Skip to content

README.md

Latest commit

 

History

History
90 lines (59 loc) · 3.42 KB

README.md

File metadata and controls

90 lines (59 loc) · 3.42 KB

Compounded

Compounded is a static, browser-based word puzzle. You get a prompt word and find partner words that combine with it to form real compound words (for example, sun + flower → sunflower). The game loads a JSON dataset and runs entirely in the browser—no backend.

How to play

  1. Read the prompt word on screen.
  2. Type a partner that forms a compound with it, or type the full compound.
  3. Find as many valid matches as you can for that prompt. Hint reveals a letter pattern for an unfound partner; Skip shows unfound answers and moves on; New word picks another prompt.

Features

  • Score, round, found count, and remaining matches
  • Guesses accepted as partner only or full compound (case-insensitive)
  • Shuffled match order each round
  • Fully static assets (HTML, CSS, JS, JSON)

Running locally

The game uses fetch() to load compounds-clean-edited.json, so open it through a local HTTP server (opening index.html directly from disk may fail in some browsers).

python -m http.server 8000

Then visit http://localhost:8000.

Any static file server works (for example npx serve).

Deployment

Host the static files on GitHub Pages, Netlify, Vercel, or any static host. Ensure compounds-clean-edited.json is deployed alongside index.html, style.css, and script.js.

Project structure

Path Purpose
index.html Page markup
style.css Layout and styling
script.js Game logic and UI
compounds-clean-edited.json Playable compound data (base + matches with partner / compound)
make_compounds.py Build compounds-clean.json from a word list + noun filter
build_noun_set.py Build noun_words.json from NLTK WordNet
inspect_compounds.py CLI to list matches for a given base in compounds-clean.json
remove_base_words.py Filter out selected bases; writes compounds-clean-edited.json
remove_partner_matches.py Remove specific base/partner pairs; writes compounds-clean-edited.json
requirements.txt Python dependency (nltk)
.github/ISSUE_TEMPLATE/ GitHub issue templates

Regenerating or editing the dataset (optional)

These steps are for maintainers who want to rebuild or curate data. Playing the game only requires the four static files above.

  1. Python 3 and dependencies:

    pip install -r requirements.txt

  2. Noun list (WordNet via NLTK):

    python build_noun_set.py

    Produces noun_words.json (WordNet is downloaded on first run if needed).

  3. Compound candidates require a wordlist.txt (one word per line, lowercase letters). That file is not tracked in this repository; supply your own list under that name next to make_compounds.py, then run:

    python make_compounds.py

    Output: compounds-clean.json.

  4. Inspect a base word:

    python inspect_compounds.py apple

  5. Curate for gameplay: adjust remove_base_words.py or remove_partner_matches.py (or edit JSON by hand), then run the script you need. Both read compounds-clean.json and write compounds-clean-edited.json, which is what the front end loads.

Future ideas

  • Daily puzzle, difficulty tiers, timer, leaderboards, mobile polish

License

MIT License