I Know What I Need, But I Don't Know How To Build It

Pick the thing you need. Then derive it: meaning, state, operations, rules, examples, representation, pseudocode — and only then the code, one rung at a time.

I need XWhat is X?What must it remember?What can happen to it?What must always hold?ExamplesRepresent the stateWhich structure?Each operationPseudocodeImplement oneTest with examplesEdge casesIntegrate

“I know I need a shopping cart. I have no idea how to implement a shopping cart.” The gap is not a missing tutorial; it is a missing method. Every stage below asks you to write before it shows you anything, because the code is the last step of a derivation, not the first thing to search for.

Which concept?

Grouped by exercise level; the level is how far the derivation goes, not how hard the code is.

Say what you need
beginner
intermediate
advanced

How much help?

Difficulty dial

Every stage's content is shown right after your attempt box.

Search Index — twelve stages

I know I need full-text search over documents. I have no idea how to implement a search index.

0 / 12 stages attempted
I need XWhat is X?What must it remember?What can happen to it?What must always hold?ExamplesRepresent the stateWhich structure?Each operationPseudocodeImplement oneTest with examplesEdge casesIntegrate
stage 1 of 12

Define the concept

In one sentence, without any code: what is it? Then answer for yourself — does it have identity, who owns it, how long does it exist, should it survive a reload or a login?

Your attempt — write before you look

Search Index = A structure, derived from a set of documents, that answers "which documents contain these words?" without reading every document.

Identity, ownership, lifetime
  • Does an index have identity? Weakly. Two indexes built from the same documents are interchangeable, because the index is derived; the documents have identity, the index has a version — "built from which documents, when".
  • Who owns it? The service that owns the documents. Nobody edits an index by hand; it is written by the same code path that writes documents, or rebuilt from them.
  • How long does it exist? As long as the documents do, but it can be thrown away at any moment and rebuilt — that is the property that makes it an index and not a database.
  • Should it survive reload? V1 in memory says no and rebuilds on start; V3 says yes because rebuilding a large corpus on every restart is the cost that forces persistence.
  • Should it survive login? Not a question the index answers. Which documents a user may see is a permission filter applied to the result ids, owned by authorisation, not by the index.
The principle
If you do not know how to build the thing, make the thing smaller until you reach something you do know how to build. Go one primitive lower →
SIMPLIFIED

The catalog is one derivation, not the only one (impl §60). A different set of examples yields different rules; a different first requirement yields a different V1; a map instead of an array is not wrong, only an answer to a different question. Compare your derivation with the reference for the differences, then decide which ones were rules and which were style.