Autocomplete System with Trie in DSA Typescript - Execution Trace

Concept Flow - Autocomplete System with Trie

Start: Insert words into Trie

↓

For each word: For each char

↓

If char node missing: Create node

↓

Move to char node

↓

Mark end of word

↓

Autocomplete query: Traverse Trie by prefix

↓

If prefix node found: DFS to collect words

↓

Return list of words starting with prefix

Build a Trie by inserting words character by character, then find autocomplete suggestions by traversing the Trie nodes matching the prefix and collecting all words below.

Execution Sample

DSA Typescript

class TrieNode {
  children: Map<string, TrieNode> = new Map();
  isWordEnd: boolean = false;
}

// Insert 'cat'
// Insert 'car'
// Autocomplete 'ca'

Insert words 'cat' and 'car' into the Trie, then find all words starting with prefix 'ca'.

Execution Table

Step	Operation	Nodes in Trie	Pointer Changes	Visual State
1	Insert 'c'	Root -> c	Create node for 'c' under root	root └─ c
2	Insert 'a' after 'c'	Root -> c -> a	Create node for 'a' under 'c'	root └─ c └─ a
3	Insert 't' after 'a'	Root -> c -> a -> t (word end)	Create node for 't' under 'a', mark isWordEnd	root └─ c └─ a └─ t*
4	Insert 'c' (existing)	Root -> c -> a -> t*	Reuse node 'c' at root	root └─ c └─ a └─ t*
5	Insert 'a' (existing)	Root -> c -> a -> t*	Reuse node 'a' under 'c'	root └─ c └─ a └─ t*
6	Insert 'r' after 'a'	Root -> c -> a -> t*, r (word end)	Create node 'r' under 'a', mark isWordEnd	root └─ c └─ a ├─ t* └─ r*
7	Autocomplete 'ca' prefix	Traverse root -> c -> a	Pointer moves to node 'a'	At node 'a' with children 't*' and 'r*'
8	Collect words from 'a' node	Words found: 'cat', 'car'	DFS visits 't*' and 'r*' nodes	Suggestions: ['cat', 'car']
9	Return suggestions	Autocomplete complete	Return ['cat', 'car']	Output: ['cat', 'car']

💡 All characters inserted and prefix traversal complete, suggestions collected.

Variable Tracker

Variable	Start	After Step 1	After Step 3	After Step 6	After Step 7	Final
Trie Root	Empty	Has child 'c'	Has path c->a->t*	Has path c->a->t*, r*	Pointer at node 'a'	Pointer at node 'a'
Current Node	Root	Node 'c'	Node 't'	Node 'r'	Node 'a'	Node 'a'
isWordEnd flags	None	None	't' true	't' and 'r' true	't' and 'r' true	't' and 'r' true
Suggestions List	Empty	Empty	Empty	Empty	Empty	['cat', 'car']

Key Moments - 3 Insights

Why do we create new nodes only if the character node does not exist?

Because existing nodes represent shared prefixes, so we reuse them to save space and avoid duplicates. See execution_table steps 4 and 5 where 'c' and 'a' nodes are reused.

How do we know when a word ends in the Trie?

We mark the last character node of a word with isWordEnd = true. In execution_table step 3 and 6, nodes 't' and 'r' are marked to indicate 'cat' and 'car' are complete words.

How does autocomplete find all words starting with a prefix?

After reaching the prefix node, we perform a depth-first search (DFS) to collect all words below it. See execution_table step 8 where DFS visits 't*' and 'r*' nodes to find 'cat' and 'car'.

Visual Quiz - 3 Questions

Look at the execution_table, what is the visual state after step 6?

A root └─ c └─ a └─ t*

B root └─ c └─ a ├─ t* └─ r*

C root └─ c └─ r*

D root └─ a └─ t*

💡 Hint

Check the Visual State column in execution_table row for step 6.

At which step does the pointer move to the prefix node 'a' during autocomplete?

A Step 7

B Step 5

C Step 3

D Step 9

💡 Hint

Look at the Pointer Changes column in execution_table for step 7.

If we insert the word 'cap' after 'car', how would the Trie nodes under 'a' change?

A Add a new child node 'p' under 'c'

B Replace node 'r' with 'p'

C Add a new child node 'p' under 'a' marked as word end

D No change, 'cap' is ignored

💡 Hint

New words add new nodes for characters not already present under the prefix node.

Concept Snapshot

Autocomplete with Trie:
- Insert words char by char, creating nodes if missing
- Mark last char node as word end
- For prefix search, traverse Trie nodes matching prefix
- From prefix node, DFS to collect all words
- Efficient prefix-based word retrieval
- Shared prefixes save space and speed up search

Full Transcript

This visualization shows how an autocomplete system uses a Trie data structure. Words are inserted one character at a time, creating new nodes only when needed. Each word's last character node is marked to indicate a complete word. To autocomplete, we traverse the Trie to the prefix node, then collect all words below it using depth-first search. The execution table tracks each insertion and traversal step, showing how nodes are created and reused. The variable tracker follows pointers and word-end flags. Key moments clarify why nodes are reused, how word ends are marked, and how autocomplete collects suggestions. The quiz tests understanding of Trie structure and traversal steps.