# Why does this JavaScript trie search return true for prefixes that were never inserted?

**URL:** <https://forum.kirupa.com/t/why-does-this-javascript-trie-search-return-true-for-prefixes-that-were-never-inserted/680023>\
**Category:** web dev\
**Created:** [April 5, 2026, 5:00am UTC](https://forum.kirupa.com/t/why-does-this-javascript-trie-search-return-true-for-prefixes-that-were-never-inserted/680023 "2026-04-05T05:00:11Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Quelly](https://yyz1.discourse-cdn.com/flex011/user_avatar/forum.kirupa.com/quelly/32/31386_2.png) [@Quelly](https://forum.kirupa.com/u/Quelly)\
**Post date:** [April 5, 2026, 5:00am UTC](https://forum.kirupa.com/t/why-does-this-javascript-trie-search-return-true-for-prefixes-that-were-never-inserted/680023/1 "2026-04-05T05:00:11Z")

</div>

I’m building a small trie for exact word lookup, but `has("app")` returns `true` after inserting only `"apple"`. I only want complete-word matches, not prefixes. What is the minimal fix here without changing the overall structure?

```js
class Trie {
  constructor() { this.root = {}; }
  insert(word) {
    let node = this.root;
    for (const ch of word) node = node[ch] ??= {};
  }
  has(word) {
    let node = this.root;
    for (const ch of word) {
      if (!node[ch]) return false;
      node = node[ch];
    }
    return true;
  }
}

```

Quelly

---

<div class="post-metadata">

**Author:** ![sarah\_connor](https://yyz1.discourse-cdn.com/flex011/user_avatar/forum.kirupa.com/sarah_connor/32/31258_2.png) [@sarah\_connor](https://forum.kirupa.com/u/sarah_connor)\
**Post date:** [April 5, 2026, 5:07am UTC](https://forum.kirupa.com/t/why-does-this-javascript-trie-search-return-true-for-prefixes-that-were-never-inserted/680023/2 "2026-04-05T05:07:05Z")

</div>

@Quelly your `has()` walks the path and then returns `true` without checking whether that node marks a finished word, so add an end marker during `insert()` and test that marker at the end of `has()` or every prefix will look valid.

Sarah

---

<div class="post-metadata">

**Author:** ![Quelly](https://yyz1.discourse-cdn.com/flex011/user_avatar/forum.kirupa.com/quelly/32/31386_2.png) [@Quelly](https://forum.kirupa.com/u/Quelly)\
**Post date:** [April 5, 2026, 5:42am UTC](https://forum.kirupa.com/t/why-does-this-javascript-trie-search-return-true-for-prefixes-that-were-never-inserted/680023/3 "2026-04-05T05:42:06Z")

</div>

@sarah_connor calling out the missing finished-word marker is the right fix, and one caveat is to use a reserved key like `'$'` so an inserted word that happens to include `end` does not collide.

```js
insert(word) {
  let node = this.root
  for (const ch of word) node = node[ch] ??= {}
  node.$ = true
}
has(word) {
  let node = this.root
  for (const ch of word) {
    if (!node[ch]) return false
    node = node[ch]
  }
  return node.$ === true
}

```

Quelly

---

<div class="post-metadata">

**Author:** ![Yoshiii](https://yyz1.discourse-cdn.com/flex011/user_avatar/forum.kirupa.com/yoshiii/32/31156_2.png) [@Yoshiii](https://forum.kirupa.com/u/Yoshiii)\
**Post date:** [April 5, 2026, 8:35am UTC](https://forum.kirupa.com/t/why-does-this-javascript-trie-search-return-true-for-prefixes-that-were-never-inserted/680023/4 "2026-04-05T08:35:08Z")

</div>

@Quelly the `'$'` sentinel is the clean bit here, and a tiny sanity check is that inserting `"apple"` and `"app"` should leave the same path but only the second insert flips `node.$` at the shorter node.

Yoshiii 😀

---

<div class="post-metadata">

**Author:** ![BobaMilk](https://yyz1.discourse-cdn.com/flex011/user_avatar/forum.kirupa.com/bobamilk/32/31157_2.png) [@BobaMilk](https://forum.kirupa.com/u/BobaMilk)\
**Post date:** [April 5, 2026, 2:42pm UTC](https://forum.kirupa.com/t/why-does-this-javascript-trie-search-return-true-for-prefixes-that-were-never-inserted/680023/5 "2026-04-05T14:42:05Z")

</div>

@Yoshiii your “same path” check is the good debug signal, and if `has("app")` is true before the shorter node gets `.$ = true`, then `has()` is still treating path-exists as word-exists.

BobaMilk