+          return m;
+        }
+
+        return m.charAt(0).toLocaleUpperCase(locale) + m.slice(1);
+      });
+    }
   }
 
   return result;