How it works

This galgje-solver is totally client-side. Unlike typical word finders that search through a massive list of words one by one, this tool uses a data structure called a Trie (Prefix Tree).

Have a look at the trie file: teije.ma/dutch_trie.json. The original curated word-dataset is provided by Stichting OpenTaal.

A standard dictionary file lists every word separately (gal, galg, galgje). This is inefficient because it repeats shared prefixes over and over. A Trie reorganizes this data into a tree of linked letters.

When you type a pattern like g?lg?e, the algorithm doesn’t scan 180,000+ words. Instead, it “walks” down the branches of the tree:

• It starts at G.
• It explores every possible letter at the wildcard ? position.
• It continues only down paths that match L.

If a branch doesn’t match your pattern, it is abandoned instantly. This reduces the search space from hundreds of thousands of words to just a few dozen logical paths, allowing the entire process to happen locally in milliseconds without ever touching a server.