Add atStart() and atEnd() convenience methods to MatchingContext.

src/MatchingContext.js

@@ -35,6 +35,15 @@ class MatchingContext {
       ? mlen
       : emptyMatchAdvance(this.match.input, index, this.replacer.regexp.unicode));
   }
+
+  atStart() {
+    return this.match.index === 0;
+  }
+
+  atEnd() {
+    const { match } = this;
+    return match.index + match[0].length >= match.input.length;
+  }
 }
 
 export default MatchingContext;

test/matching-context.spec.js

@@ -0,0 +1,26 @@
+const UnionReplacer = require('../dist/union-replacer.cjs');
+
+describe('MatchingContext', () => {
+  function startEndReplace(mctx) {
+    const l = mctx.match[0].length;
+    const s = mctx.atStart() ? 'S' : '_';
+    const e = mctx.atEnd() ? 'E' : '_';
+    return `[${s}${e}:${l}]`;
+  }
+  it('recognizes start and end', () => {
+    const r = new UnionReplacer([[/^x|y|z$/, startEndReplace, true]]);
+    expect(r.replace('xyz')).toBe('[S_:1][__:1][_E:1]');
+  });
+  it('recognizes start = end match and empty end', () => {
+    const r = new UnionReplacer([[/x|$/, startEndReplace, true]]);
+    expect(r.replace('x')).toBe('[SE:1][_E:0]');
+  });
+  it('recognizes empty start and empty end', () => {
+    const r = new UnionReplacer([[/^|$/, startEndReplace, true]]);
+    expect(r.replace('x')).toBe('[S_:0]x[_E:0]');
+  });
+  it('recognizes empty start and empty on empty string', () => {
+    const r = new UnionReplacer([[/^|$/, startEndReplace, true]]);
+    expect(r.replace('')).toBe('[SE:0]');
+  });
+});